Dry-run preview of untracked file removal
Tests ability to preview untracked file removal with git clean -n (dry-run). Evaluates safe-cleanup workflow.

These commands set up the repo before the model sees the prompt. They define the starting file structure, staged changes, and Git history.

  1. 01 git init
  2. 02 git config user.email 'test@test.com'
  3. 03 git config user.name 'Test User'
  4. 04 echo 'tracked' > tracked.txt
  5. 05 git add tracked.txt
  6. 06 git commit -m 'Initial commit'
  7. 07 echo 'untracked' > untracked.txt
  8. 08 echo 'another' > another.txt
Prompt
Show what files would be removed by git clean without actually removing them. Output ONLY the git command, nothing else.
Expected
git clean -f -n

Scoped model quality, cost, API time, and token usage for git_clean/f002.

Loading...
Loading raw attempt evidence…
anthropic/claude-fable-5.1:high PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-fable-5.1:high__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5.1:low PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-fable-5.1:low__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5.1:max PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-fable-5.1:max__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5.1:medium PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-fable-5.1:medium__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5.1:xhigh PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-fable-5.1:xhigh__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5:high PASS 100% 90 in → 19 out (18 reasoning)
git clean -n
anthropic/claude-fable-5:high__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5:low PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-fable-5:low__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5:max PASS 100% 90 in → 170 out (39 reasoning)
git clean -n
anthropic/claude-fable-5:max__json_schema PASS 100% 300 in → 137 out (50 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5:medium PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-fable-5:medium__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-fable-5:xhigh PASS 100% 90 in → 30 out (28 reasoning)
git clean -n
anthropic/claude-fable-5:xhigh__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-haiku-4.5:high PASS 100% 103 in → 156 out (144 reasoning)
``` git clean -n ```
anthropic/claude-haiku-4.5:high__json_schema PASS 100% 266 in → 248 out (233 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-haiku-4.5:low PASS 100% 103 in → 142 out (139 reasoning)
git clean -n
anthropic/claude-haiku-4.5:low__json_schema PASS 100% 266 in → 174 out (171 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-haiku-4.5:medium PASS 100% 103 in → 125 out (109 reasoning)
``` git clean -n ```
anthropic/claude-haiku-4.5:medium__json_schema PASS 100% 266 in → 230 out (219 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-haiku-4.5:none PASS 100% 74 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-haiku-4.5:none__json_schema PASS 100% 236 in → 12 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-haiku-4.5:xhigh PASS 100% 103 in → 120 out (104 reasoning)
``` git clean -n ```
anthropic/claude-haiku-4.5:xhigh__json_schema PASS 100% 266 in → 181 out (169 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.6:high PASS 100% 74 in → 12 out (0 reasoning)
``` git clean -n ```
anthropic/claude-opus-4.6:high__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.6:low PASS 100% 74 in → 9 out (0 reasoning)
`git clean -n`
anthropic/claude-opus-4.6:low__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.6:max PASS 100% 74 in → 12 out (0 reasoning)
``` git clean -n ```
anthropic/claude-opus-4.6:max__json_schema PASS 100% 237 in → 74 out (48 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.6:medium PASS 100% 74 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-opus-4.6:medium__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.6:none PASS 100% 74 in → 12 out (0 reasoning)
``` git clean -dn ```
anthropic/claude-opus-4.6:none__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.6:xhigh PASS 100% 74 in → 12 out (0 reasoning)
``` git clean -n ```
anthropic/claude-opus-4.6:xhigh__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.7:high PASS 100% 95 in → 9 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.7:high__json_schema PASS 100% 305 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.7:low PASS 100% 95 in → 9 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.7:low__json_schema PASS 100% 305 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.7:max PASS 100% 95 in → 14 out (0 reasoning)
``` git clean -n ```
anthropic/claude-opus-4.7:max__json_schema PASS 100% 305 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.7:medium PASS 100% 95 in → 9 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.7:medium__json_schema PASS 100% 305 in → 17 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.7:none PASS 100% 95 in → 9 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.7:none__json_schema PASS 100% 305 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.7:xhigh PASS 100% 95 in → 9 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.7:xhigh__json_schema PASS 100% 305 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.8:high PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.8:high__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.8:low PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.8:low__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.8:max PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.8:max__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.8:medium PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.8:medium__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.8:none PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.8:none__json_schema PASS 100% 300 in → 13 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-4.8:xhigh PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-4.8:xhigh__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-5.5:high PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-5.5:high__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-opus-5.5:low PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-5.5:low__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-opus-5.5:max PASS 100% 92 in → 137 out (126 reasoning)
``` git clean -n ```
anthropic/claude-opus-5.5:max__json_schema PASS 100% 302 in → 130 out (116 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-opus-5.5:medium PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-5.5:medium__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-opus-5.5:xhigh PASS 100% 92 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-5.5:xhigh__json_schema PASS 100% 302 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-opus-5:high PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-5:high__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-opus-5:low PASS 100% 90 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-opus-5:low__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-5:max PASS 100% 90 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-opus-5:max__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-opus-5:medium PASS 100% 90 in → 9 out (0 reasoning)
`git clean -n`
anthropic/claude-opus-5:medium__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-opus-5:xhigh PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-opus-5:xhigh__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-4.6:high PASS 100% 74 in → 92 out (74 reasoning)
``` git clean -n ```
anthropic/claude-sonnet-4.6:high__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-4.6:low PASS 100% 74 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-sonnet-4.6:low__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-4.6:max PASS 100% 74 in → 100 out (78 reasoning)
``` git clean -n ```
anthropic/claude-sonnet-4.6:max__json_schema PASS 100% 237 in → 119 out (93 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-4.6:medium PASS 100% 74 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-sonnet-4.6:medium__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-4.6:none PASS 100% 74 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-sonnet-4.6:none__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-4.6:xhigh PASS 100% 74 in → 112 out (89 reasoning)
``` git clean -n ```
anthropic/claude-sonnet-4.6:xhigh__json_schema PASS 100% 237 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-5:high PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-sonnet-5:high__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-5:low PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-sonnet-5:low__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-5:medium PASS 100% 90 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-sonnet-5:medium__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
anthropic/claude-sonnet-5:none PASS 100% 90 in → 7 out (0 reasoning)
git clean -n
anthropic/claude-sonnet-5:none__json_schema PASS 100% 300 in → 13 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
anthropic/claude-sonnet-5:xhigh PASS 100% 90 in → 11 out (0 reasoning)
``` git clean -n ```
anthropic/claude-sonnet-5:xhigh__json_schema PASS 100% 300 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
arcee-ai/trinity-large-thinking:high PASS 100% 61 in → 287 out (281 reasoning)
git clean -n
arcee-ai/trinity-large-thinking:low PASS 100% 61 in → 277 out (272 reasoning)
git clean -n
arcee-ai/trinity-large-thinking:medium PASS 100% 61 in → 254 out (248 reasoning)
git clean -n
arcee-ai/trinity-large-thinking:xhigh PASS 100% 59 in → 0 out (203 reasoning)
(empty output)
arcee-ai/trinity-mini:high PASS 100% 61 in → 348 out (380 reasoning)
git clean -n
arcee-ai/trinity-mini:high__json_schema PASS 100% 61 in → 378 out (417 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
arcee-ai/trinity-mini:low PASS 100% 61 in → 320 out (376 reasoning)
git clean -n
arcee-ai/trinity-mini:low__json_schema PASS 100% 61 in → 399 out (456 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command":"git clean -n" }
arcee-ai/trinity-mini:medium PASS 100% 61 in → 377 out (428 reasoning)
git clean -n
arcee-ai/trinity-mini:medium__json_schema PASS 100% 61 in → 358 out (401 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
arcee-ai/trinity-mini:xhigh PASS 100% 61 in → 605 out (682 reasoning)
gitclean -n
arcee-ai/trinity-mini:xhigh__json_schema PASS 100% 61 in → 344 out (377 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-flash-0731:high PASS 100% 144 in → 302 out (321 reasoning)
git clean -n
deepseek/deepseek-v4-flash-0731:high__json_schema PASS 100% 138 in → 90 out (84 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-flash-0731:low PASS 100% 59 in → 133 out (132 reasoning)
git clean -n
deepseek/deepseek-v4-flash-0731:low__json_schema PASS 100% 201 in → 115 out (105 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
deepseek/deepseek-v4-flash-0731:max PASS 100% 151 in → 239 out (234 reasoning)
git clean -n
deepseek/deepseek-v4-flash-0731:max__json_schema PASS 100% 138 in → 109 out (99 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-flash-0731:medium PASS 100% 59 in → 60 out (54 reasoning)
git clean -n
deepseek/deepseek-v4-flash-0731:medium__json_schema PASS 100% 201 in → 1,809 out (1,789 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
deepseek/deepseek-v4-flash-0731:none PASS 100% 59 in → 5 out (0 reasoning)
git clean -n
deepseek/deepseek-v4-flash-0731:none__json_schema PASS 100% 59 in → 14 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-flash-0731:xhigh PASS 100% 138 in → 67 out (62 reasoning)
git clean -n
deepseek/deepseek-v4-flash-0731:xhigh__json_schema PASS 100% 144 in → 293 out (294 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-flash:high PASS 100% 59 in → 104 out (95 reasoning)
git clean -n
deepseek/deepseek-v4-flash:high__json_schema PASS 100% 123 in → 122 out (120 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
deepseek/deepseek-v4-flash:low PASS 100% 59 in → 107 out (101 reasoning)
git clean -n
deepseek/deepseek-v4-flash:low__json_schema PASS 100% 61 in → 117 out (103 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-flash:medium PASS 100% 59 in → 109 out (103 reasoning)
git clean -n
deepseek/deepseek-v4-flash:medium__json_schema PASS 100% 61 in → 116 out (103 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-flash:none PASS 100% 59 in → 4 out (0 reasoning)
git clean -n
deepseek/deepseek-v4-flash:none__json_schema PASS 100% 61 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-flash:xhigh PASS 100% 138 in → 303 out (297 reasoning)
git clean -n
deepseek/deepseek-v4-flash:xhigh__json_schema PASS 100% 138 in → 296 out (296 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-pro:high PASS 100% 59 in → 5 out (0 reasoning)
git clean -n
deepseek/deepseek-v4-pro:high__json_schema PASS 100% 61 in → 93 out (79 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-pro:low PASS 100% 59 in → 119 out (114 reasoning)
git clean -n
deepseek/deepseek-v4-pro:low__json_schema PASS 100% 246 in → 82 out (71 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
deepseek/deepseek-v4-pro:medium PASS 100% 59 in → 163 out (166 reasoning)
git clean -n
deepseek/deepseek-v4-pro:medium__json_schema PASS 100% 59 in → 9 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
deepseek/deepseek-v4-pro:none PASS 100% 59 in → 4 out (0 reasoning)
git clean -n
deepseek/deepseek-v4-pro:none__json_schema PASS 100% 61 in → 12 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4-pro:xhigh PASS 100% 59 in → 5 out (0 reasoning)
git clean -n
deepseek/deepseek-v4-pro:xhigh__json_schema PASS 100% 324 in → 131 out (120 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
deepseek/deepseek-v4.1-flash:high PASS 100% 85 in → 101 out (95 reasoning)
git clean -n
deepseek/deepseek-v4.1-flash:high__json_schema PASS 100% 85 in → 77 out (67 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
deepseek/deepseek-v4.1-flash:low PASS 100% 85 in → 120 out (114 reasoning)
git clean -n
deepseek/deepseek-v4.1-flash:low__json_schema PASS 100% 85 in → 281 out (271 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
deepseek/deepseek-v4.1-flash:max PASS 100% 85 in → 125 out (119 reasoning)
git clean -n
deepseek/deepseek-v4.1-flash:max__json_schema PASS 100% 176 in → 42 out (31 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
deepseek/deepseek-v4.1-flash:medium PASS 100% 85 in → 76 out (77 reasoning)
git clean -n
deepseek/deepseek-v4.1-flash:medium__json_schema PASS 100% 85 in → 93 out (77 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4.1-flash:none PASS 100% 59 in → 5 out (0 reasoning)
git clean -n
deepseek/deepseek-v4.1-flash:none__json_schema PASS 100% 59 in → 12 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
deepseek/deepseek-v4.1-flash:xhigh PASS 100% 85 in → 138 out (137 reasoning)
git clean -n
deepseek/deepseek-v4.1-flash:xhigh__json_schema PASS 100% 85 in → 170 out (154 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemini-3-flash-preview:high PASS 100% 62 in → 167 out (163 reasoning)
git clean -n
google/gemini-3-flash-preview:high__json_schema PASS 100% 119 in → 178 out (169 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3-flash-preview:low PASS 100% 62 in → 315 out (311 reasoning)
git clean -n
google/gemini-3-flash-preview:low__json_schema PASS 100% 119 in → 282 out (274 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3-flash-preview:medium PASS 100% 62 in → 173 out (169 reasoning)
git clean -n
google/gemini-3-flash-preview:medium__json_schema PASS 100% 119 in → 241 out (233 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3-flash-preview:xhigh PASS 100% 62 in → 144 out (140 reasoning)
git clean -n
google/gemini-3-flash-preview:xhigh__json_schema PASS 100% 119 in → 431 out (423 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.1-flash-lite-preview:high PASS 100% 63 in → 345 out (341 reasoning)
git clean -n
google/gemini-3.1-flash-lite-preview:high__json_schema PASS 100% 63 in → 196 out (187 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.1-flash-lite-preview:low PASS 100% 62 in → 112 out (108 reasoning)
git clean -nd
google/gemini-3.1-flash-lite-preview:low__json_schema PASS 100% 63 in → 122 out (113 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command": "git clean -nd"}
google/gemini-3.1-flash-lite-preview:medium PASS 100% 63 in → 123 out (119 reasoning)
git clean -nd
google/gemini-3.1-flash-lite-preview:medium__json_schema PASS 100% 119 in → 177 out (163 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) { "command": "git clean -nd" }
google/gemini-3.1-flash-lite-preview:xhigh PASS 100% 63 in → 346 out (342 reasoning)
git clean -n
google/gemini-3.1-flash-lite-preview:xhigh__json_schema PASS 100% 63 in → 239 out (230 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.1-pro-preview:high PASS 100% 62 in → 182 out (178 reasoning)
git clean -n
google/gemini-3.1-pro-preview:high__json_schema PASS 100% 119 in → 192 out (184 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.1-pro-preview:low PASS 100% 62 in → 162 out (158 reasoning)
git clean -n
google/gemini-3.1-pro-preview:low__json_schema PASS 100% 119 in → 80 out (71 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.1-pro-preview:medium PASS 100% 62 in → 222 out (218 reasoning)
git clean -n
google/gemini-3.1-pro-preview:medium__json_schema PASS 100% 119 in → 225 out (216 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.1-pro-preview:xhigh PASS 100% 62 in → 520 out (516 reasoning)
git clean -n
google/gemini-3.1-pro-preview:xhigh__json_schema PASS 100% 119 in → 143 out (134 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.5-flash-lite:high PASS 100% 62 in → 294 out (290 reasoning)
git clean -n
google/gemini-3.5-flash-lite:high__json_schema PASS 100% 63 in → 265 out (256 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.5-flash-lite:low PASS 100% 62 in → 4 out (0 reasoning)
git clean -nd
google/gemini-3.5-flash-lite:low__json_schema PASS 100% 63 in → 9 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command": "git clean -nd"}
google/gemini-3.5-flash-lite:max PASS 100% 62 in → 311 out (307 reasoning)
git clean -n
google/gemini-3.5-flash-lite:max__json_schema PASS 100% 63 in → 245 out (237 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.5-flash-lite:medium PASS 100% 63 in → 207 out (203 reasoning)
git clean -n
google/gemini-3.5-flash-lite:medium__json_schema PASS 100% 119 in → 279 out (271 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.5-flash-lite:xhigh PASS 100% 62 in → 261 out (257 reasoning)
git clean -n
google/gemini-3.5-flash-lite:xhigh__json_schema PASS 100% 119 in → 258 out (250 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
google/gemini-3.5-flash:high PASS 100% 62 in → 346 out (342 reasoning)
git clean -n
google/gemini-3.5-flash:high__json_schema PASS 100% 119 in → 309 out (300 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.5-flash:low PASS 100% 62 in → 244 out (240 reasoning)
git clean -n
google/gemini-3.5-flash:low__json_schema PASS 100% 119 in → 99 out (90 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.5-flash:medium PASS 100% 62 in → 498 out (494 reasoning)
git clean -n
google/gemini-3.5-flash:medium__json_schema PASS 100% 119 in → 267 out (253 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemini-3.5-flash:xhigh PASS 100% 62 in → 399 out (395 reasoning)
git clean -n
google/gemini-3.5-flash:xhigh__json_schema PASS 100% 119 in → 298 out (289 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.6-flash:high PASS 100% 63 in → 277 out (273 reasoning)
git clean -n
google/gemini-3.6-flash:high__json_schema PASS 100% 119 in → 271 out (263 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.6-flash:low PASS 100% 63 in → 156 out (152 reasoning)
git clean -n
google/gemini-3.6-flash:low__json_schema PASS 100% 119 in → 8 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.6-flash:max PASS 100% 62 in → 271 out (267 reasoning)
git clean -n
google/gemini-3.6-flash:max__json_schema PASS 100% 63 in → 274 out (266 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.6-flash:medium PASS 100% 63 in → 208 out (204 reasoning)
git clean -n
google/gemini-3.6-flash:medium__json_schema PASS 100% 119 in → 281 out (273 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.6-flash:xhigh PASS 100% 63 in → 271 out (267 reasoning)
git clean -n
google/gemini-3.6-flash:xhigh__json_schema PASS 100% 63 in → 165 out (157 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.8-flash:high PASS 100% 62 in → 237 out (233 reasoning)
git clean -n
google/gemini-3.8-flash:high__json_schema PASS 100% 119 in → 263 out (255 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.8-flash:low PASS 100% 62 in → 4 out (0 reasoning)
git clean -n
google/gemini-3.8-flash:low__json_schema PASS 100% 119 in → 8 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.8-flash:max PASS 100% 62 in → 291 out (287 reasoning)
git clean -n
google/gemini-3.8-flash:max__json_schema PASS 100% 119 in → 341 out (333 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemini-3.8-flash:medium PASS 100% 62 in → 224 out (220 reasoning)
git clean -n
google/gemini-3.8-flash:medium__json_schema PASS 100% 119 in → 344 out (335 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemini-3.8-flash:xhigh PASS 100% 62 in → 133 out (129 reasoning)
git clean -n
google/gemini-3.8-flash:xhigh__json_schema PASS 100% 119 in → 291 out (283 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemma-4-26b-a4b-it:high PASS 100% 78 in → 200 out (195 reasoning)
git clean -n
google/gemma-4-26b-a4b-it:high__json_schema PASS 100% 77 in → 165 out (143 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemma-4-26b-a4b-it:low PASS 100% 78 in → 208 out (191 reasoning)
git clean -n
google/gemma-4-26b-a4b-it:low__json_schema PASS 100% 78 in → 441 out (382 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemma-4-26b-a4b-it:medium PASS 100% 78 in → 251 out (222 reasoning)
git clean -n
google/gemma-4-26b-a4b-it:medium__json_schema PASS 100% 78 in → 353 out (312 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemma-4-26b-a4b-it:none PASS 100% 75 in → 5 out (0 reasoning)
git clean -n
google/gemma-4-26b-a4b-it:none__json_schema PASS 100% 75 in → 15 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemma-4-26b-a4b-it:xhigh PASS 100% 78 in → 232 out (205 reasoning)
git clean -n
google/gemma-4-26b-a4b-it:xhigh__json_schema PASS 100% 77 in → 199 out (167 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemma-4-31b-it:high PASS 100% 78 in → 198 out (1 reasoning)
git clean -n
google/gemma-4-31b-it:high__json_schema PASS 100% 78 in → 165 out (138 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemma-4-31b-it:low PASS 100% 77 in → 122 out (102 reasoning)
git clean -n
google/gemma-4-31b-it:low__json_schema PASS 100% 78 in → 151 out (128 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
google/gemma-4-31b-it:medium PASS 100% 78 in → 153 out (136 reasoning)
git clean -n
google/gemma-4-31b-it:medium__json_schema PASS 100% 81 in → 121 out (111 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
google/gemma-4-31b-it:none PASS 100% 75 in → 5 out (0 reasoning)
git clean -n
google/gemma-4-31b-it:none__json_schema PASS 100% 75 in → 10 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
google/gemma-4-31b-it:xhigh PASS 100% 77 in → 139 out (124 reasoning)
git clean -n
google/gemma-4-31b-it:xhigh__json_schema PASS 100% 78 in → 166 out (144 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
ibm-granite/granite-4.1-8b PASS 100% 71 in → 5 out
git clean -n
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
ibm-granite/granite-4.2-8b:high PASS 100% 78 in → 1,425 out (1,509 reasoning)
git clean -n
ibm-granite/granite-4.2-8b:high__json_schema PASS 100% 78 in → 15 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
ibm-granite/granite-4.2-8b:low PASS 100% 87 in → 78 out (62 reasoning)
git clean -n -f
ibm-granite/granite-4.2-8b:low__json_schema PASS 100% 87 in → 10 out (0 reasoning)
git clean -nfd
JSON Schema Structured Output
(raw) {"command":"git clean -nfd"}
ibm-granite/granite-4.2-8b:max PASS 100% 78 in → 1,597 out (1,667 reasoning)
git clean -n
ibm-granite/granite-4.2-8b:max__json_schema PASS 100% 78 in → 15 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
ibm-granite/granite-4.2-8b:medium PASS 100% 87 in → 52 out (38 reasoning)
git clean -n -x
ibm-granite/granite-4.2-8b:medium__json_schema PASS 100% 87 in → 14 out (0 reasoning)
git clean -n -f
JSON Schema Structured Output
(raw) { "command": "git clean -n -f" }
ibm-granite/granite-4.2-8b:none PASS 100% 78 in → 4,701 out (0 reasoning)
git clean -nfd But the user said "Output ONLY the git command, nothing else." and the status shows untracked files: another.txt and untracked.txt, so the correct safe command to show what would be removed (including ignored? but ignored says none) is: git clean -ndf Wait: -n for dry-run, -d for directories, -f for force (but -n and -f can't be together? Actually -n implies dry-run, so -f is ignored in dry-run; -n -d -f is redundant but common usage is `git clean -nd` for directories without force, or `git clean -nfd`? Let's recall: - `git clean -n` shows what would be deleted (without actually doing it), applies to tracked files that are clean (safe), but for untracked files it needs `-f` or `-d`? Actually, by default `git clean` without `-f` only removes untracked files that are ignored. But the user says "Ignored files: (none)", so to remove untracked files (which are not ignored), you need `-f` (force) to remove them, because `git clean` by default doesn't remove untracked files; it only removes ignored files. Wait, the man page: - `-f`, `--force`: remove files (used with `-d`) - `-d`, `--directories`: recurse into directories - `-n`, `--dry-run`: show what would be deleted - `-i`, `--interactive`: interactive mode By default, `git clean` removes only untracked files that are ignored (i.e., match some .gitignore pattern), but since ignored is none, it wouldn't remove anything by default. To remove untracked files that are not ignored, you need `-f`. But `git clean -n` would show what would be removed by default (i.e., ignored files), which is none. But the user probably wants to see what would be removed if they run `git clean -f` (i.e., remove all untracked files), but since the status says "Untracked files: another.txt, untracked.txt" and "Ignored files: (none)", so to remove untracked files, need `-f`, and to list what would be removed, use `-n` with `-f`. The correct dry-run for removing untracked files is `git clean -ndf`? No: - `-n` means dry-run, so it shows what would be deleted without deleting. - To include untracked files in what would be deleted, you need `-f` because by default `git clean` doesn't touch untracked files (only ignored). Yes, from `git clean --help`: "Remove files that are ignored by Git." "Remove files that match a pattern in .gitignore or .git/info/exclude." "Use -f to force removal of files that match a pattern." "Use -n to see what would be deleted." So to see what would be deleted if you were to use `git clean -f`, you use `git clean -n -f`, but `-n` overrides `-f` in the sense that `-n` means don't actually delete, so `-n` and `-f` can be used together: `git clean -n -f` is valid and shows what would be deleted with force. Similarly, `-d` for directories, but here files are not directories (?? another.txt, ?? untracked.txt are files), so `-d` might not be necessary, but `-fd` includes directories, but since they're files, `-d` is fine. The safest to show all would-be removals (untracked files since ignored is none), so `git clean -nfd` or `git clean -ndf`? Order doesn't matter. Standard dry-run for clean: `git clean -nd` would only show ignored files (since `-d` doesn't include untracked non-ignored), but `-n` alone shows ignored files. To show untracked files that would be removed with `-f`, need `-f` in the dry-run. Yes: "With `-f`, `git clean` removes files that match patterns." So to simulate `-f`, use `-n` with `-f`. Thus `git clean -n -f` or `git clean -nf`, but `-d` might be needed if there are directories, but here only files, so `git clean -nfd` is safe (includes directories), but `-d` is harmless for files. The user said "without actually removing them", so dry-run, and they have untracked files not ignored, so to see what would be removed if they did `git clean -f`, the command is `git clean -n -f`, and often `-d` is added to recurse, but since files, `-d` not needed, but `-d` won't hurt. However, the simplest correct dry-run for all would-be removals (ignored + untracked with -f) is `git clean -ndf`? Let's check exact: - `git clean -n` shows ignored files (default behavior, no -f) - `git clean -n -f` shows both ignored and untracked files that match patterns (since -f allows removal of untracked) Since ignored is none, `-n` alone would show nothing, but we want to see what would be removed for untracked, so must include `-f`. Thus `git clean -n -f` or `git clean -nf`, but `-d` might be needed if untracked includes directories, but here it's files, so `-d` optional. But the status uses `??` for untracked files, which are files (not directories), so `-d` not required, but `git clean -nfd` is common to include directories. However, the minimal command that shows what would be removed for untracked files is `git clean -n -f`, because `-n` ensures no removal, `-f` enables considering untracked files for removal. Yes, from git docs: "If -f is used with -n, then `git clean` will show what would be deleted without actually deleting it." So the correct command is `git clean -n -f`, but often written as `git clean -nfd` if directories might be involved, but since status doesn't mention directories, `-d` might be unnecessary. But `-d` doesn't hurt in dry-run, and `-d` with `-n` shows directory recursion. But the user said "show what files would be removed", and files are mentioned, so `-d` might not be needed, but safest is to include `-d` because sometimes untracked could be directories, but here status shows `?? another.txt` and `?? untracked.txt`, both files, so `-d` is fine but not required. However, the precise command to show what `git clean -f` would remove (including directories) in dry-run is `git clean -nfd`. But `git clean -n -f` is sufficient for files, since `-d` is about recursing into directories; for files not in directories, `-d` doesn't change the set. Actually, `git clean -f` removes untracked files and directories matching patterns, `git clean -df` removes directories and their contents. To list all that would be removed (files and directories), use `-d` in clean. Since the status says "files" in "what files would be removed", but "files" might include directory entries, but likely they mean the untracked files listed, so `git clean -n -f` would list the untracked files, and `git clean -nfd` would also list them (same for files). The standard dry-run for `git clean` to see what gets removed (including untracked with -f) is `git clean -ndf` or `git clean -n -f -d`. Looking at typical usage: `git clean -nd` shows ignored files in directories, `git clean -n -f` shows ignored and untracked files. Since ignored is none, `git clean -n -f` would show the two untracked files. Is `-d` necessary? No, because `-d` is for recursing into directories; for files directly in root, `-d` not needed, but including `-d` won't show extra since no subdirs mentioned. To be precise and safe, `git clean -n -f` is enough, but `-d` might be expected if "files" could be in dirs, but status just lists files. However, the instruction is "Output ONLY the git command", so simplest correct is `git clean -n -f`, but `git clean -nfd` is also correct and common. Wait, the `git clean` help says: "-n, --dry-run: Do not actually remove anything. Instead, just print the list of files that would be removed." "By default, `git clean` only removes files that are ignored by Git." "With -f, `git clean` also removes files that are not ignored but match a pattern." Thus to see what `git clean -f` would remove, use `git clean -n -f`. Yes, e.g., `git clean -n -f` shows untracked files that would be removed with `-f`. `-d` is for directories; if you want to include directories, use `-d`, but for dry-run listing, `git clean -n -f` lists files and directories that `-f` would remove (since `-f` alone removes files and directories matching patterns), and `-d` is for recursing into directories during clean, but `-f` with `-d` cleans directories, while `-f` alone cleans files. Actually, `-d` alone with `-n` shows ignored directories; `-f` with `-n` shows ignored files; to see non-ignored (untracked) files, need `-f`. Thus the dry-run command that shows what would be removed if `-f` is used (i.e., typical `git clean -f` to remove untracked), is `git clean -n -f`. Adding `-d` would also consider directory recursion, but since no directory mentioned in status, it's fine, but `git clean -nfd` is often used. However, the minimal necessary for the given status (two untracked files, no ignored) to see what `git clean -f` would remove is `git clean -n -f`, but to also handle if there were directories, `-d` is safer, but the user said "files", and status shows `??` for files (not `??` for directories), so probably files. But the safest dry-run command that matches "what files would be removed by git clean" — but "git clean" without args only removes ignored, which is none, so likely they mean with `-f`, as common to remove untracked. The user said "Show what files would be removed by git clean without actually removing them." Since "git clean" by default doesn't remove untracked, only ignored, which is none, so it would remove nothing, but that's probably not what they mean; they likely mean the usual `git clean -f` to remove untracked. In the context, status shows "Untracked files", "Ignored files: (none)", so "what files would be removed" probably refers to what `git clean -f` would remove, i.e., the untracked files. Thus the correct dry-run command is `git clean -n -f`, which will list the untracked files. `git clean -nfd` would also list them, and `-d` doesn't hurt, but `-n -f` is sufficient. However, `git clean -n` alone would show nothing (since ignored none), which is not useful, so must include `-f`. Yes, so the command should include `-f` for dry-run to see untracked. Thus `git clean -n -f` is correct, and `-d` is optional; but `-fd` is common. Looking at actual git behavior: $ touch a b $ git add a $ touch c # untracked $ git clean -n # shows nothing, since c not ignored $ git clean -n -f # shows c $ git clean -n -d -f # same as above for files Since status has `?? another.txt` and `?? untracked.txt`, which are untracked files (not added), `git clean -n -f` will list them. Thus the precise command to output what would be removed (the untracked files) in dry-run is `git clean -n -f`. But sometimes people use `-ndf`, but `-n` implies dry-run, and `-f` enables removing non-ignored, so `git clean -n -f` is accurate. The flags can be in any order, so `git clean -nfd` is also fine (`-n`, `-f`, `-d`). `git clean -nfd` means dry-run, directories, force — which will list ignored directories (none) and untracked files/dirs (with force), which matches. Given the status doesn't mention directories, both work, but `-d` might be unnecessary, but including it is harmless. However, the minimal command that works is `git clean -n -f`, but to be thorough for directories, `git clean -nfd` is safer. But the user said "files" (plural), and status shows files, so probably `git clean -n -f` suffices. But let's recall the exact wording: "Show what files would be removed by git clean" — `git clean` without options only removes ignored, so "would be removed by git clean" could be ambiguous, but since ignored is none, it would remove nothing, but that seems not the intent. The status context implies they want to know what untracked files would be removed, so likely with `-f`. In practice, when someone says "git clean" in the context of removing untracked, they mean `git clean -f`, so dry-run for that is `git clean -n -f`. Thus the answer should be `git clean -n -f`, but often `-d` is added; however, `-d` is for directories, and since no dir mentioned, `-d` not needed. But `git clean -n` alone would not list untracked, so must have `-f`. Yes, so `git clean -n -f` is the simplest and correct for showing untracked files that would be removed with `-f`. `git clean -nfd` also works and is common, but `-d` might list directories if present, but not here. However, the instruction is to output ONLY the git command, nothing else, so choose the most appropriate. I think `git clean -n -f` is precise for removing untracked (since ignored none), and `-n` ensures no removal. `git clean -nfd` is also correct, but `-d` might be redundant for files. But to be safe with directories (in case `??` could be a dir, though status shows `.txt`), `-d` is useful, but the status uses `?? filename.txt`, which is typically a file. In git status, `??` can be a file or directory? No, `??` is for untracked files and directories; for directories, it might show `?? dir/`, but status says "?? another.txt" and "?? untracked.txt", so likely files. Thus `git clean -n -f` is sufficient. But let's check with actual git: $ touch another.txt untracked.txt $ git status Untracked files: another.txt untracked.txt $ git clean -n # nothing $ git clean -n -f # shows untracked files $ git clean -n -d # shows ignored directories? none $ git clean -n -d -f # shows untracked files and ignored dirs (but ignored none), so same as -n -f for files Thus `git clean -n -f` directly shows the untracked files. Yes, so `git clean -n -f` is the right dry-run for what `git clean -f` would remove. Therefore, the command is `git clean -n -f` But `-n` and `-f` must be both used; `-n` without `-f` shows ignored only, which is none, so not useful, but the user wants to know what would be removed (implied by common usage), so include `-f`. Thus output `git clean -n -f` However, `-n` comes before `-f` usually, but order doesn't matter. Standard might be `git clean -ndf`, but `-n` first. I think `git clean -n -f` is clear. But to match common "dry-run clean", often `git clean -nd` for ignored, `git clean -nfd` for all, but since ignored none, `-n -f` is fine. Given the status says "Untracked files" and "Ignored files: (none)", the user probably intends to remove untracked, so `git clean -n -f` Yes. Thus the git command is `git clean -n -f` But the `-f` is for force, which in dry-run just lists what would be removed. Yes. So output: `git clean -n -f` But let's confirm with the exact wording: "without actually removing them" — `git clean -n` alone would not remove ignored, but since ignored none, it shows nothing, which is incorrect for what they likely want. Whereas `git clean -n -f` shows the untracked files. Thus `git clean -n -f` Alternatively, `git clean -nf` (without space), but usually `git clean -n -f` or `git clean -nf`, both valid. In git, flags can be combined: `git clean -n -f` or `git clean -nf` Yes, `-n` and `-f` are separate flags but can be abbreviated with combined form? No, `-nf` is not standard; standard is `-n` and `-f` separately or `-n -f`, but `-nf` might not be recognized? Let's recall: in git, usually you write `git clean -n -f`, but `-n` and `-f` are independent, so `git clean -nf` should work as short for `-n -f`. Yes, e.g., `git commit -m` works, so `git clean -n -f` or `git clean -nf` both work; `-nf` is common shorthand. For example, `git clean -d -f` is `git clean -df`, etc. So `git clean -n -f` or `git clean -nf` are both fine, but `git clean -nf` is shorter. But to be precise and clear, `git clean -n -f` is safe, but the instruction says "Output ONLY the git command", so either is fine, but commonly `git clean -n -f` or `git clean -nfd`. However, since `-d` is not needed, `git clean -nf` might be minimal. But `-n` implies dry-run, `-f` enables force (remove untracked), so `git clean -nf` is correct. Yes, `git help clean` shows: -n, --dry-run Dry run -f, --force Remove files (used with -d) So for dry-run with force, `--dry-run --force` or `-n -f`. Thus `git clean -n -f` is accurate. I'll go with `git clean -n -f`, but often written as `git clean -nfd` if directories might be involved; however, status shows files, not directories. To strictly follow "what files would be removed", and files are listed, `-d` not necessary, so `git clean -n -f` But `git clean -n` without `-f` would output nothing (since ignored none), which is wrong if they think of removing untracked, so must include `-f`. Yes. Thus the command is `git clean -n -f` But let's see if `-f` is required in dry-run: yes, because by default clean only removes ignored. Since ignored is none, default clean removes nothing, but user likely means remove untracked, so specify dry-run for `-f`. Thus `git clean -n -f` Final answer: `git clean -n -f` But in some shells, spaces might be required, but as a command string, `git clean -n -f` is correct. I think that's it. git clean -n -f
ibm-granite/granite-4.2-8b:none__json_schema PASS 100% 78 in → 14 out (0 reasoning)
git clean -n -d
JSON Schema Structured Output
(raw) { "command": "git clean -n -d" }
ibm-granite/granite-4.2-8b:xhigh PASS 100% 78 in → 4,971 out (5,125 reasoning)
git clean -f -n
ibm-granite/granite-4.2-8b:xhigh__json_schema PASS 100% 78 in → 16 out (0 reasoning)
git clean -nfd
JSON Schema Structured Output
(raw) { "command": "git clean -nfd" }
inception/mercury-2.5:high PASS 100% 52 in → 431 out (421 reasoning)
git clean -n
inception/mercury-2.5:high__json_schema PASS 100% 251 in → 1,185 out (1,137 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
inception/mercury-2.5:low PASS 100% 52 in → 266 out (256 reasoning)
git clean -n
inception/mercury-2.5:low__json_schema PASS 100% 255 in → 329 out (253 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
inception/mercury-2.5:max PASS 100% 52 in → 383 out (372 reasoning)
git clean -n
inception/mercury-2.5:max__json_schema PASS 100% 256 in → 1,244 out (1,201 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
inception/mercury-2.5:medium PASS 100% 52 in → 290 out (281 reasoning)
git clean -n
inception/mercury-2.5:medium__json_schema PASS 100% 251 in → 1,178 out (1,132 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
inception/mercury-2.5:none PASS 100% 54 in → 7 out (0 reasoning)
git clean -n
inception/mercury-2.5:none__json_schema PASS 100% 253 in → 88 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
inception/mercury-2.5:xhigh PASS 100% 53 in → 452 out (441 reasoning)
git clean -n
inception/mercury-2.5:xhigh__json_schema PASS 100% 251 in → 1,997 out (1,948 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
inclusionai/ling-2.6-flash PASS 100% 77 in → 10 out
```sh git clean -n ```
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
liquid/lfm-2-24b-a2b PASS 100% 66 in → 12 out
git clean -dn --exclude="*.txt"
meta/muse-spark-1.1:high PASS 100% 59 in → 663 out (649 reasoning)
git clean -n
meta/muse-spark-1.1:high__json_schema PASS 100% 59 in → 840 out (819 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
meta/muse-spark-1.1:low PASS 100% 59 in → 350 out (336 reasoning)
git clean -n
meta/muse-spark-1.1:low__json_schema PASS 100% 59 in → 289 out (268 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
meta/muse-spark-1.1:max PASS 100% 59 in → 728 out (714 reasoning)
git clean -n
meta/muse-spark-1.1:max__json_schema PASS 100% 59 in → 666 out (641 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
meta/muse-spark-1.1:medium PASS 100% 59 in → 574 out (560 reasoning)
git clean -n
meta/muse-spark-1.1:medium__json_schema PASS 100% 59 in → 574 out (553 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
meta/muse-spark-1.1:xhigh PASS 100% 59 in → 633 out (619 reasoning)
git clean -n
meta/muse-spark-1.1:xhigh__json_schema PASS 100% 59 in → 865 out (844 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
meta/muse-spark-1.3-contributor:high PASS 100% 59 in → 542 out (507 reasoning)
git clean -n
meta/muse-spark-1.3-contributor:high__json_schema PASS 100% 59 in → 851 out (789 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
meta/muse-spark-1.3-contributor:low PASS 100% 59 in → 506 out (473 reasoning)
git clean -n
meta/muse-spark-1.3-contributor:low__json_schema PASS 100% 59 in → 508 out (473 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
meta/muse-spark-1.3-contributor:medium PASS 100% 59 in → 319 out (305 reasoning)
git clean -n
meta/muse-spark-1.3-contributor:medium__json_schema PASS 100% 59 in → 813 out (752 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
meta/muse-spark-1.3-contributor:xhigh PASS 100% 59 in → 529 out (494 reasoning)
git clean -n
meta/muse-spark-1.3-contributor:xhigh__json_schema PASS 100% 59 in → 1,260 out (1,203 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
minimax/minimax-m2.5:high PASS 100% 89 in → 155 out (150 reasoning)
git clean -n
minimax/minimax-m2.5:high__json_schema PASS 100% 89 in → 851 out (839 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
minimax/minimax-m2.5:low PASS 100% 89 in → 116 out (141 reasoning)
git clean -n
minimax/minimax-m2.5:low__json_schema PASS 100% 89 in → 120 out (119 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
minimax/minimax-m2.5:medium PASS 100% 89 in → 85 out (90 reasoning)
git clean -n
minimax/minimax-m2.5:medium__json_schema PASS 100% 89 in → 215 out (221 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
minimax/minimax-m2.5:xhigh PASS 100% 89 in → 229 out (223 reasoning)
git clean -n
minimax/minimax-m2.5:xhigh__json_schema PASS 100% 89 in → 157 out (145 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
minimax/minimax-m2.7:high PASS 100% 89 in → 255 out (277 reasoning)
git clean -n
minimax/minimax-m2.7:high__json_schema PASS 100% 89 in → 13 out (13 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
minimax/minimax-m2.7:low PASS 100% 89 in → 349 out (343 reasoning)
git clean -n
minimax/minimax-m2.7:low__json_schema PASS 100% 221 in → 314 out (304 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
minimax/minimax-m2.7:medium PASS 100% 92 in → 213 out (208 reasoning)
git clean -n
minimax/minimax-m2.7:medium__json_schema PASS 100% 221 in → 314 out (304 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
minimax/minimax-m2.7:xhigh PASS 100% 89 in → 296 out (289 reasoning)
git clean -n
minimax/minimax-m2.7:xhigh__json_schema PASS 100% 221 in → 204 out (194 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
minimax/minimax-m3:high PASS 100% 227 in → 42 out (39 reasoning)
git clean -n
minimax/minimax-m3:high__json_schema PASS 100% 214 in → 10 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
minimax/minimax-m3:low PASS 100% 227 in → 42 out (39 reasoning)
git clean -nd
minimax/minimax-m3:low__json_schema PASS 100% 214 in → 13 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
minimax/minimax-m3:medium PASS 100% 227 in → 50 out (45 reasoning)
git clean -n
minimax/minimax-m3:medium__json_schema PASS 100% 214 in → 12 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
minimax/minimax-m3:xhigh PASS 100% 227 in → 50 out (45 reasoning)
git clean -n
minimax/minimax-m3:xhigh__json_schema PASS 100% 214 in → 12 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
mistralai/devstral-2512 PASS 100% 58 in → 10 out
```bash git clean -n ```
mistralai/devstral-2512__json_schema PASS 100% 58 in → 10 out
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
mistralai/mistral-medium-3-5:high PASS 100% 70 in → 272 out (266 reasoning)
git clean -n
mistralai/mistral-medium-3-5:high__json_schema PASS 100% 58 in → 264 out (265 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
mistralai/mistral-medium-3-5:low PASS 100% 70 in → 257 out (273 reasoning)
git clean -n
mistralai/mistral-medium-3-5:low__json_schema PASS 100% 58 in → 177 out (190 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
mistralai/mistral-medium-3-5:medium PASS 100% 70 in → 5,921 out (6,555 reasoning)
git clean -n
mistralai/mistral-medium-3-5:medium__json_schema PASS 100% 58 in → 243 out (243 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
mistralai/mistral-medium-3-5:none PASS 100% 70 in → 5 out (0 reasoning)
git clean -n
mistralai/mistral-medium-3-5:none__json_schema PASS 100% 58 in → 10 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
mistralai/mistral-medium-3-5:xhigh PASS 100% 70 in → 627 out (693 reasoning)
git clean -n
mistralai/mistral-medium-3-5:xhigh__json_schema PASS 100% 58 in → 168 out (175 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
moonshotai/kimi-k2.6:high PASS 100% 61 in → 1,296 out (1,292 reasoning)
git clean -n
moonshotai/kimi-k2.6:high__json_schema PASS 100% 61 in → 570 out (557 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
moonshotai/kimi-k2.6:low PASS 100% 61 in → 819 out (813 reasoning)
git clean -n
moonshotai/kimi-k2.6:low__json_schema PASS 100% 61 in → 871 out (862 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
moonshotai/kimi-k2.6:medium PASS 100% 61 in → 449 out (443 reasoning)
git clean -n
moonshotai/kimi-k2.6:medium__json_schema PASS 100% 60 in → 1,864 out (1,721 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
moonshotai/kimi-k2.6:xhigh PASS 100% 61 in → 732 out (726 reasoning)
git clean -n
moonshotai/kimi-k2.6:xhigh__json_schema PASS 100% 61 in → 726 out (717 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
moonshotai/kimi-k2.7-code:high PASS 100% 61 in → 150 out (144 reasoning)
git clean -n
moonshotai/kimi-k2.7-code:high__json_schema PASS 100% 61 in → 653 out (669 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
moonshotai/kimi-k2.7-code:low PASS 100% 60 in → 183 out (177 reasoning)
git clean -n
moonshotai/kimi-k2.7-code:low__json_schema PASS 100% 144 in → 59 out (48 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
moonshotai/kimi-k2.7-code:medium PASS 100% 61 in → 170 out (164 reasoning)
git clean -n
moonshotai/kimi-k2.7-code:medium__json_schema PASS 100% 61 in → 55 out (42 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
moonshotai/kimi-k2.7-code:xhigh PASS 100% 61 in → 225 out (217 reasoning)
git clean -n -d
moonshotai/kimi-k2.7-code:xhigh__json_schema PASS 100% 61 in → 272 out (258 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
moonshotai/kimi-k3:high PASS 100% 138 in → 371 out (352 reasoning)
git clean -n
moonshotai/kimi-k3:high__json_schema PASS 100% 138 in → 149 out (130 reasoning)
git clean -n
moonshotai/kimi-k3:low PASS 100% 138 in → 210 out (188 reasoning)
``` git clean -n ```
moonshotai/kimi-k3:low__json_schema PASS 100% 138 in → 172 out (153 reasoning)
git clean -n
moonshotai/kimi-k3:max PASS 100% 138 in → 81 out (62 reasoning)
git clean -n
moonshotai/kimi-k3:max__json_schema PASS 100% 138 in → 169 out (147 reasoning)
``` git clean -n ```
moonshotai/kimi-k3:medium PASS 100% 138 in → 185 out (166 reasoning)
git clean -n
moonshotai/kimi-k3:medium__json_schema PASS 100% 138 in → 99 out (80 reasoning)
git clean -n
moonshotai/kimi-k3:xhigh PASS 100% 138 in → 251 out (232 reasoning)
git clean -n
moonshotai/kimi-k3:xhigh__json_schema PASS 100% 138 in → 133 out (114 reasoning)
git clean -n
nvidia/nemotron-3-nano-30b-a3b:high PASS 100% 71 in → 217 out (224 reasoning)
git clean -n
nvidia/nemotron-3-nano-30b-a3b:high__json_schema PASS 100% 71 in → 313 out (319 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
nvidia/nemotron-3-nano-30b-a3b:low PASS 100% 71 in → 130 out (136 reasoning)
git clean -n
nvidia/nemotron-3-nano-30b-a3b:low__json_schema PASS 100% 71 in → 223 out (233 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
nvidia/nemotron-3-nano-30b-a3b:medium PASS 100% 71 in → 157 out (167 reasoning)
git clean -n
nvidia/nemotron-3-nano-30b-a3b:medium__json_schema PASS 100% 71 in → 234 out (254 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
nvidia/nemotron-3-nano-30b-a3b:none PASS 100% 71 in → 14 out (0 reasoning)
git clean -n -d -x -e .gitignore
nvidia/nemotron-3-nano-30b-a3b:none__json_schema PASS 100% 71 in → 14 out (0 reasoning)
git clean -nxd
JSON Schema Structured Output
(raw) { "command": "git clean -nxd" }
nvidia/nemotron-3-nano-30b-a3b:xhigh PASS 100% 71 in → 212 out (219 reasoning)
git clean -n
nvidia/nemotron-3-nano-30b-a3b:xhigh__json_schema PASS 100% 71 in → 170 out (174 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
nvidia/nemotron-3-super-120b-a12b:high PASS 100% 71 in → 116 out (105 reasoning)
git clean -n
nvidia/nemotron-3-super-120b-a12b:high__json_schema PASS 100% 71 in → 85 out (69 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
nvidia/nemotron-3-super-120b-a12b:low PASS 100% 71 in → 78 out (71 reasoning)
git clean -n
nvidia/nemotron-3-super-120b-a12b:low__json_schema PASS 100% 71 in → 145 out (127 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
nvidia/nemotron-3-super-120b-a12b:medium PASS 100% 71 in → 160 out (149 reasoning)
git clean -n
nvidia/nemotron-3-super-120b-a12b:medium__json_schema PASS 100% 71 in → 153 out (139 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
nvidia/nemotron-3-super-120b-a12b:none PASS 100% 71 in → 5 out (0 reasoning)
git clean -nd
nvidia/nemotron-3-super-120b-a12b:none__json_schema PASS 100% 71 in → 13 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
nvidia/nemotron-3-super-120b-a12b:xhigh PASS 100% 71 in → 89 out (77 reasoning)
git clean -n
nvidia/nemotron-3-super-120b-a12b:xhigh__json_schema PASS 100% 71 in → 187 out (171 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
openai/gpt-5.4-mini:high PASS 100% 57 in → 131 out (121 reasoning)
git clean -n
openai/gpt-5.4-mini:high__json_schema PASS 100% 92 in → 109 out (91 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.4-mini:low PASS 100% 57 in → 78 out (68 reasoning)
git clean -n
openai/gpt-5.4-mini:low__json_schema PASS 100% 92 in → 73 out (55 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.4-mini:medium PASS 100% 57 in → 68 out (58 reasoning)
git clean -n
openai/gpt-5.4-mini:medium__json_schema PASS 100% 92 in → 88 out (70 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.4-mini:none PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.4-mini:none__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.4-mini:xhigh PASS 100% 57 in → 226 out (216 reasoning)
git clean -n
openai/gpt-5.4-mini:xhigh__json_schema PASS 100% 92 in → 304 out (286 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.4-nano:high PASS 100% 57 in → 8 out (0 reasoning)
git clean -nd
openai/gpt-5.4-nano:high__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.4-nano:low PASS 100% 57 in → 8 out (0 reasoning)
git clean -nd
openai/gpt-5.4-nano:low__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.4-nano:medium PASS 100% 57 in → 8 out (0 reasoning)
git clean -nd
openai/gpt-5.4-nano:medium__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.4-nano:none PASS 100% 57 in → 8 out (0 reasoning)
git clean -nd
openai/gpt-5.4-nano:none__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.4-nano:xhigh PASS 100% 57 in → 196 out (186 reasoning)
git clean -n
openai/gpt-5.4-nano:xhigh__json_schema PASS 100% 92 in → 267 out (249 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.4:high PASS 100% 57 in → 95 out (85 reasoning)
git clean -n
openai/gpt-5.4:high__json_schema PASS 100% 92 in → 177 out (159 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.4:low PASS 100% 57 in → 31 out (21 reasoning)
git clean -nd
openai/gpt-5.4:low__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.4:medium PASS 100% 57 in → 86 out (76 reasoning)
git clean -n
openai/gpt-5.4:medium__json_schema PASS 100% 92 in → 79 out (61 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.4:none PASS 100% 57 in → 8 out (0 reasoning)
git clean -nd
openai/gpt-5.4:none__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.4:xhigh PASS 100% 57 in → 92 out (82 reasoning)
git clean -n
openai/gpt-5.4:xhigh__json_schema PASS 100% 92 in → 534 out (516 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.5:high PASS 100% 57 in → 162 out (152 reasoning)
git clean -n
openai/gpt-5.5:high__json_schema PASS 100% 92 in → 69 out (51 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.5:low PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.5:low__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.5:medium PASS 100% 57 in → 63 out (53 reasoning)
git clean -n
openai/gpt-5.5:medium__json_schema PASS 100% 92 in → 17 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
openai/gpt-5.5:none PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.5:none__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.5:xhigh PASS 100% 57 in → 225 out (215 reasoning)
git clean -n
openai/gpt-5.5:xhigh__json_schema PASS 100% 92 in → 176 out (158 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-luna:high PASS 100% 57 in → 66 out (56 reasoning)
git clean -n
openai/gpt-5.6-luna:high__json_schema PASS 100% 92 in → 77 out (59 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-luna:low PASS 100% 57 in → 56 out (46 reasoning)
git clean -n
openai/gpt-5.6-luna:low__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-luna:max PASS 100% 57 in → 75 out (65 reasoning)
git clean -n
openai/gpt-5.6-luna:max__json_schema PASS 100% 92 in → 534 out (516 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-luna:medium PASS 100% 57 in → 48 out (38 reasoning)
git clean -n
openai/gpt-5.6-luna:medium__json_schema PASS 100% 92 in → 68 out (50 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-luna:none PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-luna:none__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-luna:xhigh PASS 100% 57 in → 79 out (69 reasoning)
git clean -n
openai/gpt-5.6-luna:xhigh__json_schema PASS 100% 92 in → 71 out (53 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-sol:high PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-sol:high__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-sol:low PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-sol:low__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-sol:max PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-sol:max__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-sol:medium PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-sol:medium__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-sol:none PASS 100% 57 in → 10 out (0 reasoning)
git clean -n -f
openai/gpt-5.6-sol:none__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-sol:xhigh PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-sol:xhigh__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-terra:high PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-terra:high__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-terra:low PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-terra:low__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.6-terra:max PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-terra:max__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-terra:medium PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-terra:medium__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-5.6-terra:none PASS 100% 57 in → 8 out (0 reasoning)
git clean -nd
openai/gpt-5.6-terra:none__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
openai/gpt-5.6-terra:xhigh PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-5.6-terra:xhigh__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-astra:high PASS 100% 57 in → 39 out (29 reasoning)
git clean -n
openai/gpt-6-astra:high__json_schema PASS 100% 92 in → 62 out (44 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-astra:low PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-6-astra:low__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-astra:max PASS 100% 57 in → 120 out (110 reasoning)
git clean -n
openai/gpt-6-astra:max__json_schema PASS 100% 92 in → 198 out (180 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-astra:medium PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-6-astra:medium__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-astra:xhigh PASS 100% 57 in → 51 out (41 reasoning)
git clean -n
openai/gpt-6-astra:xhigh__json_schema PASS 100% 92 in → 105 out (87 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-luna:high PASS 100% 57 in → 96 out (86 reasoning)
git clean -n
openai/gpt-6-luna:high__json_schema PASS 100% 92 in → 58 out (40 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-luna:low PASS 100% 57 in → 47 out (36 reasoning)
git clean -n
openai/gpt-6-luna:low__json_schema PASS 100% 92 in → 67 out (49 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-luna:max PASS 100% 57 in → 91 out (81 reasoning)
git clean -n
openai/gpt-6-luna:max__json_schema PASS 100% 92 in → 96 out (78 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-luna:medium PASS 100% 57 in → 54 out (43 reasoning)
git clean -n
openai/gpt-6-luna:medium__json_schema PASS 100% 92 in → 106 out (88 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-luna:xhigh PASS 100% 57 in → 92 out (82 reasoning)
git clean -n
openai/gpt-6-luna:xhigh__json_schema PASS 100% 92 in → 67 out (49 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-sol:high PASS 100% 57 in → 38 out (28 reasoning)
git clean -n
openai/gpt-6-sol:high__json_schema PASS 100% 92 in → 136 out (118 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-sol:low PASS 100% 57 in → 8 out (0 reasoning)
git clean -n
openai/gpt-6-sol:low__json_schema PASS 100% 92 in → 16 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-sol:max PASS 100% 57 in → 129 out (119 reasoning)
git clean -n
openai/gpt-6-sol:max__json_schema PASS 100% 92 in → 261 out (243 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-sol:medium PASS 100% 57 in → 36 out (26 reasoning)
git clean -n
openai/gpt-6-sol:medium__json_schema PASS 100% 92 in → 128 out (110 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-6-sol:xhigh PASS 100% 57 in → 76 out (66 reasoning)
git clean -n
openai/gpt-6-sol:xhigh__json_schema PASS 100% 92 in → 160 out (142 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-oss-120b:high PASS 100% 118 in → 372 out (408 reasoning)
git clean -n
openai/gpt-oss-120b:high__json_schema PASS 100% 118 in → 541 out (526 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-oss-120b:low PASS 100% 105 in → 75 out (51 reasoning)
git clean -n -d
openai/gpt-oss-120b:medium PASS 100% 116 in → 154 out (148 reasoning)
git clean -n
openai/gpt-oss-120b:medium__json_schema PASS 100% 105 in → 130 out (128 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-oss-120b:xhigh PASS 100% 118 in → 477 out (459 reasoning)
git clean -n
openai/gpt-oss-120b:xhigh__json_schema PASS 100% 105 in → 232 out (242 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-oss-20b:high PASS 100% 116 in → 268 out (282 reasoning)
git clean -n
openai/gpt-oss-20b:high__json_schema PASS 100% 105 in → 333 out (352 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
openai/gpt-oss-20b:low PASS 100% 123 in → 32 out (16 reasoning)
git clean -n -d
openai/gpt-oss-20b:low__json_schema PASS 100% 118 in → 51 out (31 reasoning)
git clean -n -d
JSON Schema Structured Output
(raw) {"command":"git clean -n -d"}
openai/gpt-oss-20b:medium PASS 100% 118 in → 202 out (198 reasoning)
git clean -n
openai/gpt-oss-20b:medium__json_schema PASS 100% 218 in → 228 out (207 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
openai/gpt-oss-20b:xhigh PASS 100% 105 in → 326 out (334 reasoning)
git clean -n
openai/gpt-oss-20b:xhigh__json_schema PASS 100% 118 in → 12 out (0 reasoning)
git clean -n -d -f
JSON Schema Structured Output
(raw) {"command":"git clean -n -d -f"}
poolside/laguna-m.1:high PASS 100% 70 in → 260 out (253 reasoning)
git clean -n
poolside/laguna-m.1:high__json_schema PASS 100% 70 in → 396 out (385 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
poolside/laguna-m.1:low PASS 100% 70 in → 243 out (236 reasoning)
git clean -n
poolside/laguna-m.1:low__json_schema PASS 100% 70 in → 629 out (619 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
poolside/laguna-m.1:medium PASS 100% 70 in → 502 out (495 reasoning)
git clean -nd
poolside/laguna-m.1:medium__json_schema PASS 100% 70 in → 408 out (393 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
poolside/laguna-m.1:none PASS 100% 70 in → 8 out (0 reasoning)
git clean -d -n
poolside/laguna-m.1:none__json_schema PASS 100% 70 in → 10 out (0 reasoning)
git clean -nX
JSON Schema Structured Output
(raw) {"command": "git clean -nX"}
poolside/laguna-m.1:xhigh PASS 100% 70 in → 455 out (448 reasoning)
git clean -n
poolside/laguna-m.1:xhigh__json_schema PASS 100% 70 in → 357 out (341 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
poolside/laguna-xs-2.1:high PASS 100% 70 in → 852 out (847 reasoning)
git clean -n
poolside/laguna-xs-2.1:high__json_schema PASS 100% 70 in → 295 out (279 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
poolside/laguna-xs-2.1:low PASS 100% 70 in → 204 out (199 reasoning)
git clean -n
poolside/laguna-xs-2.1:low__json_schema PASS 100% 70 in → 444 out (433 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
poolside/laguna-xs-2.1:medium PASS 100% 70 in → 149 out (144 reasoning)
git clean -n
poolside/laguna-xs-2.1:medium__json_schema PASS 100% 70 in → 438 out (422 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
poolside/laguna-xs-2.1:none PASS 100% 70 in → 4 out (0 reasoning)
git clean -n
poolside/laguna-xs-2.1:none__json_schema PASS 100% 70 in → 12 out (0 reasoning)
git clean -n -d
JSON Schema Structured Output
(raw) {"command": "git clean -n -d"}
poolside/laguna-xs-2.1:xhigh PASS 100% 70 in → 209 out (204 reasoning)
git clean -n
poolside/laguna-xs-2.1:xhigh__json_schema PASS 100% 70 in → 1,283 out (1,270 reasoning)
git clean -n -f
JSON Schema Structured Output
(raw) {"command": "git clean -n -f"}
poolside/laguna-xs.2:high PASS 100% 107 in → 170 out (163 reasoning)
git clean -n
poolside/laguna-xs.2:high__json_schema PASS 100% 107 in → 115 out (99 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
poolside/laguna-xs.2:low PASS 100% 107 in → 310 out (303 reasoning)
git clean -n
poolside/laguna-xs.2:low__json_schema PASS 100% 107 in → 96 out (85 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
poolside/laguna-xs.2:medium PASS 100% 107 in → 84 out (77 reasoning)
git clean -n
poolside/laguna-xs.2:medium__json_schema PASS 100% 107 in → 326 out (315 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
poolside/laguna-xs.2:none PASS 100% 107 in → 6 out (0 reasoning)
git clean -n
poolside/laguna-xs.2:none__json_schema PASS 100% 107 in → 10 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
poolside/laguna-xs.2:xhigh PASS 100% 107 in → 278 out (271 reasoning)
git clean -n
poolside/laguna-xs.2:xhigh__json_schema PASS 100% 107 in → 116 out (100 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
prism-ml/ternary-bonsai-2-27b:high PASS 100% 66 in → 87 out (80 reasoning)
git clean -n
prism-ml/ternary-bonsai-2-27b:high__json_schema PASS 100% 160 in → 139 out (128 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
prism-ml/ternary-bonsai-2-27b:low PASS 100% 66 in → 140 out (133 reasoning)
git clean -n
prism-ml/ternary-bonsai-2-27b:low__json_schema PASS 100% 160 in → 128 out (117 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
prism-ml/ternary-bonsai-2-27b:max PASS 100% 108 in → 147 out (140 reasoning)
git clean -n
prism-ml/ternary-bonsai-2-27b:max__json_schema PASS 100% 198 in → 179 out (168 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
prism-ml/ternary-bonsai-2-27b:medium PASS 100% 66 in → 153 out (146 reasoning)
git clean -fn
prism-ml/ternary-bonsai-2-27b:medium__json_schema PASS 100% 160 in → 87 out (76 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
prism-ml/ternary-bonsai-2-27b:none PASS 100% 68 in → 5 out (0 reasoning)
git clean -n
prism-ml/ternary-bonsai-2-27b:none__json_schema PASS 100% 162 in → 9 out (0 reasoning)
git clean -nd
JSON Schema Structured Output
(raw) {"command":"git clean -nd"}
prism-ml/ternary-bonsai-2-27b:xhigh PASS 100% 108 in → 129 out (122 reasoning)
git clean -n
prism-ml/ternary-bonsai-2-27b:xhigh__json_schema PASS 100% 198 in → 183 out (166 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.6-27b:high PASS 100% 66 in → 996 out (1 reasoning)
git clean -n
qwen/qwen3.6-27b:high__json_schema PASS 100% 66 in → 913 out (1 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.6-27b:low PASS 100% 66 in → 497 out (466 reasoning)
git clean -n
qwen/qwen3.6-27b:low__json_schema PASS 100% 66 in → 698 out (652 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.6-27b:medium PASS 100% 66 in → 1,254 out (1,209 reasoning)
git clean -n
qwen/qwen3.6-27b:medium__json_schema PASS 100% 68 in → 10 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
qwen/qwen3.6-27b:none PASS 100% 68 in → 5 out (0 reasoning)
git clean -n
qwen/qwen3.6-27b:none__json_schema PASS 100% 68 in → 17 out (0 reasoning)
git clean -n -f
JSON Schema Structured Output
(raw) { "command": "git clean -n -f" }
qwen/qwen3.6-27b:xhigh PASS 100% 66 in → 1,031 out (1,022 reasoning)
git clean -n
qwen/qwen3.6-27b:xhigh__json_schema PASS 100% 66 in → 933 out (898 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.6-35b-a3b:high PASS 100% 66 in → 1,085 out (1,072 reasoning)
git clean -n
qwen/qwen3.6-35b-a3b:high__json_schema PASS 100% 66 in → 749 out (708 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.6-35b-a3b:low PASS 100% 66 in → 385 out (377 reasoning)
git clean -n
qwen/qwen3.6-35b-a3b:low__json_schema PASS 100% 66 in → 791 out (737 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.6-35b-a3b:medium PASS 100% 66 in → 848 out (820 reasoning)
git clean -n
qwen/qwen3.6-35b-a3b:medium__json_schema PASS 100% 66 in → 1,000 out (909 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.6-35b-a3b:none PASS 100% 68 in → 6 out (0 reasoning)
git clean --dry-run
qwen/qwen3.6-35b-a3b:none__json_schema PASS 100% 68 in → 20 out (0 reasoning)
git clean -n -d --dry-run
JSON Schema Structured Output
(raw) { "command": "git clean -n -d --dry-run" }
qwen/qwen3.6-35b-a3b:xhigh PASS 100% 66 in → 778 out (753 reasoning)
git clean -n
qwen/qwen3.6-35b-a3b:xhigh__json_schema PASS 100% 66 in → 676 out (660 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.6-flash:high PASS 100% 66 in → 837 out (828 reasoning)
git clean -n
qwen/qwen3.6-flash:low PASS 100% 66 in → 965 out (956 reasoning)
git clean -n
qwen/qwen3.6-flash:medium PASS 100% 66 in → 1,427 out (1,417 reasoning)
git clean --dry-run
qwen/qwen3.6-flash:none PASS 100% 68 in → 4 out (0 reasoning)
git clean -n
qwen/qwen3.6-flash:xhigh PASS 100% 66 in → 830 out (821 reasoning)
git clean -n
qwen/qwen3.7-flash:high PASS 100% 66 in → 803 out (794 reasoning)
git clean -n
qwen/qwen3.7-flash:low PASS 100% 66 in → 675 out (666 reasoning)
git clean -n
qwen/qwen3.7-flash:max PASS 100% 66 in → 334 out (325 reasoning)
git clean -n
qwen/qwen3.7-flash:medium PASS 100% 66 in → 883 out (874 reasoning)
git clean -n
qwen/qwen3.7-flash:none PASS 100% 68 in → 5 out (0 reasoning)
git clean --dry-run
qwen/qwen3.7-flash:xhigh PASS 100% 66 in → 982 out (973 reasoning)
git clean -n
qwen/qwen3.7-max:high PASS 100% 66 in → 177 out (168 reasoning)
git clean -n
qwen/qwen3.7-max:low PASS 100% 66 in → 176 out (167 reasoning)
git clean -n
qwen/qwen3.7-max:medium PASS 100% 66 in → 253 out (244 reasoning)
git clean -n
qwen/qwen3.7-max:none PASS 100% 68 in → 4 out (0 reasoning)
git clean -n
qwen/qwen3.7-max:xhigh PASS 100% 66 in → 155 out (146 reasoning)
git clean -n
qwen/qwen3.7-plus:high PASS 100% 66 in → 635 out (626 reasoning)
git clean -n
qwen/qwen3.7-plus:low PASS 100% 66 in → 591 out (582 reasoning)
git clean -n
qwen/qwen3.7-plus:medium PASS 100% 66 in → 457 out (448 reasoning)
git clean -n
qwen/qwen3.7-plus:none PASS 100% 68 in → 4 out (0 reasoning)
git clean -n
qwen/qwen3.7-plus:xhigh PASS 100% 66 in → 382 out (373 reasoning)
git clean -n
qwen/qwen3.8-flash:high PASS 100% 117 in → 182 out (175 reasoning)
git clean -nd
qwen/qwen3.8-flash:high__json_schema PASS 100% 119 in → 153 out (136 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.8-flash:low PASS 100% 117 in → 209 out (202 reasoning)
git clean -nd
qwen/qwen3.8-flash:low__json_schema PASS 100% 119 in → 193 out (181 reasoning)
git clean -nxd
JSON Schema Structured Output
(raw) {"command":"git clean -nxd"}
qwen/qwen3.8-flash:max PASS 100% 117 in → 162 out (155 reasoning)
git clean -n
qwen/qwen3.8-flash:max__json_schema PASS 100% 119 in → 673 out (660 reasoning)
git clean -nxd
JSON Schema Structured Output
(raw) {"command": "git clean -nxd"}
qwen/qwen3.8-flash:medium PASS 100% 117 in → 534 out (527 reasoning)
git clean -n
qwen/qwen3.8-flash:medium__json_schema PASS 100% 119 in → 205 out (186 reasoning)
git clean -n -d
JSON Schema Structured Output
(raw) { "command": "git clean -n -d" }
qwen/qwen3.8-flash:xhigh PASS 100% 117 in → 145 out (138 reasoning)
git clean -n
qwen/qwen3.8-flash:xhigh__json_schema PASS 100% 119 in → 145 out (133 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
qwen/qwen3.8-max:high PASS 100% 104 in → 166 out (157 reasoning)
git clean -n
qwen/qwen3.8-max:high__json_schema PASS 100% 106 in → 189 out (171 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.8-max:low PASS 100% 92 in → 61 out (48 reasoning)
``` git clean -n ```
qwen/qwen3.8-max:low__json_schema PASS 100% 94 in → 80 out (62 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
qwen/qwen3.8-max:max PASS 100% 104 in → 188 out (179 reasoning)
git clean -n
qwen/qwen3.8-max:max__json_schema PASS 100% 106 in → 150 out (135 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
qwen/qwen3.8-max:medium PASS 100% 66 in → 86 out (77 reasoning)
git clean -n
qwen/qwen3.8-max:medium__json_schema PASS 100% 68 in → 76 out (63 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
qwen/qwen3.8-max:xhigh PASS 100% 104 in → 126 out (117 reasoning)
git clean -n
qwen/qwen3.8-max:xhigh__json_schema PASS 100% 106 in → 163 out (145 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
sakana/fugu-max:high PASS 100% 134 in → 53 out (43 reasoning)
git clean -n
sakana/fugu-max:high__json_schema PASS 100% 169 in → 110 out (92 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
sakana/fugu-max:low PASS 100% 134 in → 54 out (44 reasoning)
git clean -n
sakana/fugu-max:low__json_schema PASS 100% 169 in → 64 out (46 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
sakana/fugu-max:max PASS 100% 134 in → 83 out (73 reasoning)
git clean -n
sakana/fugu-max:max__json_schema PASS 100% 169 in → 256 out (238 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
sakana/fugu-max:medium PASS 100% 134 in → 47 out (37 reasoning)
git clean -n
sakana/fugu-max:medium__json_schema PASS 100% 169 in → 109 out (91 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
sakana/fugu-max:xhigh PASS 100% 134 in → 83 out (73 reasoning)
git clean -n
sakana/fugu-max:xhigh__json_schema PASS 100% 169 in → 214 out (196 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
stealth/ox-alpha:high PASS 100% 139 in → 6 out (0 reasoning)
git clean -n
stealth/ox-alpha:low PASS 100% 139 in → 6 out (0 reasoning)
git clean -n
stealth/ox-alpha:max PASS 100% 139 in → 58 out (0 reasoning)
```bash git clean -n ```
tencent/hy3:high PASS 100% 64 in → 341 out (335 reasoning)
git clean -n
tencent/hy3:high__json_schema PASS 100% 64 in → 658 out (667 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
tencent/hy3:low PASS 100% 64 in → 378 out (372 reasoning)
git clean -n
tencent/hy3:low__json_schema PASS 100% 64 in → 221 out (213 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
tencent/hy3:medium PASS 100% 64 in → 203 out (197 reasoning)
git clean -n
tencent/hy3:medium__json_schema PASS 100% 64 in → 209 out (198 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
tencent/hy3:none PASS 100% 67 in → 5 out (0 reasoning)
git clean -nd
tencent/hy3:xhigh PASS 100% 64 in → 346 out (340 reasoning)
git clean -n
tencent/hy3:xhigh__json_schema PASS 100% 64 in → 490 out (456 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
thinkingmachines/inkling-small:high PASS 100% 66 in → 152 out (142 reasoning)
git clean -n
thinkingmachines/inkling-small:high__json_schema PASS 100% 66 in → 96 out (79 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
thinkingmachines/inkling-small:low PASS 100% 66 in → 59 out (46 reasoning)
git clean -n
thinkingmachines/inkling-small:low__json_schema PASS 100% 66 in → 54 out (40 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
thinkingmachines/inkling-small:max PASS 100% 66 in → 110 out (100 reasoning)
git clean -n
thinkingmachines/inkling-small:max__json_schema PASS 100% 66 in → 207 out (193 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
thinkingmachines/inkling-small:medium PASS 100% 66 in → 29 out (19 reasoning)
git clean -n
thinkingmachines/inkling-small:medium__json_schema PASS 100% 66 in → 104 out (90 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
thinkingmachines/inkling-small:none PASS 100% 64 in → 7 out (0 reasoning)
git clean -n
thinkingmachines/inkling-small:xhigh PASS 100% 66 in → 143 out (133 reasoning)
git clean -n
thinkingmachines/inkling-small:xhigh__json_schema PASS 100% 66 in → 110 out (96 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
thinkingmachines/inkling:high PASS 100% 66 in → 76 out (65 reasoning)
git clean -n
thinkingmachines/inkling:high__json_schema PASS 100% 66 in → 109 out (96 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
thinkingmachines/inkling:low PASS 100% 66 in → 27 out (17 reasoning)
git clean -n
thinkingmachines/inkling:max PASS 100% 66 in → 216 out (206 reasoning)
git clean -n
thinkingmachines/inkling:medium PASS 100% 66 in → 73 out (63 reasoning)
git clean -n
thinkingmachines/inkling:medium__json_schema PASS 100% 66 in → 98 out (82 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
thinkingmachines/inkling:none PASS 100% 64 in → 8 out (0 reasoning)
git clean -n
thinkingmachines/inkling:none__json_schema PASS 100% 64 in → 12 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n"}
thinkingmachines/inkling:xhigh PASS 100% 66 in → 130 out (120 reasoning)
git clean -n
thinkingmachines/inkling:xhigh__json_schema PASS 100% 66 in → 61 out (45 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
x-ai/grok-4.3:high PASS 100% 236 in → 197 out (193 reasoning)
git clean -n
x-ai/grok-4.3:high__json_schema PASS 100% 292 in → 385 out (377 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.3:low PASS 100% 242 in → 394 out (290 reasoning)
git clean -n The command `git clean -n` (or equivalently `--dry-run`) performs a dry-run of `git clean`, listing only the untracked files that would be deleted while leaving the working tree untouched. Because the supplied status shows exactly the two untracked files `another.txt` and `untracked.txt` and no ignored files, the output of the dry-run will be those two paths. No other flags (e.g., `-d` or `-x`) are required for this scenario.
x-ai/grok-4.3:low__json_schema PASS 100% 298 in → 308 out (299 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
x-ai/grok-4.3:max PASS 100% 236 in → 210 out (206 reasoning)
git clean -n
x-ai/grok-4.3:max__json_schema PASS 100% 292 in → 349 out (337 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
x-ai/grok-4.3:medium PASS 100% 242 in → 422 out (418 reasoning)
git clean -n
x-ai/grok-4.3:medium__json_schema PASS 100% 298 in → 447 out (435 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
x-ai/grok-4.3:none PASS 100% 234 in → 6 out (0 reasoning)
git clean -n -d
x-ai/grok-4.3:none__json_schema PASS 100% 290 in → 8 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.3:xhigh PASS 100% 236 in → 167 out (160 reasoning)
``` git clean -n ```
x-ai/grok-4.3:xhigh__json_schema PASS 100% 292 in → 439 out (431 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.5:high PASS 100% 257 in → 315 out (311 reasoning)
git clean -n
x-ai/grok-4.5:high__json_schema PASS 100% 323 in → 302 out (293 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.5:low PASS 100% 257 in → 134 out (126 reasoning)
``` git clean -n ```
x-ai/grok-4.5:low__json_schema PASS 100% 323 in → 253 out (244 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.5:max PASS 100% 257 in → 313 out (309 reasoning)
git clean -n
x-ai/grok-4.5:max__json_schema PASS 100% 323 in → 471 out (462 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
x-ai/grok-4.5:medium PASS 100% 257 in → 322 out (318 reasoning)
git clean -n
x-ai/grok-4.5:medium__json_schema PASS 100% 323 in → 317 out (308 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.5:xhigh PASS 100% 257 in → 173 out (169 reasoning)
git clean -n
x-ai/grok-4.5:xhigh__json_schema PASS 100% 323 in → 419 out (410 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
x-ai/grok-4.7:high PASS 100% 1,293 in → 278 out (274 reasoning)
git clean -n
x-ai/grok-4.7:high__json_schema PASS 100% 1,359 in → 264 out (255 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.7:low PASS 100% 1,293 in → 215 out (211 reasoning)
git clean -n
x-ai/grok-4.7:low__json_schema PASS 100% 1,359 in → 83 out (74 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.7:max PASS 100% 1,294 in → 335 out (331 reasoning)
git clean -n
x-ai/grok-4.7:max__json_schema PASS 100% 1,360 in → 268 out (259 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.7:medium PASS 100% 1,293 in → 171 out (167 reasoning)
git clean -n
x-ai/grok-4.7:medium__json_schema PASS 100% 1,359 in → 248 out (239 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
x-ai/grok-4.7:xhigh PASS 100% 1,294 in → 143 out (139 reasoning)
git clean -n
x-ai/grok-4.7:xhigh__json_schema PASS 100% 1,360 in → 546 out (537 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
xiaomi/mimo-v2.6-flash:high PASS 100% 59 in → 29 out (18 reasoning)
``` git clean -n ```
xiaomi/mimo-v2.6-flash:low PASS 100% 59 in → 25 out (18 reasoning)
git clean -n
xiaomi/mimo-v2.6-flash:low__json_schema PASS 100% 59 in → 37 out (22 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
xiaomi/mimo-v2.6-flash:max PASS 100% 59 in → 11 out (5 reasoning)
git clean -n
xiaomi/mimo-v2.6-flash:max__json_schema PASS 100% 59 in → 34 out (19 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
xiaomi/mimo-v2.6-flash:medium PASS 100% 59 in → 27 out (16 reasoning)
``` git clean -n ```
xiaomi/mimo-v2.6-flash:medium__json_schema PASS 100% 59 in → 19 out (5 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
xiaomi/mimo-v2.6-flash:none PASS 100% 61 in → 5 out (0 reasoning)
git clean -n
xiaomi/mimo-v2.6-flash:xhigh PASS 100% 59 in → 30 out (19 reasoning)
``` git clean -n ```
xiaomi/mimo-v2.6-flash:xhigh__json_schema PASS 100% 59 in → 32 out (16 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
xiaomi/mimo-v2.6-pro:high PASS 100% 59 in → 15 out (9 reasoning)
git clean -n
xiaomi/mimo-v2.6-pro:high__json_schema PASS 100% 59 in → 24 out (10 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
xiaomi/mimo-v2.6-pro:low PASS 100% 59 in → 11 out (5 reasoning)
git clean -n
xiaomi/mimo-v2.6-pro:low__json_schema PASS 100% 59 in → 20 out (6 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
xiaomi/mimo-v2.6-pro:max PASS 100% 59 in → 14 out (8 reasoning)
git clean -n
xiaomi/mimo-v2.6-pro:max__json_schema PASS 100% 59 in → 19 out (5 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
xiaomi/mimo-v2.6-pro:medium PASS 100% 59 in → 11 out (5 reasoning)
git clean -nd
xiaomi/mimo-v2.6-pro:medium__json_schema PASS 100% 59 in → 41 out (27 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
xiaomi/mimo-v2.6-pro:xhigh PASS 100% 59 in → 17 out (11 reasoning)
git clean -n
xiaomi/mimo-v2.6-pro:xhigh__json_schema PASS 100% 59 in → 49 out (35 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
z-ai/glm-4.7-flash:high PASS 100% 57 in → 360 out (358 reasoning)
git clean -n
z-ai/glm-4.7-flash:low PASS 100% 57 in → 414 out (407 reasoning)
git clean -n
z-ai/glm-4.7-flash:medium PASS 100% 57 in → 408 out (395 reasoning)
git clean -n
z-ai/glm-4.7-flash:medium__json_schema PASS 100% 57 in → 291 out (271 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
z-ai/glm-4.7-flash:none PASS 100% 57 in → 5 out (0 reasoning)
git clean -n
z-ai/glm-4.7-flash:none__json_schema PASS 100% 57 in → 10 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
z-ai/glm-4.7-flash:xhigh PASS 100% 57 in → 341 out (337 reasoning)
git clean -n
z-ai/glm-4.7-flash:xhigh__json_schema PASS 100% 57 in → 260 out (232 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
z-ai/glm-5.2:high PASS 100% 64 in → 209 out (204 reasoning)
git clean -n
z-ai/glm-5.2:high__json_schema PASS 100% 57 in → 447 out (394 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
z-ai/glm-5.2:low PASS 100% 64 in → 261 out (247 reasoning)
git clean -n
z-ai/glm-5.2:low__json_schema PASS 100% 66 in → 219 out (206 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
z-ai/glm-5.2:medium PASS 100% 64 in → 281 out (276 reasoning)
git clean -n
z-ai/glm-5.2:medium__json_schema PASS 100% 239 in → 10 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command":"git clean -n"}
z-ai/glm-5.2:none PASS 100% 58 in → 4 out (0 reasoning)
git clean -n
z-ai/glm-5.2:none__json_schema PASS 100% 60 in → 12 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
z-ai/glm-5.2:xhigh PASS 100% 64 in → 386 out (0 reasoning)
git clean -n
z-ai/glm-5.2:xhigh__json_schema PASS 100% 64 in → 281 out (253 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
z-ai/glm-5.3-flash:high PASS 100% 64 in → 25 out (0 reasoning)
``` git clean -n ``` *(also: `git clean --dry-run` — same effect)*
z-ai/glm-5.3-flash:high__json_schema PASS 100% 64 in → 20 out (7 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
z-ai/glm-5.3-flash:low PASS 100% 64 in → 6 out (0 reasoning)
git clean -n
z-ai/glm-5.3-flash:low__json_schema PASS 100% 64 in → 13 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
z-ai/glm-5.3-flash:max PASS 100% 64 in → 155 out (149 reasoning)
git clean -n
z-ai/glm-5.3-flash:max__json_schema PASS 100% 64 in → 98 out (86 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
z-ai/glm-5.3-flash:medium PASS 100% 64 in → 8 out (0 reasoning)
`git clean -n`
z-ai/glm-5.3-flash:medium__json_schema PASS 100% 64 in → 11 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) {"command": "git clean -n"}
z-ai/glm-5.3-flash:xhigh PASS 100% 64 in → 10 out (0 reasoning)
``` git clean -n ```
z-ai/glm-5.3-flash:xhigh__json_schema PASS 100% 64 in → 13 out (0 reasoning)
git clean -n
JSON Schema Structured Output
(raw) { "command": "git clean -n" }
Invalid JSON. Output:
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output:
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output:
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output:
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output: git clean -fd
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output: {"cmd":["bash","-lc","git clean -n"]
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting ',' delimiter: line 1 column 37 (char 36)
Failure: Failed to parse structured JSON response: Expecting ',' delimiter: line 1 column 37 (char 36)
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-2e820d9f-562a-962b-8654-33ec60185d74","request_id":"2e820d9f-562a-962b-8654-33ec60185d74"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-9b8bdd4c-6e64-9068-bcc4-dbc266ead9c0","request_id":"9b8bdd4c-6e64-9068-bcc4-dbc266ead9c0"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-93bee89c-ee03-9f2e-9674-b809e292af54","request_id":"93bee89c-ee03-9f2e-9674-b809e292af54"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-f0ed976d-99b2-9d4b-b426-022a061c94dc","request_id":"f0ed976d-99b2-9d4b-b426-022a061c94dc"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-a1e156f1-a746-9e9f-9be6-cdd14d8f8e53","request_id":"a1e156f1-a746-9e9f-9be6-cdd14d8f8e53"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': 'data: {"error":{"code":"invalid_parameter_error","param":null,"message":"\'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error"},"id":"chatcmpl-65c49cb1-df59-9e96-947a-a47fc22c6a62"}\n\n', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': 'data: {"error":{"code":"invalid_parameter_error","param":null,"message":"\'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error"},"id":"chatcmpl-adedaacd-0d92-968f-b1d5-fe1596cacdfe"}\n\n', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': 'data: {"error":{"code":"invalid_parameter_error","param":null,"message":"\'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error"},"id":"chatcmpl-9a688032-59a0-9cf0-a402-f9041a8ed709"}\n\n', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': 'data: {"error":{"code":"invalid_parameter_error","param":null,"message":"\'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error"},"id":"chatcmpl-0113f4af-e486-99ed-b181-86b851cf9aac"}\n\n', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': 'data: {"error":{"code":"invalid_parameter_error","param":null,"message":"\'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error"},"id":"chatcmpl-d13e0e5e-85d0-97c2-ac31-fcbca5325993"}\n\n', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': 'data: {"error":{"code":"invalid_parameter_error","param":null,"message":"\'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error"},"id":"chatcmpl-ccb95563-d7bf-9010-9e54-611573ab67c1"}\n\n', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-7db79f12-63f2-97cb-8b2a-0ed509241394","request_id":"7db79f12-63f2-97cb-8b2a-0ed509241394"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-d6b7667d-da49-9d37-89a8-f1848e369000","request_id":"d6b7667d-da49-9d37-89a8-f1848e369000"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-9bf226ff-9a3c-9e83-b342-3f31228fdeb2","request_id":"9bf226ff-9a3c-9e83-b342-3f31228fdeb2"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-e105237c-890c-981f-8e6c-75c6eab8f121","request_id":"e105237c-890c-981f-8e6c-75c6eab8f121"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-97f4ec9a-3627-940e-bc19-7295fa704125","request_id":"97f4ec9a-3627-940e-bc19-7295fa704125"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-54ee2921-d859-9d33-b71d-4db40b84d46e","request_id":"54ee2921-d859-9d33-b71d-4db40b84d46e"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-d6a25371-1947-92a3-9c11-8ea01bfc2fa0","request_id":"d6a25371-1947-92a3-9c11-8ea01bfc2fa0"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-c0b40ffb-68fa-9025-b772-6f510661a8d5","request_id":"c0b40ffb-68fa-9025-b772-6f510661a8d5"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-4d9e083a-4bb3-94bc-b8db-35f756217024","request_id":"4d9e083a-4bb3-94bc-b8db-35f756217024"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
(empty output)
JSON Schema Structured Output
Failure: Error code: 400 - {'error': {'message': 'Provider returned error', 'code': 400, 'metadata': {'raw': '{"error":{"message":"<400> InternalError.Algo.InvalidParameter: \'messages\' must contain the word \'json\' in some form, to use \'response_format\' of type \'json_object\'.","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-c684ce64-c69f-9093-a518-dd974a982254","request_id":"c684ce64-c69f-9093-a518-dd974a982254"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
Invalid JSON. Output: git clean -n
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output: git clean -n
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output: git clean -n
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
(empty output)
JSON Schema Structured Output
Failure: Model 'tencent/hy3:none' violated the reasoning_level=none invariant: provider reported 413 reasoning token(s)
Invalid JSON. Output: git clean -f -n{"command": "git clean -f -n"}
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output: git clean -n
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output: git clean -n
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output: { "command": "git clean -n" }{ "
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Extra data: line 1 column 30 (char 29)
Failure: Failed to parse structured JSON response: Extra data: line 1 column 30 (char 29)
Invalid JSON. Output: { "command": "git clean -n" }{ "
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Extra data: line 1 column 30 (char 29)
Failure: Failed to parse structured JSON response: Extra data: line 1 column 30 (char 29)
Invalid JSON. Output:
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Invalid JSON. Output:
JSON Schema Structured Output
Structured Output Error
Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)
Failure: Failed to parse structured JSON response: Expecting value: line 1 column 1 (char 0)