Show file content at a specific revision
Tests ability to show file content at a specific revision. Evaluates historical-file-content retrieval.

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 'version one' > doc.txt
  5. 05 git add doc.txt
  6. 06 git commit -m 'First version'
  7. 07 FIRST=$(git rev-parse HEAD)
  8. 08 echo 'version two' > doc.txt
  9. 09 git add doc.txt
  10. 10 git commit -m 'Second version'
Prompt
Using git show HEAD~1:doc.txt, what was the content of doc.txt in the first commit? Output ONLY the file content, nothing else.
Expected
version one

Scoped model quality, cost, API time, and token usage for git_show/f005.

Loading...
Loading raw attempt evidence…
anthropic/claude-fable-5:high PASS 100% 254 in → 16 out (11 reasoning)
version one
anthropic/claude-fable-5:high__json_schema PASS 100% 459 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
anthropic/claude-fable-5:low PASS 100% 253 in → 4 out (0 reasoning)
version one
anthropic/claude-fable-5:low__json_schema PASS 100% 463 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
anthropic/claude-fable-5:medium PASS 100% 252 in → 13 out (8 reasoning)
version one
anthropic/claude-fable-5:medium__json_schema PASS 100% 454 in → 13 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
anthropic/claude-fable-5:xhigh PASS 100% 252 in → 15 out (13 reasoning)
version one
anthropic/claude-fable-5:xhigh__json_schema PASS 100% 456 in → 24 out (13 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
anthropic/claude-haiku-4.5:high PASS 100% 235 in → 223 out (210 reasoning)
version one
anthropic/claude-haiku-4.5:high__json_schema PASS 100% 402 in → 417 out (390 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
anthropic/claude-haiku-4.5:low PASS 100% 232 in → 388 out (368 reasoning)
version one
anthropic/claude-haiku-4.5:low__json_schema PASS 100% 392 in → 210 out (182 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
anthropic/claude-haiku-4.5:medium PASS 100% 239 in → 275 out (244 reasoning)
version one
anthropic/claude-haiku-4.5:medium__json_schema PASS 100% 394 in → 189 out (170 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
anthropic/claude-haiku-4.5:none PASS 100% 208 in → 5 out (0 reasoning)
version one
anthropic/claude-haiku-4.5:none__json_schema PASS 100% 368 in → 10 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
anthropic/claude-haiku-4.5:xhigh PASS 100% 236 in → 260 out (220 reasoning)
version one
anthropic/claude-haiku-4.5:xhigh__json_schema PASS 100% 399 in → 200 out (183 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
anthropic/claude-opus-4.6:high PASS 100% 204 in → 5 out (0 reasoning)
version one
anthropic/claude-opus-4.6:high__json_schema PASS 100% 363 in → 12 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.6:low PASS 100% 203 in → 5 out (0 reasoning)
version one
anthropic/claude-opus-4.6:low__json_schema PASS 100% 368 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.6:max PASS 100% 201 in → 100 out (77 reasoning)
version one
anthropic/claude-opus-4.6:max__json_schema PASS 100% 367 in → 93 out (68 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.6:medium PASS 100% 202 in → 5 out (0 reasoning)
version one
anthropic/claude-opus-4.6:medium__json_schema PASS 100% 365 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.6:none PASS 100% 205 in → 5 out (0 reasoning)
version one
anthropic/claude-opus-4.6:none__json_schema PASS 100% 367 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.6:xhigh PASS 100% 204 in → 5 out (0 reasoning)
version one
anthropic/claude-opus-4.6:xhigh__json_schema PASS 100% 367 in → 12 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.7:high PASS 100% 260 in → 6 out (0 reasoning)
version one
anthropic/claude-opus-4.7:high__json_schema PASS 100% 464 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.7:low PASS 100% 260 in → 6 out (0 reasoning)
version one
anthropic/claude-opus-4.7:low__json_schema PASS 100% 462 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.7:max PASS 100% 257 in → 122 out (44 reasoning)
version one
anthropic/claude-opus-4.7:max__json_schema PASS 100% 458 in → 168 out (68 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.7:medium PASS 100% 255 in → 6 out (0 reasoning)
version one
anthropic/claude-opus-4.7:medium__json_schema PASS 100% 462 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.7:none PASS 100% 260 in → 6 out (0 reasoning)
version one
anthropic/claude-opus-4.7:none__json_schema PASS 100% 464 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.7:xhigh PASS 100% 262 in → 6 out (0 reasoning)
version one
anthropic/claude-opus-4.7:xhigh__json_schema PASS 100% 466 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.8:high PASS 100% 255 in → 66 out (36 reasoning)
version one
anthropic/claude-opus-4.8:high__json_schema PASS 100% 460 in → 66 out (35 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.8:low PASS 100% 250 in → 25 out (14 reasoning)
version one
anthropic/claude-opus-4.8:low__json_schema PASS 100% 453 in → 13 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.8:max PASS 100% 254 in → 325 out (79 reasoning)
version one
anthropic/claude-opus-4.8:max__json_schema PASS 100% 460 in → 154 out (64 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
anthropic/claude-opus-4.8:medium PASS 100% 256 in → 47 out (28 reasoning)
version one
anthropic/claude-opus-4.8:medium__json_schema PASS 100% 460 in → 34 out (14 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.8:none PASS 100% 254 in → 4 out (0 reasoning)
version one
anthropic/claude-opus-4.8:none__json_schema PASS 100% 454 in → 12 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-opus-4.8:xhigh PASS 100% 256 in → 122 out (51 reasoning)
version one
anthropic/claude-opus-4.8:xhigh__json_schema PASS 100% 460 in → 77 out (38 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-sonnet-4.6:high PASS 100% 207 in → 103 out (82 reasoning)
version one
anthropic/claude-sonnet-4.6:high__json_schema PASS 100% 366 in → 62 out (36 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
anthropic/claude-sonnet-4.6:low PASS 100% 207 in → 5 out (0 reasoning)
version one
anthropic/claude-sonnet-4.6:low__json_schema PASS 100% 368 in → 12 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
anthropic/claude-sonnet-4.6:max PASS 100% 202 in → 152 out (132 reasoning)
version one
anthropic/claude-sonnet-4.6:max__json_schema PASS 100% 367 in → 121 out (92 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
anthropic/claude-sonnet-4.6:medium PASS 100% 203 in → 5 out (0 reasoning)
version one
anthropic/claude-sonnet-4.6:medium__json_schema PASS 100% 365 in → 12 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
anthropic/claude-sonnet-4.6:none__json_schema PASS 100% 367 in → 12 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
anthropic/claude-sonnet-4.6:xhigh PASS 100% 200 in → 73 out (51 reasoning)
version one
anthropic/claude-sonnet-4.6:xhigh__json_schema PASS 100% 368 in → 90 out (65 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
anthropic/claude-sonnet-5:high PASS 100% 254 in → 4 out (0 reasoning)
version one
anthropic/claude-sonnet-5:high__json_schema PASS 100% 457 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
anthropic/claude-sonnet-5:low PASS 100% 249 in → 4 out (0 reasoning)
version one
anthropic/claude-sonnet-5:low__json_schema PASS 100% 462 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
anthropic/claude-sonnet-5:medium PASS 100% 256 in → 4 out (0 reasoning)
version one
anthropic/claude-sonnet-5:medium__json_schema PASS 100% 457 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
anthropic/claude-sonnet-5:none PASS 100% 253 in → 4 out (0 reasoning)
version one
anthropic/claude-sonnet-5:none__json_schema PASS 100% 457 in → 10 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
anthropic/claude-sonnet-5:xhigh PASS 100% 250 in → 56 out (39 reasoning)
version one
anthropic/claude-sonnet-5:xhigh__json_schema PASS 100% 466 in → 28 out (12 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
arcee-ai/trinity-large-thinking:high PASS 100% 171 in → 503 out (498 reasoning)
version one
arcee-ai/trinity-large-thinking:low PASS 100% 181 in → 747 out (744 reasoning)
version one
arcee-ai/trinity-large-thinking:medium PASS 100% 170 in → 509 out (504 reasoning)
version one
arcee-ai/trinity-large-thinking:xhigh PASS 100% 175 in → 379 out (374 reasoning)
version one
arcee-ai/trinity-mini:low PASS 100% 177 in → 364 out (381 reasoning)
version one
arcee-ai/trinity-mini:xhigh PASS 100% 171 in → 1,140 out (1,326 reasoning)
version one
deepseek/deepseek-v4-flash-0731:high PASS 100% 248 in → 111 out (116 reasoning)
version one
deepseek/deepseek-v4-flash-0731:high__json_schema PASS 100% 175 in → 116 out (105 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
deepseek/deepseek-v4-flash-0731:low PASS 100% 172 in → 85 out (82 reasoning)
version one
deepseek/deepseek-v4-flash-0731:low__json_schema PASS 100% 171 in → 116 out (113 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
deepseek/deepseek-v4-flash-0731:max PASS 100% 248 in → 177 out (194 reasoning)
version one
deepseek/deepseek-v4-flash-0731:medium PASS 100% 170 in → 82 out (85 reasoning)
version one
deepseek/deepseek-v4-flash-0731:medium__json_schema PASS 100% 177 in → 111 out (99 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
deepseek/deepseek-v4-flash-0731:none PASS 100% 174 in → 2 out (0 reasoning)
version one
deepseek/deepseek-v4-flash-0731:none__json_schema PASS 100% 171 in → 7 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
deepseek/deepseek-v4-flash-0731:xhigh PASS 100% 173 in → 110 out (112 reasoning)
version one
deepseek/deepseek-v4-flash-0731:xhigh__json_schema PASS 100% 176 in → 330 out (339 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one\n" }
deepseek/deepseek-v4-flash:high PASS 100% 172 in → 155 out (159 reasoning)
version one
deepseek/deepseek-v4-flash:high__json_schema PASS 100% 174 in → 115 out (103 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one\n" }
deepseek/deepseek-v4-flash:low PASS 100% 171 in → 182 out (178 reasoning)
version one
deepseek/deepseek-v4-flash:low__json_schema PASS 100% 173 in → 121 out (110 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
deepseek/deepseek-v4-flash:medium PASS 100% 176 in → 137 out (134 reasoning)
version one
deepseek/deepseek-v4-flash:medium__json_schema PASS 100% 175 in → 118 out (107 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
deepseek/deepseek-v4-flash:none PASS 100% 170 in → 2 out (0 reasoning)
version one
deepseek/deepseek-v4-flash:none__json_schema PASS 100% 177 in → 10 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
deepseek/deepseek-v4-flash:xhigh PASS 100% 255 in → 136 out (132 reasoning)
version one
deepseek/deepseek-v4-flash:xhigh__json_schema PASS 100% 251 in → 169 out (160 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
deepseek/deepseek-v4-pro:high PASS 100% 169 in → 192 out (188 reasoning)
version one
deepseek/deepseek-v4-pro:low PASS 100% 172 in → 3 out (0 reasoning)
version one
deepseek/deepseek-v4-pro:low__json_schema PASS 100% 176 in → 252 out (240 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
deepseek/deepseek-v4-pro:medium PASS 100% 174 in → 134 out (130 reasoning)
version one
deepseek/deepseek-v4-pro:none PASS 100% 173 in → 3 out (0 reasoning)
version one
deepseek/deepseek-v4-pro:none__json_schema PASS 100% 174 in → 8 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
deepseek/deepseek-v4-pro:xhigh PASS 100% 171 in → 87 out (83 reasoning)
version one
deepseek/deepseek-v4-pro:xhigh__json_schema PASS 100% 251 in → 1,090 out (1,081 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
google/gemini-3-flash-preview:high PASS 100% 208 in → 265 out (263 reasoning)
version one
google/gemini-3-flash-preview:high__json_schema PASS 100% 265 in → 155 out (149 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3-flash-preview:low PASS 100% 208 in → 247 out (245 reasoning)
version one
google/gemini-3-flash-preview:low__json_schema PASS 100% 262 in → 267 out (261 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3-flash-preview:medium PASS 100% 206 in → 288 out (286 reasoning)
version one
google/gemini-3-flash-preview:medium__json_schema PASS 100% 262 in → 315 out (304 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemini-3-flash-preview:xhigh PASS 100% 203 in → 320 out (318 reasoning)
version one
google/gemini-3-flash-preview:xhigh__json_schema PASS 100% 264 in → 328 out (321 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
google/gemini-3.1-flash-lite-preview:high PASS 100% 207 in → 509 out (507 reasoning)
version one
google/gemini-3.1-flash-lite-preview:high__json_schema PASS 100% 209 in → 618 out (612 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.1-flash-lite-preview:low PASS 100% 210 in → 135 out (133 reasoning)
version one
google/gemini-3.1-flash-lite-preview:low__json_schema PASS 100% 262 in → 142 out (135 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
google/gemini-3.1-flash-lite-preview:medium PASS 100% 209 in → 274 out (272 reasoning)
version one
google/gemini-3.1-flash-lite-preview:medium__json_schema PASS 100% 211 in → 140 out (128 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemini-3.1-flash-lite-preview:xhigh PASS 100% 206 in → 418 out (416 reasoning)
version one
google/gemini-3.1-flash-lite-preview:xhigh__json_schema PASS 100% 212 in → 264 out (257 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
google/gemini-3.1-pro-preview:high PASS 100% 209 in → 327 out (325 reasoning)
version one
google/gemini-3.1-pro-preview:high__json_schema PASS 100% 263 in → 883 out (875 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
google/gemini-3.1-pro-preview:low PASS 100% 212 in → 268 out (266 reasoning)
version one
google/gemini-3.1-pro-preview:low__json_schema PASS 100% 262 in → 239 out (233 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.1-pro-preview:medium PASS 100% 210 in → 314 out (312 reasoning)
version one
google/gemini-3.1-pro-preview:medium__json_schema PASS 100% 257 in → 291 out (285 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.1-pro-preview:xhigh PASS 100% 208 in → 390 out (388 reasoning)
version one
google/gemini-3.1-pro-preview:xhigh__json_schema PASS 100% 262 in → 134 out (127 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
google/gemini-3.5-flash-lite:high PASS 100% 208 in → 436 out (434 reasoning)
version one
google/gemini-3.5-flash-lite:high__json_schema PASS 100% 264 in → 856 out (844 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemini-3.5-flash-lite:low PASS 100% 210 in → 206 out (204 reasoning)
version one
google/gemini-3.5-flash-lite:low__json_schema PASS 100% 263 in → 14 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one\n" }
google/gemini-3.5-flash-lite:max PASS 100% 208 in → 494 out (492 reasoning)
version one
google/gemini-3.5-flash-lite:max__json_schema PASS 100% 264 in → 594 out (588 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.5-flash-lite:medium PASS 100% 211 in → 354 out (352 reasoning)
version one
google/gemini-3.5-flash-lite:medium__json_schema PASS 100% 207 in → 308 out (301 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
google/gemini-3.5-flash-lite:xhigh PASS 100% 208 in → 393 out (391 reasoning)
version one
google/gemini-3.5-flash-lite:xhigh__json_schema PASS 100% 263 in → 492 out (486 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.5-flash:high PASS 100% 205 in → 329 out (327 reasoning)
version one
google/gemini-3.5-flash:high__json_schema PASS 100% 263 in → 372 out (366 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.5-flash:low PASS 100% 208 in → 106 out (104 reasoning)
version one
google/gemini-3.5-flash:low__json_schema PASS 100% 262 in → 161 out (152 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
google/gemini-3.5-flash:medium PASS 100% 209 in → 319 out (317 reasoning)
version one
google/gemini-3.5-flash:medium__json_schema PASS 100% 265 in → 475 out (466 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one\n"}
google/gemini-3.5-flash:xhigh PASS 100% 208 in → 410 out (408 reasoning)
version one
google/gemini-3.5-flash:xhigh__json_schema PASS 100% 264 in → 399 out (392 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
google/gemini-3.6-flash:high PASS 100% 210 in → 282 out (280 reasoning)
version one
google/gemini-3.6-flash:high__json_schema PASS 100% 267 in → 465 out (459 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.6-flash:low PASS 100% 209 in → 82 out (80 reasoning)
version one
google/gemini-3.6-flash:low__json_schema PASS 100% 210 in → 12 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemini-3.6-flash:max PASS 100% 209 in → 399 out (397 reasoning)
version one
google/gemini-3.6-flash:max__json_schema PASS 100% 261 in → 234 out (228 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.6-flash:medium PASS 100% 206 in → 344 out (342 reasoning)
version one
google/gemini-3.6-flash:medium__json_schema PASS 100% 211 in → 335 out (329 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemini-3.6-flash:xhigh PASS 100% 207 in → 236 out (234 reasoning)
version one
google/gemini-3.6-flash:xhigh__json_schema PASS 100% 268 in → 242 out (236 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
google/gemma-4-26b-a4b-it:high PASS 100% 222 in → 498 out (410 reasoning)
version one
google/gemma-4-26b-a4b-it:high__json_schema PASS 100% 224 in → 283 out (224 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemma-4-26b-a4b-it:low PASS 100% 222 in → 712 out (709 reasoning)
version one
google/gemma-4-26b-a4b-it:low__json_schema PASS 100% 227 in → 332 out (257 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemma-4-26b-a4b-it:medium PASS 100% 227 in → 384 out (381 reasoning)
version one
google/gemma-4-26b-a4b-it:medium__json_schema PASS 100% 222 in → 476 out (387 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemma-4-26b-a4b-it:none PASS 100% 219 in → 3 out (0 reasoning)
version one
google/gemma-4-26b-a4b-it:none__json_schema PASS 100% 221 in → 13 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemma-4-26b-a4b-it:xhigh PASS 100% 227 in → 456 out (453 reasoning)
version one
google/gemma-4-26b-a4b-it:xhigh__json_schema PASS 100% 223 in → 378 out (275 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemma-4-31b-it:high PASS 100% 223 in → 270 out (229 reasoning)
version one
google/gemma-4-31b-it:high__json_schema PASS 100% 226 in → 289 out (225 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemma-4-31b-it:low PASS 100% 226 in → 330 out (1 reasoning)
version one
google/gemma-4-31b-it:low__json_schema PASS 100% 226 in → 277 out (225 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
google/gemma-4-31b-it:medium PASS 100% 223 in → 316 out (1 reasoning)
version one
google/gemma-4-31b-it:medium__json_schema PASS 100% 229 in → 322 out (260 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemma-4-31b-it:none PASS 100% 221 in → 3 out (0 reasoning)
version one
google/gemma-4-31b-it:none__json_schema PASS 100% 225 in → 13 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
google/gemma-4-31b-it:xhigh PASS 100% 226 in → 248 out (210 reasoning)
version one
google/gemma-4-31b-it:xhigh__json_schema PASS 100% 221 in → 262 out (210 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
ibm-granite/granite-4.1-8b PASS 100% 194 in → 3 out
version one
inclusionai/ling-2.6-flash PASS 100% 217 in → 3 out
version one
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
liquid/lfm-2-24b-a2b PASS 100% 184 in → 3 out
version one
minimax/minimax-m2.5:high__json_schema PASS 100% 199 in → 239 out (227 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
minimax/minimax-m2.5:low PASS 100% 202 in → 441 out (480 reasoning)
version one
minimax/minimax-m2.5:low__json_schema PASS 100% 224 in → 126 out (124 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
minimax/minimax-m2.5:medium__json_schema PASS 100% 223 in → 137 out (127 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
minimax/minimax-m2.5:xhigh PASS 100% 202 in → 193 out (218 reasoning)
version one
minimax/minimax-m2.5:xhigh__json_schema PASS 100% 201 in → 398 out (387 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
minimax/minimax-m2.7:high PASS 100% 201 in → 298 out (324 reasoning)
version one
minimax/minimax-m2.7:high__json_schema PASS 100% 336 in → 813 out (805 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
minimax/minimax-m2.7:low PASS 100% 203 in → 195 out (190 reasoning)
version one
minimax/minimax-m2.7:low__json_schema PASS 100% 333 in → 872 out (864 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
minimax/minimax-m2.7:medium PASS 100% 187 in → 93 out (94 reasoning)
version one
minimax/minimax-m2.7:medium__json_schema PASS 100% 333 in → 267 out (259 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
minimax/minimax-m2.7:xhigh PASS 100% 201 in → 334 out (330 reasoning)
version one
minimax/minimax-m2.7:xhigh__json_schema PASS 100% 333 in → 380 out (372 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
minimax/minimax-m3:high PASS 100% 334 in → 84 out (85 reasoning)
version one
minimax/minimax-m3:high__json_schema PASS 100% 288 in → 7 out (6 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
minimax/minimax-m3:low PASS 100% 337 in → 87 out (87 reasoning)
version one
minimax/minimax-m3:low__json_schema PASS 100% 331 in → 8 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
minimax/minimax-m3:medium PASS 100% 339 in → 74 out (73 reasoning)
version one
minimax/minimax-m3:medium__json_schema PASS 100% 292 in → 10 out (9 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
minimax/minimax-m3:xhigh PASS 100% 336 in → 70 out (72 reasoning)
version one
minimax/minimax-m3:xhigh__json_schema PASS 100% 323 in → 8 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
mistralai/devstral-2512 PASS 100% 196 in → 3 out
version one
mistralai/devstral-2512__json_schema PASS 100% 200 in → 8 out
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
mistralai/mistral-medium-3-5:high PASS 100% 205 in → 334 out (355 reasoning)
version one
mistralai/mistral-medium-3-5:high__json_schema PASS 100% 194 in → 297 out (304 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
mistralai/mistral-medium-3-5:low PASS 100% 208 in → 277 out (293 reasoning)
version one
mistralai/mistral-medium-3-5:low__json_schema PASS 100% 197 in → 176 out (172 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
mistralai/mistral-medium-3-5:medium PASS 100% 204 in → 238 out (251 reasoning)
version one
mistralai/mistral-medium-3-5:medium__json_schema PASS 100% 197 in → 199 out (210 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
mistralai/mistral-medium-3-5:none PASS 100% 210 in → 3 out (0 reasoning)
version one
mistralai/mistral-medium-3-5:none__json_schema PASS 100% 195 in → 8 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
mistralai/mistral-medium-3-5:xhigh PASS 100% 211 in → 351 out (376 reasoning)
version one
mistralai/mistral-medium-3-5:xhigh__json_schema PASS 100% 195 in → 474 out (506 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
moonshotai/kimi-k2.6:high PASS 100% 175 in → 580 out (627 reasoning)
version one
moonshotai/kimi-k2.6:high__json_schema PASS 100% 169 in → 521 out (509 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
moonshotai/kimi-k2.6:low PASS 100% 172 in → 859 out (855 reasoning)
version one
moonshotai/kimi-k2.6:low__json_schema PASS 100% 171 in → 1,147 out (456 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
moonshotai/kimi-k2.6:medium PASS 100% 171 in → 424 out (422 reasoning)
version one
moonshotai/kimi-k2.6:medium__json_schema PASS 100% 170 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
moonshotai/kimi-k2.6:xhigh PASS 100% 176 in → 317 out (315 reasoning)
version one
moonshotai/kimi-k2.6:xhigh__json_schema PASS 100% 172 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
moonshotai/kimi-k2.7-code:high PASS 100% 174 in → 150 out (146 reasoning)
version one
moonshotai/kimi-k2.7-code:high__json_schema PASS 100% 172 in → 190 out (199 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
moonshotai/kimi-k2.7-code:low PASS 100% 171 in → 136 out (132 reasoning)
version one
moonshotai/kimi-k2.7-code:low__json_schema PASS 100% 170 in → 147 out (149 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
moonshotai/kimi-k2.7-code:medium PASS 100% 172 in → 126 out (122 reasoning)
version one
moonshotai/kimi-k2.7-code:xhigh PASS 100% 171 in → 99 out (95 reasoning)
version one
moonshotai/kimi-k2.7-code:xhigh__json_schema PASS 100% 172 in → 146 out (134 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
moonshotai/kimi-k3:high PASS 100% 251 in → 129 out (112 reasoning)
version one
moonshotai/kimi-k3:high__json_schema PASS 100% 246 in → 138 out (121 reasoning)
version one
moonshotai/kimi-k3:low PASS 100% 248 in → 83 out (66 reasoning)
version one
moonshotai/kimi-k3:low__json_schema PASS 100% 247 in → 90 out (73 reasoning)
version one
moonshotai/kimi-k3:max PASS 100% 243 in → 176 out (159 reasoning)
version one
moonshotai/kimi-k3:max__json_schema PASS 100% 248 in → 129 out (112 reasoning)
version one
moonshotai/kimi-k3:medium PASS 100% 251 in → 146 out (129 reasoning)
version one
moonshotai/kimi-k3:medium__json_schema PASS 100% 250 in → 86 out (69 reasoning)
version one
moonshotai/kimi-k3:xhigh PASS 100% 249 in → 167 out (150 reasoning)
version one
moonshotai/kimi-k3:xhigh__json_schema PASS 100% 249 in → 106 out (89 reasoning)
version one
nvidia/nemotron-3-nano-30b-a3b:high PASS 100% 209 in → 131 out (142 reasoning)
version one
nvidia/nemotron-3-nano-30b-a3b:high__json_schema PASS 100% 213 in → 170 out (174 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one"}
nvidia/nemotron-3-nano-30b-a3b:low PASS 100% 212 in → 293 out (319 reasoning)
version one
nvidia/nemotron-3-nano-30b-a3b:low__json_schema PASS 100% 210 in → 239 out (262 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
nvidia/nemotron-3-nano-30b-a3b:medium PASS 100% 209 in → 168 out (186 reasoning)
version one
nvidia/nemotron-3-nano-30b-a3b:medium__json_schema PASS 100% 210 in → 116 out (118 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
nvidia/nemotron-3-nano-30b-a3b:xhigh PASS 100% 209 in → 189 out (206 reasoning)
version one
nvidia/nemotron-3-nano-30b-a3b:xhigh__json_schema PASS 100% 212 in → 2,458 out (2,660 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
nvidia/nemotron-3-super-120b-a12b:high PASS 100% 212 in → 126 out (117 reasoning)
version one
nvidia/nemotron-3-super-120b-a12b:high__json_schema PASS 100% 207 in → 200 out (182 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one\n" }
nvidia/nemotron-3-super-120b-a12b:low PASS 100% 211 in → 161 out (144 reasoning)
version one
nvidia/nemotron-3-super-120b-a12b:low__json_schema PASS 100% 207 in → 219 out (203 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
nvidia/nemotron-3-super-120b-a12b:medium PASS 100% 208 in → 89 out (94 reasoning)
version one
nvidia/nemotron-3-super-120b-a12b:medium__json_schema PASS 100% 210 in → 136 out (121 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
nvidia/nemotron-3-super-120b-a12b:none PASS 100% 211 in → 3 out (0 reasoning)
version one
nvidia/nemotron-3-super-120b-a12b:none__json_schema PASS 100% 213 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
nvidia/nemotron-3-super-120b-a12b:xhigh PASS 100% 210 in → 165 out (175 reasoning)
version one
nvidia/nemotron-3-super-120b-a12b:xhigh__json_schema PASS 100% 209 in → 232 out (211 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
openai/gpt-5.4-mini:high PASS 100% 173 in → 81 out (73 reasoning)
version one
openai/gpt-5.4-mini:high__json_schema PASS 100% 205 in → 97 out (81 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4-mini:low PASS 100% 171 in → 50 out (42 reasoning)
version one
openai/gpt-5.4-mini:low__json_schema PASS 100% 204 in → 63 out (47 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4-mini:medium PASS 100% 172 in → 64 out (56 reasoning)
version one
openai/gpt-5.4-mini:medium__json_schema PASS 100% 206 in → 61 out (45 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4-mini:none PASS 100% 170 in → 12 out (0 reasoning)
version one
openai/gpt-5.4-mini:none__json_schema PASS 100% 200 in → 21 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.4-mini:xhigh PASS 100% 168 in → 256 out (248 reasoning)
version one
openai/gpt-5.4-mini:xhigh__json_schema PASS 100% 202 in → 167 out (151 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4-nano:high PASS 100% 171 in → 66 out (58 reasoning)
version one
openai/gpt-5.4-nano:high__json_schema PASS 100% 202 in → 147 out (130 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.4-nano:low PASS 100% 173 in → 51 out (43 reasoning)
version one
openai/gpt-5.4-nano:low__json_schema PASS 100% 203 in → 62 out (46 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4-nano:medium PASS 100% 174 in → 114 out (106 reasoning)
version one
openai/gpt-5.4-nano:medium__json_schema PASS 100% 200 in → 83 out (66 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.4-nano:none PASS 100% 166 in → 6 out (0 reasoning)
version one
openai/gpt-5.4-nano:none__json_schema PASS 100% 204 in → 14 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4-nano:xhigh PASS 100% 169 in → 168 out (160 reasoning)
version one
openai/gpt-5.4-nano:xhigh__json_schema PASS 100% 209 in → 535 out (519 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4:high PASS 100% 172 in → 59 out (51 reasoning)
version one
openai/gpt-5.4:high__json_schema PASS 100% 206 in → 92 out (76 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4:low PASS 100% 170 in → 43 out (35 reasoning)
version one
openai/gpt-5.4:low__json_schema PASS 100% 204 in → 42 out (26 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4:medium PASS 100% 169 in → 61 out (53 reasoning)
version one
openai/gpt-5.4:medium__json_schema PASS 100% 205 in → 72 out (56 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4:none PASS 100% 172 in → 6 out (0 reasoning)
version one
openai/gpt-5.4:none__json_schema PASS 100% 202 in → 14 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.4:xhigh PASS 100% 174 in → 101 out (93 reasoning)
version one
openai/gpt-5.4:xhigh__json_schema PASS 100% 204 in → 288 out (272 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.5:high PASS 100% 169 in → 60 out (52 reasoning)
version one
openai/gpt-5.5:high__json_schema PASS 100% 201 in → 91 out (74 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.5:low PASS 100% 169 in → 33 out (25 reasoning)
version one
openai/gpt-5.5:low__json_schema PASS 100% 206 in → 37 out (20 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.5:medium PASS 100% 167 in → 38 out (30 reasoning)
version one
openai/gpt-5.5:medium__json_schema PASS 100% 206 in → 46 out (29 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.5:none PASS 100% 171 in → 6 out (0 reasoning)
version one
openai/gpt-5.5:none__json_schema PASS 100% 202 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.5:xhigh PASS 100% 172 in → 126 out (118 reasoning)
version one
openai/gpt-5.5:xhigh__json_schema PASS 100% 204 in → 121 out (104 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.6-luna:high PASS 100% 172 in → 36 out (28 reasoning)
version one
openai/gpt-5.6-luna:high__json_schema PASS 100% 205 in → 42 out (26 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-luna:low PASS 100% 168 in → 26 out (18 reasoning)
version one
openai/gpt-5.6-luna:low__json_schema PASS 100% 205 in → 32 out (16 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-luna:max PASS 100% 173 in → 47 out (39 reasoning)
version one
openai/gpt-5.6-luna:max__json_schema PASS 100% 201 in → 280 out (264 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-luna:medium PASS 100% 169 in → 30 out (22 reasoning)
version one
openai/gpt-5.6-luna:medium__json_schema PASS 100% 206 in → 40 out (24 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-luna:none PASS 100% 170 in → 6 out (0 reasoning)
version one
openai/gpt-5.6-luna:none__json_schema PASS 100% 202 in → 14 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-luna:xhigh PASS 100% 172 in → 43 out (35 reasoning)
version one
openai/gpt-5.6-luna:xhigh__json_schema PASS 100% 206 in → 294 out (278 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-sol:high PASS 100% 170 in → 6 out (0 reasoning)
version one
openai/gpt-5.6-sol:high__json_schema PASS 100% 203 in → 14 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-sol:low PASS 100% 168 in → 6 out (0 reasoning)
version one
openai/gpt-5.6-sol:low__json_schema PASS 100% 206 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.6-sol:max PASS 100% 171 in → 6 out (0 reasoning)
version one
openai/gpt-5.6-sol:max__json_schema PASS 100% 205 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.6-sol:medium PASS 100% 170 in → 6 out (0 reasoning)
version one
openai/gpt-5.6-sol:medium__json_schema PASS 100% 207 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.6-sol:none PASS 100% 172 in → 6 out (0 reasoning)
version one
openai/gpt-5.6-sol:none__json_schema PASS 100% 207 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.6-sol:xhigh PASS 100% 169 in → 6 out (0 reasoning)
version one
openai/gpt-5.6-sol:xhigh__json_schema PASS 100% 203 in → 15 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
openai/gpt-5.6-terra:high PASS 100% 170 in → 28 out (20 reasoning)
version one
openai/gpt-5.6-terra:high__json_schema PASS 100% 203 in → 14 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-terra:low PASS 100% 171 in → 30 out (22 reasoning)
version one
openai/gpt-5.6-terra:low__json_schema PASS 100% 202 in → 29 out (13 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-terra:max PASS 100% 169 in → 73 out (65 reasoning)
version one
openai/gpt-5.6-terra:max__json_schema PASS 100% 202 in → 54 out (38 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-terra:medium PASS 100% 170 in → 31 out (23 reasoning)
version one
openai/gpt-5.6-terra:medium__json_schema PASS 100% 202 in → 14 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-terra:none PASS 100% 173 in → 6 out (0 reasoning)
version one
openai/gpt-5.6-terra:none__json_schema PASS 100% 206 in → 14 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-5.6-terra:xhigh PASS 100% 171 in → 31 out (23 reasoning)
version one
openai/gpt-5.6-terra:xhigh__json_schema PASS 100% 202 in → 46 out (30 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-oss-120b:high PASS 100% 230 in → 333 out (361 reasoning)
version one
openai/gpt-oss-120b:high__json_schema PASS 100% 231 in → 10 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
openai/gpt-oss-120b:low PASS 100% 231 in → 49 out (40 reasoning)
version one
openai/gpt-oss-120b:low__json_schema PASS 100% 220 in → 54 out (38 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
openai/gpt-oss-120b:medium PASS 100% 228 in → 130 out (133 reasoning)
version one
openai/gpt-oss-120b:medium__json_schema PASS 100% 219 in → 205 out (214 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-oss-120b:xhigh PASS 100% 229 in → 346 out (366 reasoning)
version one
openai/gpt-oss-120b:xhigh__json_schema PASS 100% 231 in → 448 out (491 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
openai/gpt-oss-20b:high PASS 100% 221 in → 336 out (354 reasoning)
version one
openai/gpt-oss-20b:high__json_schema PASS 100% 232 in → 397 out (386 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-oss-20b:low PASS 100% 234 in → 38 out (27 reasoning)
version one
openai/gpt-oss-20b:low__json_schema PASS 100% 230 in → 69 out (59 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
openai/gpt-oss-20b:medium PASS 100% 230 in → 167 out (165 reasoning)
version one
openai/gpt-oss-20b:medium__json_schema PASS 100% 230 in → 219 out (202 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
openai/gpt-oss-20b:xhigh PASS 100% 234 in → 744 out (740 reasoning)
version one
openai/gpt-oss-20b:xhigh__json_schema PASS 100% 234 in → 355 out (334 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
poolside/laguna-m.1:high PASS 100% 208 in → 284 out (279 reasoning)
version one
poolside/laguna-m.1:high__json_schema PASS 100% 210 in → 364 out (350 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one\n" }
poolside/laguna-m.1:low PASS 100% 211 in → 291 out (286 reasoning)
version one
poolside/laguna-m.1:low__json_schema PASS 100% 210 in → 290 out (281 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-m.1:medium PASS 100% 211 in → 243 out (238 reasoning)
version one
poolside/laguna-m.1:medium__json_schema PASS 100% 209 in → 226 out (217 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-m.1:none PASS 100% 207 in → 4 out (0 reasoning)
version one
poolside/laguna-m.1:none__json_schema PASS 100% 208 in → 13 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one\n" }
poolside/laguna-m.1:xhigh PASS 100% 211 in → 205 out (200 reasoning)
version one
poolside/laguna-m.1:xhigh__json_schema PASS 100% 207 in → 465 out (452 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
poolside/laguna-xs-2.1:high PASS 100% 210 in → 252 out (249 reasoning)
version one
poolside/laguna-xs-2.1:high__json_schema PASS 100% 208 in → 470 out (461 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-xs-2.1:low PASS 100% 209 in → 309 out (306 reasoning)
version one
poolside/laguna-xs-2.1:low__json_schema PASS 100% 206 in → 246 out (237 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-xs-2.1:medium PASS 100% 209 in → 285 out (282 reasoning)
version one
poolside/laguna-xs-2.1:medium__json_schema PASS 100% 208 in → 230 out (221 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-xs-2.1:none__json_schema PASS 100% 210 in → 8 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-xs-2.1:xhigh PASS 100% 207 in → 408 out (405 reasoning)
version one
poolside/laguna-xs-2.1:xhigh__json_schema PASS 100% 208 in → 305 out (296 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-xs.2:high PASS 100% 245 in → 273 out (268 reasoning)
version one
poolside/laguna-xs.2:high__json_schema PASS 100% 245 in → 544 out (535 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-xs.2:low PASS 100% 240 in → 395 out (390 reasoning)
version one
poolside/laguna-xs.2:low__json_schema PASS 100% 246 in → 294 out (286 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
poolside/laguna-xs.2:medium PASS 100% 244 in → 547 out (542 reasoning)
version one
poolside/laguna-xs.2:medium__json_schema PASS 100% 249 in → 391 out (382 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
poolside/laguna-xs.2:none PASS 100% 245 in → 4 out (0 reasoning)
version one
poolside/laguna-xs.2:none__json_schema PASS 100% 246 in → 8 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one\n"}
poolside/laguna-xs.2:xhigh PASS 100% 246 in → 591 out (586 reasoning)
version one
poolside/laguna-xs.2:xhigh__json_schema PASS 100% 243 in → 567 out (559 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
qwen/qwen3.6-27b:high PASS 100% 205 in → 917 out (1 reasoning)
version one
qwen/qwen3.6-27b:high__json_schema PASS 100% 202 in → 348 out (314 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
qwen/qwen3.6-27b:low PASS 100% 203 in → 281 out (280 reasoning)
version one
qwen/qwen3.6-27b:low__json_schema PASS 100% 209 in → 693 out (673 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
qwen/qwen3.6-27b:medium PASS 100% 204 in → 872 out (1 reasoning)
version one
qwen/qwen3.6-27b:medium__json_schema PASS 100% 202 in → 316 out (283 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
qwen/qwen3.6-27b:none PASS 100% 207 in → 3 out (0 reasoning)
version one
qwen/qwen3.6-27b:xhigh PASS 100% 202 in → 869 out (862 reasoning)
version one
qwen/qwen3.6-27b:xhigh__json_schema PASS 100% 204 in → 353 out (1 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
qwen/qwen3.6-35b-a3b:high PASS 100% 206 in → 919 out (885 reasoning)
version one
qwen/qwen3.6-35b-a3b:high__json_schema PASS 100% 207 in → 532 out (540 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
qwen/qwen3.6-35b-a3b:low PASS 100% 206 in → 960 out (942 reasoning)
version one
qwen/qwen3.6-35b-a3b:low__json_schema PASS 100% 205 in → 972 out (914 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
qwen/qwen3.6-35b-a3b:medium PASS 100% 206 in → 884 out (864 reasoning)
version one
qwen/qwen3.6-35b-a3b:medium__json_schema PASS 100% 207 in → 913 out (877 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
qwen/qwen3.6-35b-a3b:none PASS 100% 208 in → 2 out (0 reasoning)
version one
qwen/qwen3.6-35b-a3b:xhigh PASS 100% 204 in → 552 out (546 reasoning)
version one
qwen/qwen3.6-35b-a3b:xhigh__json_schema PASS 100% 206 in → 1,107 out (1,047 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
qwen/qwen3.6-flash:high PASS 100% 204 in → 1,127 out (1,121 reasoning)
version one
qwen/qwen3.6-flash:low PASS 100% 209 in → 747 out (740 reasoning)
version one
qwen/qwen3.6-flash:medium PASS 100% 206 in → 852 out (846 reasoning)
version one
qwen/qwen3.6-flash:none PASS 100% 211 in → 2 out (0 reasoning)
version one
qwen/qwen3.6-flash:xhigh PASS 100% 204 in → 814 out (808 reasoning)
version one
qwen/qwen3.7-flash:high PASS 100% 204 in → 409 out (402 reasoning)
version one
qwen/qwen3.7-flash:low PASS 100% 206 in → 846 out (840 reasoning)
version one
qwen/qwen3.7-flash:max PASS 100% 204 in → 1,030 out (1,023 reasoning)
version one
qwen/qwen3.7-flash:medium PASS 100% 207 in → 789 out (782 reasoning)
version one
qwen/qwen3.7-flash:none PASS 100% 205 in → 2 out (0 reasoning)
version one
qwen/qwen3.7-flash:xhigh PASS 100% 207 in → 585 out (579 reasoning)
version one
qwen/qwen3.7-max:high PASS 100% 207 in → 303 out (297 reasoning)
version one
qwen/qwen3.7-max:low PASS 100% 206 in → 270 out (263 reasoning)
version one
qwen/qwen3.7-max:medium PASS 100% 202 in → 369 out (363 reasoning)
version one
qwen/qwen3.7-max:none PASS 100% 205 in → 2 out (0 reasoning)
version one
qwen/qwen3.7-max:xhigh PASS 100% 207 in → 154 out (148 reasoning)
version one
qwen/qwen3.7-plus:high PASS 100% 204 in → 258 out (251 reasoning)
version one
qwen/qwen3.7-plus:low PASS 100% 204 in → 412 out (405 reasoning)
version one
qwen/qwen3.7-plus:medium PASS 100% 207 in → 133 out (127 reasoning)
version one
qwen/qwen3.7-plus:none PASS 100% 210 in → 2 out (0 reasoning)
version one
qwen/qwen3.7-plus:xhigh PASS 100% 203 in → 429 out (422 reasoning)
version one
tencent/hy3:high PASS 100% 176 in → 543 out (593 reasoning)
version one
tencent/hy3:high__json_schema PASS 100% 176 in → 728 out (789 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
tencent/hy3:low PASS 100% 179 in → 293 out (295 reasoning)
version one
tencent/hy3:low__json_schema PASS 100% 179 in → 549 out (532 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
tencent/hy3:medium PASS 100% 176 in → 406 out (420 reasoning)
version one
tencent/hy3:medium__json_schema PASS 100% 176 in → 554 out (546 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
tencent/hy3:none PASS 100% 180 in → 3 out (0 reasoning)
version one
tencent/hy3:none__json_schema PASS 100% 177 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one\n" }
tencent/hy3:xhigh PASS 100% 179 in → 266 out (262 reasoning)
version one
tencent/hy3:xhigh__json_schema PASS 100% 175 in → 328 out (343 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
thinkingmachines/inkling-small:high PASS 100% 180 in → 117 out (109 reasoning)
version one
thinkingmachines/inkling-small:high__json_schema PASS 100% 180 in → 142 out (128 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
thinkingmachines/inkling-small:low PASS 100% 182 in → 45 out (43 reasoning)
version one
thinkingmachines/inkling-small:low__json_schema PASS 100% 180 in → 20 out (6 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
thinkingmachines/inkling-small:max PASS 100% 181 in → 273 out (265 reasoning)
version one
thinkingmachines/inkling-small:max__json_schema PASS 100% 186 in → 176 out (162 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
thinkingmachines/inkling-small:medium PASS 100% 182 in → 105 out (104 reasoning)
version one
thinkingmachines/inkling-small:medium__json_schema PASS 100% 185 in → 87 out (72 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one\n" }
thinkingmachines/inkling-small:none PASS 100% 177 in → 5 out (0 reasoning)
version one
thinkingmachines/inkling-small:xhigh PASS 100% 177 in → 51 out (47 reasoning)
version one
thinkingmachines/inkling-small:xhigh__json_schema PASS 100% 178 in → 117 out (103 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
thinkingmachines/inkling:high PASS 100% 181 in → 120 out (124 reasoning)
version one
thinkingmachines/inkling:low PASS 100% 178 in → 28 out (20 reasoning)
version one
thinkingmachines/inkling:max PASS 100% 182 in → 201 out (192 reasoning)
version one
thinkingmachines/inkling:max__json_schema PASS 100% 182 in → 190 out (175 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
thinkingmachines/inkling:medium PASS 100% 179 in → 49 out (40 reasoning)
version one
thinkingmachines/inkling:medium__json_schema PASS 100% 176 in → 54 out (39 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
thinkingmachines/inkling:xhigh PASS 100% 180 in → 164 out (174 reasoning)
version one
thinkingmachines/inkling:xhigh__json_schema PASS 100% 183 in → 113 out (102 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
x-ai/grok-4.3:high PASS 100% 345 in → 175 out (173 reasoning)
version one
x-ai/grok-4.3:high__json_schema PASS 100% 405 in → 617 out (611 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
x-ai/grok-4.3:low PASS 100% 354 in → 327 out (325 reasoning)
version one
x-ai/grok-4.3:low__json_schema PASS 100% 405 in → 369 out (362 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
x-ai/grok-4.3:max PASS 100% 347 in → 343 out (341 reasoning)
version one
x-ai/grok-4.3:max__json_schema PASS 100% 404 in → 534 out (528 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
x-ai/grok-4.3:medium PASS 100% 356 in → 646 out (644 reasoning)
version one
x-ai/grok-4.3:medium__json_schema PASS 100% 411 in → 467 out (457 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
x-ai/grok-4.3:none__json_schema PASS 100% 397 in → 6 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
x-ai/grok-4.3:xhigh PASS 100% 352 in → 282 out (280 reasoning)
version one
x-ai/grok-4.3:xhigh__json_schema PASS 100% 403 in → 410 out (404 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
x-ai/grok-4.5:high PASS 100% 379 in → 341 out (339 reasoning)
version one
x-ai/grok-4.5:high__json_schema PASS 100% 441 in → 479 out (472 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
x-ai/grok-4.5:low PASS 100% 378 in → 124 out (122 reasoning)
version one
x-ai/grok-4.5:low__json_schema PASS 100% 439 in → 274 out (267 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
x-ai/grok-4.5:max PASS 100% 378 in → 414 out (412 reasoning)
version one
x-ai/grok-4.5:max__json_schema PASS 100% 442 in → 628 out (618 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
x-ai/grok-4.5:medium PASS 100% 380 in → 275 out (273 reasoning)
version one
x-ai/grok-4.5:medium__json_schema PASS 100% 441 in → 549 out (542 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
x-ai/grok-4.5:xhigh PASS 100% 373 in → 303 out (301 reasoning)
version one
x-ai/grok-4.5:xhigh__json_schema PASS 100% 442 in → 362 out (355 reasoning)
version one
JSON Schema Structured Output
(raw) {"content":"version one"}
z-ai/glm-4.7-flash:high PASS 100% 175 in → 452 out (418 reasoning)
version one
z-ai/glm-4.7-flash:low PASS 100% 175 in → 453 out (423 reasoning)
version one
z-ai/glm-4.7-flash:medium PASS 100% 174 in → 393 out (373 reasoning)
version one
z-ai/glm-4.7-flash:none PASS 100% 177 in → 2 out (0 reasoning)
version one
z-ai/glm-4.7-flash:none__json_schema PASS 100% 172 in → 8 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) {"content": "version one"}
z-ai/glm-4.7-flash:xhigh PASS 100% 177 in → 332 out (297 reasoning)
version one
z-ai/glm-5.2:high PASS 100% 180 in → 280 out (276 reasoning)
version one
z-ai/glm-5.2:low PASS 100% 182 in → 289 out (285 reasoning)
version one
z-ai/glm-5.2:low__json_schema PASS 100% 186 in → 383 out (369 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
z-ai/glm-5.2:medium PASS 100% 182 in → 98 out (104 reasoning)
version one
z-ai/glm-5.2:none PASS 100% 173 in → 3 out (0 reasoning)
version one
z-ai/glm-5.2:none__json_schema PASS 100% 180 in → 11 out (0 reasoning)
version one
JSON Schema Structured Output
(raw) { "content": "version one" }
z-ai/glm-5.2:xhigh PASS 100% 181 in → 307 out (298 reasoning)
version one
anthropic/claude-fable-5:max FAIL 0% 251 in → 4 out (0 reasoning)
(empty output)
Failure: Expected 'version one', got ''
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)
anthropic/claude-sonnet-4.6:none FAIL 0% 206 in → 37 out (0 reasoning)
Looking at the diff, the second commit changed "version one" to "version two". So `HEAD~1` (the first commit) contained: version one
Failure: Expected 'version one', got 'Looking at the diff, the second commit changed "version one" to "version two". So `HEAD~1` (the first commit) contained: version one'
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)
arcee-ai/trinity-mini:high FAIL 0% 173 in → 574 out (640 reasoning)
-version one
Failure: Expected 'version one', got ' -version one'
arcee-ai/trinity-mini:high__json_schema FAIL 0% 175 in → 546 out (597 reasoning)
-version one
JSON Schema Structured Output
(raw) { "content": "-version one" }
Failure: Expected 'version one', got '-version one'
arcee-ai/trinity-mini:low__json_schema FAIL 0% 171 in → 278 out (314 reasoning)
-version one
JSON Schema Structured Output
(raw) { "content": "-version one" }
Failure: Expected 'version one', got '-version one'
arcee-ai/trinity-mini:medium FAIL 0% 175 in → 450 out (515 reasoning)
-version one
Failure: Expected 'version one', got ' -version one'
arcee-ai/trinity-mini:medium__json_schema FAIL 0% 175 in → 578 out (647 reasoning)
-version one
JSON Schema Structured Output
(raw) { "content": "-version one" }
Failure: Expected 'version one', got '-version one'
arcee-ai/trinity-mini:xhigh__json_schema FAIL 0% 174 in → 342 out (371 reasoning)
-version one
JSON Schema Structured Output
(raw) { "content": "-version one" }
Failure: Expected 'version one', got '-version one'
Invalid JSON. Output: version one
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: version one
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: version one
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)
version two
JSON Schema Structured Output
(raw) { "content": "version two" }
Failure: Expected 'version one', got 'version two'
Invalid JSON. Output: version one
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)
minimax/minimax-m2.5:high FAIL 0% 209 in → 188 out (179 reasoning)
one
Failure: Expected 'version one', got 'one'
minimax/minimax-m2.5:medium FAIL 0% 207 in → 131 out (122 reasoning)
(empty output)
Failure: Expected 'version one', got ''
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)
nvidia/nemotron-3-nano-30b-a3b:none FAIL 0% 211 in → 3 out (0 reasoning)
version two
Failure: Expected 'version one', got 'version two'
nvidia/nemotron-3-nano-30b-a3b:none__json_schema FAIL 0% 208 in → 8 out (0 reasoning)
(empty output)
JSON Schema Structured Output
(raw) { "content": "" }
Failure: Expected 'version one', got ''
poolside/laguna-xs-2.1:none FAIL 0% 206 in → 6 out (0 reasoning)
version one </think>version one
Failure: Expected 'version one', got 'version one </think>version one'
qwen/qwen3.6-27b:none__json_schema FAIL 0% 207 in → 8 out (0 reasoning)
version_one
JSON Schema Structured Output
(raw) {"content": "version_one"}
Failure: Expected 'version one', got 'version_one'
qwen/qwen3.6-35b-a3b:none__json_schema FAIL 0% 207 in → 7 out (0 reasoning)
version two
JSON Schema Structured Output
(raw) {"content":"version two"}
Failure: Expected 'version one', got 'version two'
(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-1b993280-d0ea-9e4f-b67f-1a3a9e54ad15","request_id":"1b993280-d0ea-9e4f-b67f-1a3a9e54ad15"}', '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-2d7dd797-2896-9955-8b1f-edba7789d59d","request_id":"2d7dd797-2896-9955-8b1f-edba7789d59d"}', '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-aa260a1c-1d39-902b-b50c-fe84332a91ef","request_id":"aa260a1c-1d39-902b-b50c-fe84332a91ef"}', '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-2c10e99d-f59d-9ba3-b0a3-478d49e8dbf5","request_id":"2c10e99d-f59d-9ba3-b0a3-478d49e8dbf5"}', '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-2dde968d-6154-9abf-986e-41e36525954c","request_id":"2dde968d-6154-9abf-986e-41e36525954c"}', '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-0eab8160-d4de-92e5-a330-96598dfccbb6"}\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-240219d8-6b96-9532-8d48-bd6112c9223f"}\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-a4346013-4434-9f9d-aa76-76be45bd0c01"}\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-7d216a85-5181-9e3d-b5bd-c21e72ccab80"}\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-79e4dee3-406a-9a4d-8199-63e5c30a5448"}\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-667006f5-2e4c-9365-b9b3-2983c8a0b8d7"}\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-746586b2-fda8-9a01-a23f-31d1aaec49e8","request_id":"746586b2-fda8-9a01-a23f-31d1aaec49e8"}', '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-d2223161-19ed-9174-90fd-37a2d5d0b8f3","request_id":"d2223161-19ed-9174-90fd-37a2d5d0b8f3"}', '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-0397bd94-9a59-9161-ad80-87f4009ce2d3","request_id":"0397bd94-9a59-9161-ad80-87f4009ce2d3"}', '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-3c20a328-6c9b-95f6-8216-8556c86ee032","request_id":"3c20a328-6c9b-95f6-8216-8556c86ee032"}', '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-81bc6538-bf02-9c43-b73c-139063ddfcf5","request_id":"81bc6538-bf02-9c43-b73c-139063ddfcf5"}', '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-cd9a468d-b5d4-9a55-a2b6-a2a520dcee6f","request_id":"cd9a468d-b5d4-9a55-a2b6-a2a520dcee6f"}', '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-a8162197-fa3d-9296-bddd-5f73cfac46e8","request_id":"a8162197-fa3d-9296-bddd-5f73cfac46e8"}', '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-97db0a2d-7770-9646-9e6d-1e510e6bcf34","request_id":"97db0a2d-7770-9646-9e6d-1e510e6bcf34"}', '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-ebfce2d4-0b82-90a6-9318-70df532a0885","request_id":"ebfce2d4-0b82-90a6-9318-70df532a0885"}', '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-7afd0d97-01aa-97e8-8604-8ffffe3dde7c","request_id":"7afd0d97-01aa-97e8-8604-8ffffe3dde7c"}', 'provider_name': 'Alibaba', 'is_byok': False}}, 'user_id': 'user_3FrVBkmFqVIuZ58dEh7AU8vhXUV'}
Invalid JSON. Output: version one{"content": "version one"}
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: version one
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: version one
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)
thinkingmachines/inkling:none FAIL 0% 183 in → 5 out (0 reasoning)
version one<|end_message|>
Failure: Expected 'version one', got 'version one<|end_message|>'
Invalid JSON. Output: version one{"content": "version one"}
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)
x-ai/grok-4.3:none FAIL 0% 346 in → 4 out (0 reasoning)
File content unavailable.
Failure: Expected 'version one', got 'File content unavailable.'
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: version one
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: version one
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: version one
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)