Openai / gpt-oss-20b
low
93.1%
190 / 204 fixtures · 1 run(s)
45,888 input / 18,508 total output / 12,189 reasoning within output tokens $0.00476041
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-13.2%
Text: 93.1% →
JSON: 79.9%
+4
Gained
JSON pass / text fail
−31
Lost
Text pass / JSON fail
159
Unchanged Pass
Both pass
10
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| branch_cleanup | 100% | 50% | -50% |
| git_log_format | 100% | 66.7% | -33.3% |
| tag_management | 100% | 66.7% | -33.3% |
| git_bisect | 100% | 75% | -25% |
| git_show | 100% | 83.3% | -16.7% |
| rebase | 83.3% | 66.7% | -16.7% |
| submodule_usage | 91.7% | 83.3% | -8.3% |
| cherry_pick | 75% | 66.7% | -8.3% |
| commit_squash | 100% | 91.7% | -8.3% |
| git_grep | 100% | 91.7% | -8.3% |
| merge_conflicts | 83.3% | 75% | -8.3% |
| worktree_usage | 83.3% | 75% | -8.3% |
| blame_forensics | 91.7% | 91.7% | + 0% |
| commit_messages | 100% | 100% | + 0% |
| git_clean | 83.3% | 83.3% | + 0% |
| reflog | 91.7% | 91.7% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
Changed Fixtures (35)
f002 -loss f003 -loss f006 -loss f008 -loss f009 -loss f011 -loss f011 -loss f007 -loss f003 -loss f005 -loss f010 -loss f005 -loss f006 -loss f008 +gain f012 +gain f004 -loss f001 -loss f004 -loss f006 -loss f007 -loss f002 -loss f010 -loss f001 +gain f006 -loss f011 -loss f010 -loss f011 -loss f010 +gain f012 -loss f009 -loss f007 -loss f008 -loss f009 -loss f011 -loss f010 -loss
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
276
Input
303
Total output
298
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
263
Input
60
Total output
47
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
286
Input
44
Total output
30
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
218
Input
45
Total output
36
Reasoning within output
blame_forensics f005
100%
Passed
Tokens
274
Input
57
Total output
39
Reasoning within output
blame_forensics f006
0%
Failed
Tokens
255
Input
243
Total output
212
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
242
Input
145
Total output
141
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
274
Input
224
Total output
229
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
222
Input
134
Total output
124
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
402
Input
71
Total output
67
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
265
Input
117
Total output
104
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
172
Input
120
Total output
110
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
165
Input
80
Total output
67
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
171
Input
76
Total output
60
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
148
Input
45
Total output
39
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
188
Input
183
Total output
178
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
148
Input
61
Total output
53
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
166
Input
50
Total output
37
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
156
Input
77
Total output
72
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
172
Input
86
Total output
70
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
123
Input
38
Total output
30
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
170
Input
30
Total output
15
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
130
Input
15
Total output
5
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
205
Input
50
Total output
31
Reasoning within output
cherry_pick f001
0%
Failed
Tokens
147
Input
84
Total output
78
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
169
Input
48
Total output
25
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
130
Input
87
Total output
82
Reasoning within output
cherry_pick f004
0%
Failed
Tokens
186
Input
104
Total output
85
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
203
Input
51
Total output
27
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
191
Input
114
Total output
73
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
166
Input
38
Total output
22
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
180
Input
30
Total output
15
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
150
Input
30
Total output
15
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
263
Input
113
Total output
97
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
216
Input
79
Total output
46
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
315
Input
464
Total output
413
Reasoning within output
commit_messages f001
92.7%
Passed
Tokens
160
Input
43
Total output
27
Reasoning within output
commit_messages f002
83%
Passed
Tokens
243
Input
63
Total output
40
Reasoning within output
commit_messages f003
94.3%
Passed
Tokens
128
Input
28
Total output
10
Reasoning within output
commit_messages f004
94%
Passed
Tokens
156
Input
40
Total output
24
Reasoning within output
commit_messages f005
88.3%
Passed
Tokens
157
Input
45
Total output
32
Reasoning within output
commit_messages f006
91%
Passed
Tokens
179
Input
45
Total output
30
Reasoning within output
commit_messages f007
91.7%
Passed
Tokens
221
Input
49
Total output
33
Reasoning within output
commit_messages f008
92.7%
Passed
Tokens
127
Input
41
Total output
30
Reasoning within output
commit_messages f009
91%
Passed
Tokens
179
Input
55
Total output
35
Reasoning within output
commit_messages f010
91.3%
Passed
Tokens
319
Input
66
Total output
49
Reasoning within output
commit_messages f011
84.3%
Passed
Tokens
197
Input
191
Total output
173
Reasoning within output
commit_messages f012
91.7%
Passed
Tokens
184
Input
37
Total output
23
Reasoning within output
commit_squash f001
100%
Passed
Tokens
194
Input
92
Total output
66
Reasoning within output
commit_squash f002
100%
Passed
Tokens
171
Input
75
Total output
59
Reasoning within output
commit_squash f003
100%
Passed
Tokens
173
Input
148
Total output
100
Reasoning within output
commit_squash f004
100%
Passed
Tokens
172
Input
43
Total output
22
Reasoning within output
commit_squash f005
100%
Passed
Tokens
171
Input
37
Total output
24
Reasoning within output
commit_squash f006
100%
Passed
Tokens
156
Input
110
Total output
89
Reasoning within output
commit_squash f007
100%
Passed
Tokens
142
Input
41
Total output
15
Reasoning within output
commit_squash f008
100%
Passed
Tokens
173
Input
66
Total output
45
Reasoning within output
commit_squash f009
100%
Passed
Tokens
160
Input
55
Total output
26
Reasoning within output
commit_squash f010
100%
Passed
Tokens
166
Input
53
Total output
23
Reasoning within output
commit_squash f011
100%
Passed
Tokens
184
Input
52
Total output
22
Reasoning within output
commit_squash f012
100%
Passed
Tokens
173
Input
80
Total output
42
Reasoning within output
git_bisect f001
100%
Passed
Tokens
214
Input
27
Total output
11
Reasoning within output
git_bisect f002
100%
Passed
Tokens
254
Input
166
Total output
49
Reasoning within output
git_bisect f003
100%
Passed
Tokens
237
Input
108
Total output
81
Reasoning within output
git_bisect f004
100%
Passed
Tokens
252
Input
41
Total output
27
Reasoning within output
git_bisect f005
100%
Passed
Tokens
246
Input
425
Total output
421
Reasoning within output
git_bisect f006
100%
Passed
Tokens
260
Input
69
Total output
54
Reasoning within output
git_bisect f007
100%
Passed
Tokens
235
Input
83
Total output
51
Reasoning within output
git_bisect f008
100%
Passed
Tokens
250
Input
107
Total output
94
Reasoning within output
git_bisect f009
100%
Passed
Tokens
237
Input
85
Total output
37
Reasoning within output
git_bisect f010
100%
Passed
Tokens
239
Input
101
Total output
61
Reasoning within output
git_bisect f011
100%
Passed
Tokens
245
Input
127
Total output
106
Reasoning within output
git_bisect f012
100%
Passed
Tokens
277
Input
67
Total output
51
Reasoning within output
git_clean f001
100%
Passed
Tokens
112
Input
71
Total output
54
Reasoning within output
git_clean f002
100%
Passed
Tokens
123
Input
32
Total output
16
Reasoning within output
git_clean f003
100%
Passed
Tokens
115
Input
141
Total output
137
Reasoning within output
git_clean f004
100%
Passed
Tokens
120
Input
46
Total output
28
Reasoning within output
git_clean f005
100%
Passed
Tokens
94
Input
218
Total output
223
Reasoning within output
git_clean f006
100%
Passed
Tokens
130
Input
97
Total output
71
Reasoning within output
git_clean f007
100%
Passed
Tokens
120
Input
49
Total output
32
Reasoning within output
git_clean f008
25%
Failed
Tokens
127
Input
27
Total output
12
Reasoning within output
git_clean f009
100%
Passed
Tokens
110
Input
85
Total output
73
Reasoning within output
git_clean f010
100%
Passed
Tokens
120
Input
241
Total output
235
Reasoning within output
git_clean f011
100%
Passed
Tokens
106
Input
94
Total output
78
Reasoning within output
git_clean f012
83.3%
Failed
Tokens
108
Input
207
Total output
184
Reasoning within output
git_grep f001
100%
Passed
Tokens
112
Input
46
Total output
43
Reasoning within output
git_grep f002
100%
Passed
Tokens
124
Input
18
Total output
7
Reasoning within output
git_grep f003
100%
Passed
Tokens
143
Input
17
Total output
4
Reasoning within output
git_grep f004
100%
Passed
Tokens
137
Input
46
Total output
22
Reasoning within output
git_grep f005
100%
Passed
Tokens
170
Input
18
Total output
6
Reasoning within output
git_grep f006
100%
Passed
Tokens
100
Input
40
Total output
34
Reasoning within output
git_grep f007
100%
Passed
Tokens
213
Input
147
Total output
129
Reasoning within output
git_grep f008
100%
Passed
Tokens
117
Input
23
Total output
13
Reasoning within output
git_grep f009
100%
Passed
Tokens
111
Input
27
Total output
13
Reasoning within output
git_grep f010
100%
Passed
Tokens
127
Input
17
Total output
6
Reasoning within output
git_grep f011
100%
Passed
Tokens
167
Input
17
Total output
4
Reasoning within output
git_grep f012
100%
Passed
Tokens
122
Input
32
Total output
15
Reasoning within output
git_log_format f001
100%
Passed
Tokens
752
Input
22
Total output
11
Reasoning within output
git_log_format f002
100%
Passed
Tokens
737
Input
42
Total output
32
Reasoning within output
git_log_format f003
100%
Passed
Tokens
613
Input
44
Total output
27
Reasoning within output
git_log_format f004
100%
Passed
Tokens
728
Input
24
Total output
14
Reasoning within output
git_log_format f005
100%
Passed
Tokens
624
Input
166
Total output
161
Reasoning within output
git_log_format f006
100%
Passed
Tokens
741
Input
55
Total output
33
Reasoning within output
git_log_format f007
100%
Passed
Tokens
771
Input
26
Total output
8
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,002
Input
14
Total output
4
Reasoning within output
git_log_format f009
100%
Passed
Tokens
644
Input
16
Total output
4
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,122
Input
17
Total output
4
Reasoning within output
git_log_format f011
100%
Passed
Tokens
387
Input
52
Total output
35
Reasoning within output
git_log_format f012
100%
Passed
Tokens
380
Input
22
Total output
11
Reasoning within output
git_show f001
100%
Passed
Tokens
231
Input
37
Total output
25
Reasoning within output
git_show f002
100%
Passed
Tokens
224
Input
326
Total output
324
Reasoning within output
git_show f003
100%
Passed
Tokens
620
Input
365
Total output
359
Reasoning within output
git_show f004
100%
Passed
Tokens
448
Input
25
Total output
9
Reasoning within output
git_show f005
100%
Passed
Tokens
234
Input
38
Total output
27
Reasoning within output
git_show f006
100%
Passed
Tokens
243
Input
22
Total output
12
Reasoning within output
git_show f007
100%
Passed
Tokens
229
Input
30
Total output
13
Reasoning within output
git_show f008
100%
Passed
Tokens
222
Input
66
Total output
26
Reasoning within output
git_show f009
100%
Passed
Tokens
320
Input
37
Total output
18
Reasoning within output
git_show f010
100%
Passed
Tokens
213
Input
31
Total output
19
Reasoning within output
git_show f011
100%
Passed
Tokens
255
Input
20
Total output
8
Reasoning within output
git_show f012
100%
Passed
Tokens
333
Input
36
Total output
25
Reasoning within output
merge_conflicts f001
0%
Failed
Tokens
143
Input
119
Total output
116
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
165
Input
72
Total output
56
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
138
Input
73
Total output
71
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
187
Input
289
Total output
292
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
192
Input
260
Total output
244
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
163
Input
90
Total output
54
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
181
Input
166
Total output
158
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
163
Input
43
Total output
27
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
160
Input
65
Total output
54
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
228
Input
83
Total output
56
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
204
Input
91
Total output
53
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
280
Input
131
Total output
75
Reasoning within output
rebase f001
0%
Failed
Tokens
150
Input
68
Total output
56
Reasoning within output
rebase f002
100%
Passed
Tokens
166
Input
44
Total output
21
Reasoning within output
rebase f003
100%
Passed
Tokens
158
Input
31
Total output
18
Reasoning within output
rebase f004
100%
Passed
Tokens
199
Input
69
Total output
40
Reasoning within output
rebase f005
100%
Passed
Tokens
199
Input
85
Total output
72
Reasoning within output
rebase f006
100%
Passed
Tokens
188
Input
310
Total output
291
Reasoning within output
rebase f007
100%
Passed
Tokens
182
Input
40
Total output
24
Reasoning within output
rebase f008
100%
Passed
Tokens
163
Input
42
Total output
27
Reasoning within output
rebase f009
100%
Passed
Tokens
159
Input
29
Total output
15
Reasoning within output
rebase f010
100%
Passed
Tokens
257
Input
65
Total output
33
Reasoning within output
rebase f011
100%
Passed
Tokens
207
Input
74
Total output
33
Reasoning within output
rebase f012
0%
Failed
Tokens
309
Input
116
Total output
59
Reasoning within output
reflog f001
100%
Passed
Tokens
251
Input
82
Total output
36
Reasoning within output
reflog f002
100%
Passed
Tokens
241
Input
213
Total output
24
Reasoning within output
reflog f003
100%
Passed
Tokens
227
Input
376
Total output
35
Reasoning within output
reflog f004
100%
Passed
Tokens
315
Input
397
Total output
116
Reasoning within output
reflog f005
100%
Passed
Tokens
363
Input
122
Total output
19
Reasoning within output
reflog f006
100%
Passed
Tokens
257
Input
282
Total output
29
Reasoning within output
reflog f007
100%
Passed
Tokens
303
Input
289
Total output
129
Reasoning within output
reflog f008
100%
Passed
Tokens
360
Input
192
Total output
33
Reasoning within output
reflog f009
100%
Passed
Tokens
328
Input
198
Total output
44
Reasoning within output
reflog f010
0%
Failed
Tokens
298
Input
47
Total output
18
Reasoning within output
reflog f011
100%
Passed
Tokens
336
Input
197
Total output
72
Reasoning within output
reflog f012
100%
Passed
Tokens
338
Input
574
Total output
44
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
185
Input
40
Total output
12
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
329
Input
101
Total output
25
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
173
Input
107
Total output
86
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
158
Input
34
Total output
10
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
180
Input
107
Total output
58
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
230
Input
36
Total output
5
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
255
Input
50
Total output
21
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
222
Input
70
Total output
9
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
256
Input
71
Total output
16
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
180
Input
137
Total output
85
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
167
Input
46
Total output
13
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
185
Input
91
Total output
36
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
90
Input
33
Total output
13
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
151
Input
68
Total output
44
Reasoning within output
submodule_usage f003
75%
Failed
Tokens
153
Input
74
Total output
8
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
126
Input
55
Total output
46
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
112
Input
102
Total output
71
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
140
Input
22
Total output
9
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
111
Input
148
Total output
129
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
145
Input
52
Total output
38
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
146
Input
74
Total output
69
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
150
Input
189
Total output
182
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
141
Input
31
Total output
17
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
111
Input
71
Total output
49
Reasoning within output
tag_management f001
100%
Passed
Tokens
102
Input
40
Total output
19
Reasoning within output
tag_management f002
100%
Passed
Tokens
101
Input
74
Total output
40
Reasoning within output
tag_management f003
100%
Passed
Tokens
105
Input
50
Total output
24
Reasoning within output
tag_management f004
100%
Passed
Tokens
114
Input
26
Total output
14
Reasoning within output
tag_management f005
100%
Passed
Tokens
121
Input
49
Total output
29
Reasoning within output
tag_management f006
100%
Passed
Tokens
110
Input
147
Total output
82
Reasoning within output
tag_management f007
100%
Passed
Tokens
124
Input
52
Total output
24
Reasoning within output
tag_management f008
100%
Passed
Tokens
124
Input
142
Total output
72
Reasoning within output
tag_management f009
100%
Passed
Tokens
127
Input
51
Total output
20
Reasoning within output
tag_management f010
100%
Passed
Tokens
106
Input
28
Total output
8
Reasoning within output
tag_management f011
100%
Passed
Tokens
107
Input
41
Total output
30
Reasoning within output
tag_management f012
100%
Passed
Tokens
139
Input
89
Total output
62
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
185
Input
172
Total output
163
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
185
Input
59
Total output
31
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
244
Input
35
Total output
17
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
228
Input
26
Total output
13
Reasoning within output
worktree_usage f005
33.3%
Failed
Tokens
181
Input
23
Total output
4
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
180
Input
42
Total output
16
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
167
Input
67
Total output
40
Reasoning within output
worktree_usage f008
50%
Failed
Tokens
205
Input
75
Total output
22
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
273
Input
24
Total output
10
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
245
Input
41
Total output
16
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
247
Input
36
Total output
21
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
175
Input
40
Total output
15
Reasoning within output