Openai / gpt-oss-20b
medium
96.1%
196 / 204 fixtures · 1 run(s)
46,289 input / 53,293 total output / 47,692 reasoning within output tokens $0.01038587
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-12.7%
Text: 96.1% →
JSON: 83.3%
+3
Gained
JSON pass / text fail
−29
Lost
Text pass / JSON fail
167
Unchanged Pass
Both pass
5
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|---|---|---|
| f010 | 0% (0/1) | 100% (1/1) | -100% |
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| reflog | 100% | 50% | -50% |
| cherry_pick | 83.3% | 41.7% | -41.7% |
| branch_cleanup | 100% | 75% | -25% |
| merge_conflicts | 91.7% | 66.7% | -25% |
| rebase | 91.7% | 75% | -16.7% |
| tag_management | 100% | 83.3% | -16.7% |
| commit_messages | 91.7% | 100% | + 8.3% |
| git_bisect | 100% | 91.7% | -8.3% |
| git_clean | 100% | 91.7% | -8.3% |
| git_log_format | 100% | 91.7% | -8.3% |
| git_show | 100% | 91.7% | -8.3% |
| stash_recovery | 100% | 91.7% | -8.3% |
| worktree_usage | 83.3% | 75% | -8.3% |
| blame_forensics | 100% | 100% | + 0% |
| commit_squash | 91.7% | 91.7% | + 0% |
| git_grep | 100% | 100% | + 0% |
| submodule_usage | 100% | 100% | + 0% |
Changed Fixtures (32)
f003 -loss f009 -loss f011 -loss f001 -loss f002 -loss f003 -loss f004 +gain f008 -loss f009 -loss f011 -loss f010 +gain f003 +gain f006 -loss f011 -loss f012 -loss f007 -loss f009 -loss f009 -loss f010 -loss f011 -loss f010 -loss f011 -loss f001 -loss f007 -loss f008 -loss f009 -loss f011 -loss f012 -loss f005 -loss f008 -loss f012 -loss f007 -loss
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
281
Input
243
Total output
237
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
270
Input
256
Total output
211
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
284
Input
168
Total output
162
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
232
Input
334
Total output
320
Reasoning within output
blame_forensics f005
100%
Passed
Tokens
283
Input
151
Total output
138
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
269
Input
1,837
Total output
1,874
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
243
Input
262
Total output
267
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
253
Input
273
Total output
285
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
231
Input
175
Total output
182
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
426
Input
348
Total output
339
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
248
Input
155
Total output
130
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
188
Input
395
Total output
414
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
157
Input
249
Total output
233
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
164
Input
192
Total output
179
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
147
Input
249
Total output
261
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
186
Input
221
Total output
236
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
160
Input
130
Total output
117
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
184
Input
165
Total output
173
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
157
Input
206
Total output
216
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
177
Input
201
Total output
204
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
139
Input
114
Total output
119
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
158
Input
194
Total output
173
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
133
Input
100
Total output
102
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
209
Input
267
Total output
229
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
149
Input
287
Total output
279
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
174
Input
417
Total output
382
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
145
Input
268
Total output
255
Reasoning within output
cherry_pick f004
0%
Failed
Tokens
198
Input
516
Total output
580
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
191
Input
164
Total output
174
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
189
Input
440
Total output
421
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
180
Input
104
Total output
87
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
165
Input
54
Total output
43
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
151
Input
82
Total output
73
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
261
Input
130
Total output
105
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
215
Input
158
Total output
112
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
313
Input
407
Total output
355
Reasoning within output
commit_messages f001
71%
Passed
Tokens
160
Input
129
Total output
111
Reasoning within output
commit_messages f002
86.7%
Passed
Tokens
254
Input
271
Total output
181
Reasoning within output
commit_messages f003
93.7%
Passed
Tokens
139
Input
157
Total output
111
Reasoning within output
commit_messages f004
61.7%
Passed
Tokens
167
Input
41
Total output
30
Reasoning within output
commit_messages f005
88.3%
Passed
Tokens
155
Input
114
Total output
109
Reasoning within output
commit_messages f006
93.3%
Passed
Tokens
179
Input
277
Total output
238
Reasoning within output
commit_messages f007
79.3%
Passed
Tokens
232
Input
51
Total output
35
Reasoning within output
commit_messages f008
81.7%
Passed
Tokens
131
Input
113
Total output
87
Reasoning within output
commit_messages f009
85%
Passed
Tokens
168
Input
210
Total output
156
Reasoning within output
commit_messages f010
49%
Failed
Tokens
322
Input
250
Total output
182
Reasoning within output
commit_messages f011
78.3%
Passed
Tokens
199
Input
454
Total output
430
Reasoning within output
commit_messages f012
88.3%
Passed
Tokens
180
Input
104
Total output
103
Reasoning within output
commit_squash f001
100%
Passed
Tokens
188
Input
345
Total output
347
Reasoning within output
commit_squash f002
100%
Passed
Tokens
171
Input
218
Total output
204
Reasoning within output
commit_squash f003
50%
Failed
Tokens
178
Input
1,693
Total output
1,797
Reasoning within output
commit_squash f004
100%
Passed
Tokens
156
Input
227
Total output
210
Reasoning within output
commit_squash f005
100%
Passed
Tokens
170
Input
231
Total output
249
Reasoning within output
commit_squash f006
100%
Passed
Tokens
154
Input
177
Total output
166
Reasoning within output
commit_squash f007
100%
Passed
Tokens
156
Input
535
Total output
523
Reasoning within output
commit_squash f008
100%
Passed
Tokens
168
Input
225
Total output
199
Reasoning within output
commit_squash f009
100%
Passed
Tokens
162
Input
336
Total output
325
Reasoning within output
commit_squash f010
100%
Passed
Tokens
176
Input
112
Total output
97
Reasoning within output
commit_squash f011
100%
Passed
Tokens
190
Input
261
Total output
234
Reasoning within output
commit_squash f012
100%
Passed
Tokens
181
Input
391
Total output
371
Reasoning within output
git_bisect f001
100%
Passed
Tokens
214
Input
144
Total output
115
Reasoning within output
git_bisect f002
100%
Passed
Tokens
256
Input
133
Total output
129
Reasoning within output
git_bisect f003
100%
Passed
Tokens
250
Input
425
Total output
410
Reasoning within output
git_bisect f004
100%
Passed
Tokens
235
Input
569
Total output
580
Reasoning within output
git_bisect f005
100%
Passed
Tokens
257
Input
251
Total output
246
Reasoning within output
git_bisect f006
100%
Passed
Tokens
271
Input
216
Total output
185
Reasoning within output
git_bisect f007
100%
Passed
Tokens
233
Input
341
Total output
303
Reasoning within output
git_bisect f008
100%
Passed
Tokens
252
Input
291
Total output
317
Reasoning within output
git_bisect f009
100%
Passed
Tokens
240
Input
329
Total output
318
Reasoning within output
git_bisect f010
100%
Passed
Tokens
252
Input
463
Total output
460
Reasoning within output
git_bisect f011
100%
Passed
Tokens
248
Input
233
Total output
217
Reasoning within output
git_bisect f012
100%
Passed
Tokens
271
Input
630
Total output
618
Reasoning within output
git_clean f001
100%
Passed
Tokens
107
Input
191
Total output
184
Reasoning within output
git_clean f002
100%
Passed
Tokens
118
Input
202
Total output
198
Reasoning within output
git_clean f003
100%
Passed
Tokens
113
Input
184
Total output
193
Reasoning within output
git_clean f004
100%
Passed
Tokens
102
Input
152
Total output
141
Reasoning within output
git_clean f005
100%
Passed
Tokens
107
Input
581
Total output
626
Reasoning within output
git_clean f006
100%
Passed
Tokens
130
Input
524
Total output
517
Reasoning within output
git_clean f007
100%
Passed
Tokens
122
Input
305
Total output
297
Reasoning within output
git_clean f008
100%
Passed
Tokens
127
Input
187
Total output
174
Reasoning within output
git_clean f009
100%
Passed
Tokens
110
Input
377
Total output
417
Reasoning within output
git_clean f010
100%
Passed
Tokens
122
Input
337
Total output
368
Reasoning within output
git_clean f011
100%
Passed
Tokens
117
Input
245
Total output
228
Reasoning within output
git_clean f012
100%
Passed
Tokens
123
Input
508
Total output
526
Reasoning within output
git_grep f001
100%
Passed
Tokens
110
Input
95
Total output
82
Reasoning within output
git_grep f002
100%
Passed
Tokens
109
Input
89
Total output
79
Reasoning within output
git_grep f003
100%
Passed
Tokens
143
Input
108
Total output
97
Reasoning within output
git_grep f004
100%
Passed
Tokens
135
Input
105
Total output
90
Reasoning within output
git_grep f005
100%
Passed
Tokens
172
Input
43
Total output
31
Reasoning within output
git_grep f006
100%
Passed
Tokens
87
Input
180
Total output
193
Reasoning within output
git_grep f007
100%
Passed
Tokens
215
Input
243
Total output
234
Reasoning within output
git_grep f008
100%
Passed
Tokens
117
Input
105
Total output
94
Reasoning within output
git_grep f009
100%
Passed
Tokens
113
Input
118
Total output
115
Reasoning within output
git_grep f010
100%
Passed
Tokens
122
Input
97
Total output
96
Reasoning within output
git_grep f011
100%
Passed
Tokens
165
Input
133
Total output
109
Reasoning within output
git_grep f012
100%
Passed
Tokens
122
Input
132
Total output
120
Reasoning within output
git_log_format f001
100%
Passed
Tokens
789
Input
56
Total output
36
Reasoning within output
git_log_format f002
100%
Passed
Tokens
750
Input
129
Total output
143
Reasoning within output
git_log_format f003
100%
Passed
Tokens
626
Input
258
Total output
225
Reasoning within output
git_log_format f004
100%
Passed
Tokens
754
Input
76
Total output
54
Reasoning within output
git_log_format f005
100%
Passed
Tokens
626
Input
108
Total output
92
Reasoning within output
git_log_format f006
100%
Passed
Tokens
754
Input
113
Total output
112
Reasoning within output
git_log_format f007
100%
Passed
Tokens
766
Input
47
Total output
30
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,001
Input
33
Total output
22
Reasoning within output
git_log_format f009
100%
Passed
Tokens
622
Input
70
Total output
69
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,110
Input
79
Total output
58
Reasoning within output
git_log_format f011
100%
Passed
Tokens
421
Input
136
Total output
130
Reasoning within output
git_log_format f012
100%
Passed
Tokens
373
Input
61
Total output
43
Reasoning within output
git_show f001
100%
Passed
Tokens
226
Input
78
Total output
75
Reasoning within output
git_show f002
100%
Passed
Tokens
226
Input
189
Total output
198
Reasoning within output
git_show f003
100%
Passed
Tokens
620
Input
191
Total output
160
Reasoning within output
git_show f004
100%
Passed
Tokens
444
Input
89
Total output
65
Reasoning within output
git_show f005
100%
Passed
Tokens
230
Input
167
Total output
165
Reasoning within output
git_show f006
100%
Passed
Tokens
241
Input
39
Total output
28
Reasoning within output
git_show f007
100%
Passed
Tokens
211
Input
76
Total output
59
Reasoning within output
git_show f008
100%
Passed
Tokens
225
Input
128
Total output
75
Reasoning within output
git_show f009
100%
Passed
Tokens
319
Input
220
Total output
184
Reasoning within output
git_show f010
100%
Passed
Tokens
232
Input
91
Total output
90
Reasoning within output
git_show f011
100%
Passed
Tokens
254
Input
50
Total output
48
Reasoning within output
git_show f012
100%
Passed
Tokens
358
Input
106
Total output
95
Reasoning within output
merge_conflicts f001
100%
Passed
Tokens
139
Input
476
Total output
520
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
159
Input
442
Total output
369
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
123
Input
860
Total output
963
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
186
Input
654
Total output
687
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
192
Input
221
Total output
241
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
176
Input
380
Total output
336
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
186
Input
211
Total output
222
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
165
Input
89
Total output
80
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
158
Input
460
Total output
514
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
232
Input
417
Total output
448
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
204
Input
339
Total output
275
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
278
Input
401
Total output
394
Reasoning within output
rebase f001
100%
Passed
Tokens
150
Input
346
Total output
332
Reasoning within output
rebase f002
100%
Passed
Tokens
167
Input
208
Total output
193
Reasoning within output
rebase f003
100%
Passed
Tokens
158
Input
161
Total output
164
Reasoning within output
rebase f004
100%
Passed
Tokens
197
Input
449
Total output
427
Reasoning within output
rebase f005
100%
Passed
Tokens
196
Input
158
Total output
176
Reasoning within output
rebase f006
100%
Passed
Tokens
187
Input
515
Total output
496
Reasoning within output
rebase f007
100%
Passed
Tokens
186
Input
113
Total output
106
Reasoning within output
rebase f008
100%
Passed
Tokens
162
Input
90
Total output
86
Reasoning within output
rebase f009
100%
Passed
Tokens
157
Input
94
Total output
88
Reasoning within output
rebase f010
100%
Passed
Tokens
261
Input
240
Total output
207
Reasoning within output
rebase f011
100%
Passed
Tokens
211
Input
195
Total output
148
Reasoning within output
rebase f012
0%
Failed
Tokens
300
Input
994
Total output
1,041
Reasoning within output
reflog f001
100%
Passed
Tokens
260
Input
235
Total output
180
Reasoning within output
reflog f002
100%
Passed
Tokens
260
Input
261
Total output
99
Reasoning within output
reflog f003
100%
Passed
Tokens
239
Input
460
Total output
198
Reasoning within output
reflog f004
100%
Passed
Tokens
332
Input
834
Total output
543
Reasoning within output
reflog f005
100%
Passed
Tokens
382
Input
408
Total output
227
Reasoning within output
reflog f006
100%
Passed
Tokens
279
Input
364
Total output
104
Reasoning within output
reflog f007
100%
Passed
Tokens
298
Input
336
Total output
125
Reasoning within output
reflog f008
100%
Passed
Tokens
389
Input
399
Total output
201
Reasoning within output
reflog f009
100%
Passed
Tokens
340
Input
546
Total output
228
Reasoning within output
reflog f010
100%
Passed
Tokens
294
Input
372
Total output
274
Reasoning within output
reflog f011
100%
Passed
Tokens
329
Input
892
Total output
508
Reasoning within output
reflog f012
100%
Passed
Tokens
327
Input
1,221
Total output
325
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
185
Input
140
Total output
100
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
328
Input
224
Total output
151
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
171
Input
48
Total output
22
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
173
Input
112
Total output
86
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
182
Input
126
Total output
96
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
233
Input
102
Total output
74
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
248
Input
391
Total output
314
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
209
Input
75
Total output
33
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
241
Input
111
Total output
74
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
167
Input
103
Total output
75
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
171
Input
455
Total output
440
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
185
Input
172
Total output
143
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
101
Input
72
Total output
54
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
139
Input
146
Total output
144
Reasoning within output
submodule_usage f003
100%
Passed
Tokens
159
Input
918
Total output
870
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
136
Input
109
Total output
104
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
115
Input
216
Total output
193
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
135
Input
146
Total output
144
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
111
Input
174
Total output
138
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
142
Input
560
Total output
604
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
140
Input
131
Total output
129
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
149
Input
190
Total output
183
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
137
Input
93
Total output
90
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
111
Input
162
Total output
146
Reasoning within output
tag_management f001
100%
Passed
Tokens
112
Input
138
Total output
121
Reasoning within output
tag_management f002
100%
Passed
Tokens
113
Input
80
Total output
50
Reasoning within output
tag_management f003
100%
Passed
Tokens
117
Input
73
Total output
53
Reasoning within output
tag_management f004
100%
Passed
Tokens
129
Input
40
Total output
30
Reasoning within output
tag_management f005
100%
Passed
Tokens
136
Input
88
Total output
81
Reasoning within output
tag_management f006
100%
Passed
Tokens
122
Input
293
Total output
223
Reasoning within output
tag_management f007
100%
Passed
Tokens
124
Input
365
Total output
289
Reasoning within output
tag_management f008
100%
Passed
Tokens
128
Input
267
Total output
195
Reasoning within output
tag_management f009
100%
Passed
Tokens
117
Input
107
Total output
86
Reasoning within output
tag_management f010
100%
Passed
Tokens
108
Input
61
Total output
43
Reasoning within output
tag_management f011
100%
Passed
Tokens
111
Input
129
Total output
124
Reasoning within output
tag_management f012
100%
Passed
Tokens
126
Input
149
Total output
132
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
169
Input
146
Total output
146
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
187
Input
492
Total output
470
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
239
Input
139
Total output
122
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
249
Input
137
Total output
126
Reasoning within output
worktree_usage f005
33.3%
Failed
Tokens
175
Input
137
Total output
137
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
176
Input
231
Total output
207
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
181
Input
233
Total output
206
Reasoning within output
worktree_usage f008
0%
Failed
Tokens
200
Input
582
Total output
543
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
278
Input
143
Total output
155
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
250
Input
450
Total output
421
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
241
Input
186
Total output
187
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
163
Input
120
Total output
88
Reasoning within output