Google / gemma-4-26b-a4b-it
high
96.1%
196 / 204 fixtures · 1 run(s)
43,880 input / 136,112 total output / 115,375 reasoning within output tokens $0.06592534
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-4.9%
Text: 96.1% →
JSON: 91.2%
+2
Gained
JSON pass / text fail
−12
Lost
Text pass / JSON fail
184
Unchanged Pass
Both pass
6
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| branch_cleanup | 100% | 75% | -25% |
| stash_recovery | 100% | 83.3% | -16.7% |
| git_clean | 91.7% | 83.3% | -8.3% |
| merge_conflicts | 91.7% | 83.3% | -8.3% |
| rebase | 91.7% | 83.3% | -8.3% |
| cherry_pick | 83.3% | 75% | -8.3% |
| git_bisect | 100% | 91.7% | -8.3% |
| blame_forensics | 100% | 100% | + 0% |
| commit_messages | 100% | 100% | + 0% |
| commit_squash | 100% | 100% | + 0% |
| git_grep | 100% | 100% | + 0% |
| git_log_format | 100% | 100% | + 0% |
| git_show | 91.7% | 91.7% | + 0% |
| reflog | 100% | 100% | + 0% |
| submodule_usage | 100% | 100% | + 0% |
| tag_management | 100% | 100% | + 0% |
| worktree_usage | 83.3% | 83.3% | + 0% |
Changed Fixtures (14)
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
275
Input
1,151
Total output
1,146
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
273
Input
629
Total output
509
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
309
Input
365
Total output
358
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
215
Input
440
Total output
418
Reasoning within output
blame_forensics f005
100%
Passed
Tokens
286
Input
1,538
Total output
1,205
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
259
Input
670
Total output
492
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
234
Input
491
Total output
379
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
276
Input
597
Total output
511
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
218
Input
753
Total output
674
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
477
Input
982
Total output
713
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
272
Input
507
Total output
358
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
171
Input
528
Total output
426
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
115
Input
370
Total output
314
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
130
Input
336
Total output
291
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
108
Input
793
Total output
739
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
148
Input
324
Total output
252
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
118
Input
279
Total output
231
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
142
Input
337
Total output
284
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
120
Input
277
Total output
233
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
134
Input
325
Total output
250
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
92
Input
262
Total output
244
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
127
Input
242
Total output
173
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
80
Input
312
Total output
310
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
171
Input
410
Total output
294
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
111
Input
608
Total output
574
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
128
Input
369
Total output
303
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
99
Input
392
Total output
385
Reasoning within output
cherry_pick f004
0%
Failed
Tokens
179
Input
10,587
Total output
8,868
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
168
Input
408
Total output
355
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
156
Input
370
Total output
249
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
148
Input
291
Total output
280
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
139
Input
194
Total output
157
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
120
Input
169
Total output
137
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
242
Input
936
Total output
864
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
180
Input
317
Total output
217
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
299
Input
4,963
Total output
4,713
Reasoning within output
commit_messages f001
94.3%
Passed
Tokens
127
Input
268
Total output
206
Reasoning within output
commit_messages f002
91.7%
Passed
Tokens
261
Input
498
Total output
466
Reasoning within output
commit_messages f003
90%
Passed
Tokens
99
Input
293
Total output
282
Reasoning within output
commit_messages f004
94.3%
Passed
Tokens
134
Input
307
Total output
266
Reasoning within output
commit_messages f005
91%
Passed
Tokens
125
Input
480
Total output
415
Reasoning within output
commit_messages f006
87.7%
Passed
Tokens
167
Input
990
Total output
973
Reasoning within output
commit_messages f007
91.7%
Passed
Tokens
225
Input
254
Total output
215
Reasoning within output
commit_messages f008
91%
Passed
Tokens
92
Input
363
Total output
302
Reasoning within output
commit_messages f009
86.7%
Passed
Tokens
148
Input
694
Total output
626
Reasoning within output
commit_messages f010
91.7%
Passed
Tokens
348
Input
465
Total output
428
Reasoning within output
commit_messages f011
82.3%
Passed
Tokens
169
Input
1,146
Total output
946
Reasoning within output
commit_messages f012
91%
Passed
Tokens
145
Input
428
Total output
411
Reasoning within output
commit_squash f001
100%
Passed
Tokens
149
Input
2,871
Total output
2,857
Reasoning within output
commit_squash f002
100%
Passed
Tokens
131
Input
684
Total output
572
Reasoning within output
commit_squash f003
100%
Passed
Tokens
137
Input
1,105
Total output
865
Reasoning within output
commit_squash f004
100%
Passed
Tokens
124
Input
589
Total output
493
Reasoning within output
commit_squash f005
100%
Passed
Tokens
128
Input
1,045
Total output
930
Reasoning within output
commit_squash f006
100%
Passed
Tokens
109
Input
305
Total output
250
Reasoning within output
commit_squash f007
100%
Passed
Tokens
114
Input
348
Total output
270
Reasoning within output
commit_squash f008
100%
Passed
Tokens
127
Input
440
Total output
343
Reasoning within output
commit_squash f009
100%
Passed
Tokens
120
Input
897
Total output
885
Reasoning within output
commit_squash f010
100%
Passed
Tokens
135
Input
376
Total output
360
Reasoning within output
commit_squash f011
100%
Passed
Tokens
141
Input
287
Total output
271
Reasoning within output
commit_squash f012
100%
Passed
Tokens
140
Input
467
Total output
391
Reasoning within output
git_bisect f001
100%
Passed
Tokens
201
Input
187
Total output
179
Reasoning within output
git_bisect f002
100%
Passed
Tokens
227
Input
328
Total output
320
Reasoning within output
git_bisect f003
100%
Passed
Tokens
217
Input
302
Total output
296
Reasoning within output
git_bisect f004
100%
Passed
Tokens
223
Input
229
Total output
222
Reasoning within output
git_bisect f005
100%
Passed
Tokens
229
Input
333
Total output
225
Reasoning within output
git_bisect f006
100%
Passed
Tokens
253
Input
490
Total output
337
Reasoning within output
git_bisect f007
100%
Passed
Tokens
231
Input
229
Total output
221
Reasoning within output
git_bisect f008
100%
Passed
Tokens
229
Input
490
Total output
482
Reasoning within output
git_bisect f009
100%
Passed
Tokens
221
Input
228
Total output
222
Reasoning within output
git_bisect f010
100%
Passed
Tokens
231
Input
288
Total output
192
Reasoning within output
git_bisect f011
100%
Passed
Tokens
231
Input
265
Total output
164
Reasoning within output
git_bisect f012
100%
Passed
Tokens
255
Input
265
Total output
257
Reasoning within output
git_clean f001
100%
Passed
Tokens
66
Input
182
Total output
177
Reasoning within output
git_clean f002
100%
Passed
Tokens
78
Input
200
Total output
195
Reasoning within output
git_clean f003
100%
Passed
Tokens
78
Input
217
Total output
194
Reasoning within output
git_clean f004
100%
Passed
Tokens
78
Input
451
Total output
445
Reasoning within output
git_clean f005
100%
Passed
Tokens
65
Input
289
Total output
251
Reasoning within output
git_clean f006
100%
Passed
Tokens
96
Input
983
Total output
870
Reasoning within output
git_clean f007
100%
Passed
Tokens
85
Input
285
Total output
234
Reasoning within output
git_clean f008
100%
Passed
Tokens
92
Input
428
Total output
390
Reasoning within output
git_clean f009
100%
Passed
Tokens
88
Input
331
Total output
280
Reasoning within output
git_clean f010
100%
Passed
Tokens
79
Input
252
Total output
229
Reasoning within output
git_clean f011
60%
Failed
Tokens
80
Input
626
Total output
564
Reasoning within output
git_clean f012
100%
Passed
Tokens
85
Input
271
Total output
263
Reasoning within output
git_grep f001
100%
Passed
Tokens
63
Input
200
Total output
169
Reasoning within output
git_grep f002
100%
Passed
Tokens
75
Input
169
Total output
130
Reasoning within output
git_grep f003
100%
Passed
Tokens
115
Input
209
Total output
152
Reasoning within output
git_grep f004
100%
Passed
Tokens
91
Input
304
Total output
225
Reasoning within output
git_grep f005
100%
Passed
Tokens
130
Input
241
Total output
239
Reasoning within output
git_grep f006
100%
Passed
Tokens
48
Input
772
Total output
801
Reasoning within output
git_grep f007
100%
Passed
Tokens
191
Input
1,947
Total output
1,365
Reasoning within output
git_grep f008
100%
Passed
Tokens
71
Input
161
Total output
136
Reasoning within output
git_grep f009
100%
Passed
Tokens
65
Input
126
Total output
95
Reasoning within output
git_grep f010
100%
Passed
Tokens
78
Input
335
Total output
287
Reasoning within output
git_grep f011
100%
Passed
Tokens
142
Input
640
Total output
638
Reasoning within output
git_grep f012
100%
Passed
Tokens
80
Input
249
Total output
174
Reasoning within output
git_log_format f001
100%
Passed
Tokens
951
Input
595
Total output
486
Reasoning within output
git_log_format f002
100%
Passed
Tokens
922
Input
434
Total output
367
Reasoning within output
git_log_format f003
100%
Passed
Tokens
756
Input
692
Total output
380
Reasoning within output
git_log_format f004
100%
Passed
Tokens
918
Input
663
Total output
365
Reasoning within output
git_log_format f005
100%
Passed
Tokens
762
Input
634
Total output
490
Reasoning within output
git_log_format f006
100%
Passed
Tokens
950
Input
430
Total output
211
Reasoning within output
git_log_format f007
100%
Passed
Tokens
954
Input
315
Total output
235
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,248
Input
276
Total output
174
Reasoning within output
git_log_format f009
100%
Passed
Tokens
773
Input
425
Total output
339
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,469
Input
632
Total output
445
Reasoning within output
git_log_format f011
100%
Passed
Tokens
450
Input
254
Total output
183
Reasoning within output
git_log_format f012
100%
Passed
Tokens
424
Input
331
Total output
238
Reasoning within output
git_show f001
100%
Passed
Tokens
212
Input
232
Total output
226
Reasoning within output
git_show f002
0%
Failed
Tokens
223
Input
3,449
Total output
2,937
Reasoning within output
git_show f003
100%
Passed
Tokens
727
Input
489
Total output
279
Reasoning within output
git_show f004
100%
Passed
Tokens
499
Input
306
Total output
235
Reasoning within output
git_show f005
100%
Passed
Tokens
222
Input
498
Total output
410
Reasoning within output
git_show f006
100%
Passed
Tokens
231
Input
263
Total output
201
Reasoning within output
git_show f007
100%
Passed
Tokens
219
Input
291
Total output
285
Reasoning within output
git_show f008
100%
Passed
Tokens
227
Input
354
Total output
192
Reasoning within output
git_show f009
100%
Passed
Tokens
331
Input
706
Total output
525
Reasoning within output
git_show f010
100%
Passed
Tokens
220
Input
414
Total output
346
Reasoning within output
git_show f011
100%
Passed
Tokens
242
Input
303
Total output
244
Reasoning within output
git_show f012
100%
Passed
Tokens
377
Input
323
Total output
256
Reasoning within output
merge_conflicts f001
100%
Passed
Tokens
98
Input
2,514
Total output
2,509
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
116
Input
444
Total output
436
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
89
Input
1,388
Total output
1,382
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
166
Input
1,839
Total output
1,812
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
156
Input
551
Total output
532
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
142
Input
1,482
Total output
1,158
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
152
Input
400
Total output
389
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
123
Input
203
Total output
185
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
114
Input
446
Total output
417
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
206
Input
1,037
Total output
942
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
159
Input
771
Total output
599
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
263
Input
6,337
Total output
5,452
Reasoning within output
rebase f001
100%
Passed
Tokens
111
Input
3,117
Total output
3,108
Reasoning within output
rebase f002
100%
Passed
Tokens
138
Input
236
Total output
188
Reasoning within output
rebase f003
100%
Passed
Tokens
116
Input
417
Total output
403
Reasoning within output
rebase f004
100%
Passed
Tokens
177
Input
710
Total output
558
Reasoning within output
rebase f005
100%
Passed
Tokens
163
Input
1,401
Total output
1,384
Reasoning within output
rebase f006
100%
Passed
Tokens
155
Input
1,737
Total output
1,403
Reasoning within output
rebase f007
100%
Passed
Tokens
153
Input
334
Total output
273
Reasoning within output
rebase f008
100%
Passed
Tokens
137
Input
193
Total output
168
Reasoning within output
rebase f009
100%
Passed
Tokens
116
Input
396
Total output
358
Reasoning within output
rebase f010
100%
Passed
Tokens
235
Input
3,046
Total output
2,716
Reasoning within output
rebase f011
100%
Passed
Tokens
171
Input
1,417
Total output
1,121
Reasoning within output
rebase f012
0%
Failed
Tokens
298
Input
1,753
Total output
1,368
Reasoning within output
reflog f001
100%
Passed
Tokens
261
Input
798
Total output
471
Reasoning within output
reflog f002
100%
Passed
Tokens
260
Input
595
Total output
307
Reasoning within output
reflog f003
100%
Passed
Tokens
213
Input
645
Total output
327
Reasoning within output
reflog f004
100%
Passed
Tokens
367
Input
873
Total output
382
Reasoning within output
reflog f005
100%
Passed
Tokens
413
Input
643
Total output
339
Reasoning within output
reflog f006
100%
Passed
Tokens
272
Input
552
Total output
287
Reasoning within output
reflog f007
100%
Passed
Tokens
309
Input
689
Total output
316
Reasoning within output
reflog f008
100%
Passed
Tokens
394
Input
890
Total output
443
Reasoning within output
reflog f009
100%
Passed
Tokens
357
Input
754
Total output
518
Reasoning within output
reflog f010
100%
Passed
Tokens
304
Input
527
Total output
380
Reasoning within output
reflog f011
100%
Passed
Tokens
354
Input
1,343
Total output
1,063
Reasoning within output
reflog f012
100%
Passed
Tokens
352
Input
911
Total output
464
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
151
Input
180
Total output
162
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
323
Input
327
Total output
314
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
143
Input
134
Total output
117
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
136
Input
178
Total output
156
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
146
Input
256
Total output
207
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
212
Input
242
Total output
226
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
236
Input
204
Total output
178
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
208
Input
195
Total output
163
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
237
Input
289
Total output
229
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
144
Input
178
Total output
151
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
137
Input
301
Total output
283
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
159
Input
178
Total output
166
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
52
Input
131
Total output
105
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
114
Input
835
Total output
827
Reasoning within output
submodule_usage f003
100%
Passed
Tokens
122
Input
1,010
Total output
882
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
112
Input
174
Total output
170
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
59
Input
630
Total output
603
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
110
Input
660
Total output
656
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
59
Input
288
Total output
230
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
118
Input
303
Total output
309
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
107
Input
181
Total output
177
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
119
Input
626
Total output
671
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
122
Input
205
Total output
177
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
56
Input
179
Total output
140
Reasoning within output
tag_management f001
100%
Passed
Tokens
71
Input
255
Total output
248
Reasoning within output
tag_management f002
100%
Passed
Tokens
69
Input
197
Total output
143
Reasoning within output
tag_management f003
100%
Passed
Tokens
77
Input
273
Total output
262
Reasoning within output
tag_management f004
100%
Passed
Tokens
84
Input
156
Total output
112
Reasoning within output
tag_management f005
100%
Passed
Tokens
95
Input
232
Total output
224
Reasoning within output
tag_management f006
100%
Passed
Tokens
81
Input
927
Total output
912
Reasoning within output
tag_management f007
100%
Passed
Tokens
76
Input
248
Total output
236
Reasoning within output
tag_management f008
100%
Passed
Tokens
79
Input
1,896
Total output
1,476
Reasoning within output
tag_management f009
100%
Passed
Tokens
83
Input
266
Total output
198
Reasoning within output
tag_management f010
100%
Passed
Tokens
61
Input
134
Total output
126
Reasoning within output
tag_management f011
100%
Passed
Tokens
64
Input
237
Total output
222
Reasoning within output
tag_management f012
100%
Passed
Tokens
99
Input
606
Total output
596
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
167
Input
242
Total output
190
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
174
Input
524
Total output
424
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
254
Input
173
Total output
140
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
252
Input
76
Total output
80
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
157
Input
273
Total output
259
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
160
Input
552
Total output
540
Reasoning within output
worktree_usage f007
0%
Failed
Tokens
168
Input
1,639
Total output
1,386
Reasoning within output
worktree_usage f008
50%
Failed
Tokens
181
Input
1,775
Total output
1,523
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
284
Input
182
Total output
177
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
253
Input
275
Total output
239
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
253
Input
612
Total output
495
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
158
Input
156
Total output
112
Reasoning within output