Google / gemma-4-26b-a4b-it
low
96.6%
197 / 204 fixtures · 1 run(s)
43,808 input / 137,207 total output / 115,867 reasoning within output tokens $0.05773684
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-6.9%
Text: 96.6% →
JSON: 89.7%
+2
Gained
JSON pass / text fail
−16
Lost
Text pass / JSON fail
181
Unchanged Pass
Both pass
5
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| git_show | 91.7% | 75% | -16.7% |
| merge_conflicts | 91.7% | 75% | -16.7% |
| rebase | 91.7% | 75% | -16.7% |
| worktree_usage | 91.7% | 75% | -16.7% |
| blame_forensics | 100% | 91.7% | -8.3% |
| branch_cleanup | 100% | 91.7% | -8.3% |
| git_clean | 100% | 91.7% | -8.3% |
| git_log_format | 100% | 91.7% | -8.3% |
| reflog | 100% | 91.7% | -8.3% |
| submodule_usage | 100% | 91.7% | -8.3% |
| cherry_pick | 83.3% | 83.3% | + 0% |
| commit_messages | 100% | 100% | + 0% |
| commit_squash | 100% | 100% | + 0% |
| git_bisect | 100% | 100% | + 0% |
| git_grep | 91.7% | 91.7% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
| tag_management | 100% | 100% | + 0% |
Changed Fixtures (18)
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
278
Input
466
Total output
461
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
272
Input
693
Total output
580
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
305
Input
335
Total output
228
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
221
Input
641
Total output
635
Reasoning within output
blame_forensics f005
100%
Passed
Tokens
285
Input
1,314
Total output
1,178
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
265
Input
983
Total output
747
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
234
Input
357
Total output
278
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
272
Input
565
Total output
502
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
224
Input
505
Total output
444
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
472
Input
866
Total output
721
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
278
Input
485
Total output
407
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
168
Input
727
Total output
623
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
116
Input
563
Total output
505
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
132
Input
728
Total output
634
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
106
Input
432
Total output
385
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
148
Input
331
Total output
255
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
119
Input
249
Total output
244
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
140
Input
344
Total output
263
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
118
Input
304
Total output
240
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
135
Input
350
Total output
270
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
90
Input
466
Total output
432
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
127
Input
315
Total output
262
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
79
Input
230
Total output
213
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
165
Input
312
Total output
248
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
111
Input
870
Total output
852
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
128
Input
318
Total output
274
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
101
Input
1,144
Total output
1,148
Reasoning within output
cherry_pick f004
0%
Failed
Tokens
181
Input
7,989
Total output
7,099
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
171
Input
537
Total output
501
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
156
Input
404
Total output
307
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
150
Input
258
Total output
247
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
139
Input
172
Total output
145
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
119
Input
248
Total output
232
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
246
Input
1,612
Total output
1,499
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
181
Input
453
Total output
332
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
304
Input
3,028
Total output
2,551
Reasoning within output
commit_messages f001
91.7%
Passed
Tokens
128
Input
268
Total output
263
Reasoning within output
commit_messages f002
87.7%
Passed
Tokens
261
Input
577
Total output
572
Reasoning within output
commit_messages f003
92.7%
Passed
Tokens
99
Input
423
Total output
346
Reasoning within output
commit_messages f004
93.7%
Passed
Tokens
134
Input
262
Total output
228
Reasoning within output
commit_messages f005
87.3%
Passed
Tokens
124
Input
386
Total output
345
Reasoning within output
commit_messages f006
85%
Passed
Tokens
167
Input
999
Total output
833
Reasoning within output
commit_messages f007
91%
Passed
Tokens
225
Input
336
Total output
285
Reasoning within output
commit_messages f008
87.7%
Passed
Tokens
92
Input
299
Total output
243
Reasoning within output
commit_messages f009
81%
Passed
Tokens
149
Input
534
Total output
477
Reasoning within output
commit_messages f010
88.3%
Passed
Tokens
348
Input
497
Total output
435
Reasoning within output
commit_messages f011
66%
Passed
Tokens
169
Input
1,154
Total output
935
Reasoning within output
commit_messages f012
89.7%
Passed
Tokens
146
Input
467
Total output
414
Reasoning within output
commit_squash f001
100%
Passed
Tokens
149
Input
2,389
Total output
1,900
Reasoning within output
commit_squash f002
100%
Passed
Tokens
130
Input
509
Total output
427
Reasoning within output
commit_squash f003
100%
Passed
Tokens
136
Input
696
Total output
682
Reasoning within output
commit_squash f004
100%
Passed
Tokens
129
Input
557
Total output
459
Reasoning within output
commit_squash f005
100%
Passed
Tokens
124
Input
829
Total output
763
Reasoning within output
commit_squash f006
100%
Passed
Tokens
113
Input
785
Total output
691
Reasoning within output
commit_squash f007
100%
Passed
Tokens
113
Input
374
Total output
286
Reasoning within output
commit_squash f008
100%
Passed
Tokens
125
Input
320
Total output
280
Reasoning within output
commit_squash f009
100%
Passed
Tokens
117
Input
4,928
Total output
4,090
Reasoning within output
commit_squash f010
100%
Passed
Tokens
134
Input
335
Total output
267
Reasoning within output
commit_squash f011
100%
Passed
Tokens
139
Input
333
Total output
256
Reasoning within output
commit_squash f012
100%
Passed
Tokens
138
Input
331
Total output
242
Reasoning within output
git_bisect f001
100%
Passed
Tokens
199
Input
214
Total output
206
Reasoning within output
git_bisect f002
100%
Passed
Tokens
222
Input
165
Total output
103
Reasoning within output
git_bisect f003
100%
Passed
Tokens
223
Input
545
Total output
376
Reasoning within output
git_bisect f004
100%
Passed
Tokens
229
Input
284
Total output
276
Reasoning within output
git_bisect f005
100%
Passed
Tokens
228
Input
371
Total output
242
Reasoning within output
git_bisect f006
100%
Passed
Tokens
253
Input
325
Total output
279
Reasoning within output
git_bisect f007
100%
Passed
Tokens
221
Input
415
Total output
297
Reasoning within output
git_bisect f008
100%
Passed
Tokens
229
Input
207
Total output
201
Reasoning within output
git_bisect f009
100%
Passed
Tokens
224
Input
225
Total output
150
Reasoning within output
git_bisect f010
100%
Passed
Tokens
225
Input
209
Total output
169
Reasoning within output
git_bisect f011
100%
Passed
Tokens
227
Input
249
Total output
177
Reasoning within output
git_bisect f012
100%
Passed
Tokens
250
Input
267
Total output
181
Reasoning within output
git_clean f001
100%
Passed
Tokens
66
Input
441
Total output
387
Reasoning within output
git_clean f002
100%
Passed
Tokens
78
Input
208
Total output
191
Reasoning within output
git_clean f003
100%
Passed
Tokens
78
Input
243
Total output
211
Reasoning within output
git_clean f004
100%
Passed
Tokens
78
Input
736
Total output
672
Reasoning within output
git_clean f005
100%
Passed
Tokens
65
Input
303
Total output
267
Reasoning within output
git_clean f006
100%
Passed
Tokens
96
Input
1,076
Total output
972
Reasoning within output
git_clean f007
100%
Passed
Tokens
85
Input
429
Total output
370
Reasoning within output
git_clean f008
100%
Passed
Tokens
92
Input
228
Total output
198
Reasoning within output
git_clean f009
100%
Passed
Tokens
89
Input
614
Total output
527
Reasoning within output
git_clean f010
100%
Passed
Tokens
80
Input
216
Total output
190
Reasoning within output
git_clean f011
100%
Passed
Tokens
80
Input
843
Total output
756
Reasoning within output
git_clean f012
100%
Passed
Tokens
85
Input
500
Total output
492
Reasoning within output
git_grep f001
100%
Passed
Tokens
64
Input
173
Total output
143
Reasoning within output
git_grep f002
100%
Passed
Tokens
77
Input
140
Total output
109
Reasoning within output
git_grep f003
100%
Passed
Tokens
116
Input
151
Total output
118
Reasoning within output
git_grep f004
0%
Failed
Tokens
92
Input
256
Total output
187
Reasoning within output
git_grep f005
100%
Passed
Tokens
130
Input
226
Total output
204
Reasoning within output
git_grep f006
100%
Passed
Tokens
49
Input
1,818
Total output
1,850
Reasoning within output
git_grep f007
100%
Passed
Tokens
192
Input
845
Total output
842
Reasoning within output
git_grep f008
100%
Passed
Tokens
71
Input
175
Total output
153
Reasoning within output
git_grep f009
100%
Passed
Tokens
64
Input
189
Total output
158
Reasoning within output
git_grep f010
100%
Passed
Tokens
79
Input
268
Total output
232
Reasoning within output
git_grep f011
100%
Passed
Tokens
142
Input
450
Total output
448
Reasoning within output
git_grep f012
100%
Passed
Tokens
79
Input
234
Total output
157
Reasoning within output
git_log_format f001
100%
Passed
Tokens
931
Input
612
Total output
365
Reasoning within output
git_log_format f002
100%
Passed
Tokens
926
Input
472
Total output
417
Reasoning within output
git_log_format f003
100%
Passed
Tokens
755
Input
426
Total output
337
Reasoning within output
git_log_format f004
100%
Passed
Tokens
936
Input
415
Total output
290
Reasoning within output
git_log_format f005
100%
Passed
Tokens
762
Input
786
Total output
481
Reasoning within output
git_log_format f006
100%
Passed
Tokens
950
Input
540
Total output
249
Reasoning within output
git_log_format f007
100%
Passed
Tokens
945
Input
322
Total output
213
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,217
Input
170
Total output
127
Reasoning within output
git_log_format f009
100%
Passed
Tokens
750
Input
264
Total output
189
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,449
Input
653
Total output
436
Reasoning within output
git_log_format f011
100%
Passed
Tokens
450
Input
344
Total output
228
Reasoning within output
git_log_format f012
100%
Passed
Tokens
425
Input
306
Total output
211
Reasoning within output
git_show f001
100%
Passed
Tokens
213
Input
237
Total output
170
Reasoning within output
git_show f002
0%
Failed
Tokens
218
Input
1,711
Total output
1,707
Reasoning within output
git_show f003
100%
Passed
Tokens
739
Input
428
Total output
421
Reasoning within output
git_show f004
100%
Passed
Tokens
492
Input
260
Total output
252
Reasoning within output
git_show f005
100%
Passed
Tokens
222
Input
712
Total output
709
Reasoning within output
git_show f006
100%
Passed
Tokens
229
Input
258
Total output
193
Reasoning within output
git_show f007
100%
Passed
Tokens
218
Input
313
Total output
212
Reasoning within output
git_show f008
100%
Passed
Tokens
221
Input
312
Total output
174
Reasoning within output
git_show f009
100%
Passed
Tokens
331
Input
416
Total output
341
Reasoning within output
git_show f010
100%
Passed
Tokens
218
Input
344
Total output
273
Reasoning within output
git_show f011
100%
Passed
Tokens
244
Input
268
Total output
224
Reasoning within output
git_show f012
100%
Passed
Tokens
375
Input
431
Total output
307
Reasoning within output
merge_conflicts f001
100%
Passed
Tokens
98
Input
4,262
Total output
3,992
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
115
Input
811
Total output
690
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
89
Input
378
Total output
360
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
166
Input
1,762
Total output
1,553
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
156
Input
528
Total output
455
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
141
Input
988
Total output
795
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
152
Input
305
Total output
240
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
123
Input
137
Total output
130
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
115
Input
303
Total output
279
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
206
Input
2,717
Total output
2,664
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
159
Input
345
Total output
262
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
263
Input
7,991
Total output
7,179
Reasoning within output
rebase f001
100%
Passed
Tokens
111
Input
3,602
Total output
3,400
Reasoning within output
rebase f002
100%
Passed
Tokens
137
Input
288
Total output
228
Reasoning within output
rebase f003
100%
Passed
Tokens
113
Input
333
Total output
323
Reasoning within output
rebase f004
100%
Passed
Tokens
181
Input
1,387
Total output
1,089
Reasoning within output
rebase f005
100%
Passed
Tokens
162
Input
1,393
Total output
1,239
Reasoning within output
rebase f006
100%
Passed
Tokens
158
Input
1,466
Total output
1,172
Reasoning within output
rebase f007
100%
Passed
Tokens
154
Input
336
Total output
271
Reasoning within output
rebase f008
100%
Passed
Tokens
135
Input
256
Total output
222
Reasoning within output
rebase f009
100%
Passed
Tokens
116
Input
389
Total output
383
Reasoning within output
rebase f010
100%
Passed
Tokens
237
Input
2,489
Total output
2,275
Reasoning within output
rebase f011
100%
Passed
Tokens
169
Input
401
Total output
318
Reasoning within output
rebase f012
0%
Failed
Tokens
295
Input
769
Total output
559
Reasoning within output
reflog f001
100%
Passed
Tokens
269
Input
570
Total output
336
Reasoning within output
reflog f002
100%
Passed
Tokens
250
Input
635
Total output
459
Reasoning within output
reflog f003
100%
Passed
Tokens
217
Input
517
Total output
291
Reasoning within output
reflog f004
100%
Passed
Tokens
371
Input
1,113
Total output
612
Reasoning within output
reflog f005
100%
Passed
Tokens
432
Input
806
Total output
421
Reasoning within output
reflog f006
100%
Passed
Tokens
264
Input
446
Total output
324
Reasoning within output
reflog f007
100%
Passed
Tokens
308
Input
753
Total output
310
Reasoning within output
reflog f008
100%
Passed
Tokens
390
Input
736
Total output
374
Reasoning within output
reflog f009
100%
Passed
Tokens
357
Input
722
Total output
437
Reasoning within output
reflog f010
100%
Passed
Tokens
310
Input
739
Total output
523
Reasoning within output
reflog f011
100%
Passed
Tokens
372
Input
1,662
Total output
944
Reasoning within output
reflog f012
100%
Passed
Tokens
361
Input
777
Total output
356
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
152
Input
238
Total output
204
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
332
Input
221
Total output
165
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
143
Input
171
Total output
149
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
136
Input
154
Total output
124
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
147
Input
188
Total output
154
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
212
Input
246
Total output
195
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
236
Input
184
Total output
146
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
207
Input
179
Total output
147
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
238
Input
183
Total output
162
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
144
Input
182
Total output
164
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
138
Input
248
Total output
212
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
159
Input
225
Total output
194
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
52
Input
135
Total output
119
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
112
Input
794
Total output
800
Reasoning within output
submodule_usage f003
100%
Passed
Tokens
120
Input
679
Total output
653
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
112
Input
208
Total output
224
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
59
Input
660
Total output
584
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
106
Input
536
Total output
553
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
59
Input
242
Total output
202
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
114
Input
238
Total output
213
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
110
Input
197
Total output
162
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
118
Input
588
Total output
591
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
122
Input
160
Total output
151
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
56
Input
163
Total output
136
Reasoning within output
tag_management f001
100%
Passed
Tokens
72
Input
197
Total output
157
Reasoning within output
tag_management f002
100%
Passed
Tokens
69
Input
213
Total output
154
Reasoning within output
tag_management f003
100%
Passed
Tokens
76
Input
465
Total output
388
Reasoning within output
tag_management f004
100%
Passed
Tokens
80
Input
189
Total output
170
Reasoning within output
tag_management f005
100%
Passed
Tokens
95
Input
390
Total output
289
Reasoning within output
tag_management f006
100%
Passed
Tokens
80
Input
1,564
Total output
1,305
Reasoning within output
tag_management f007
100%
Passed
Tokens
75
Input
303
Total output
236
Reasoning within output
tag_management f008
100%
Passed
Tokens
80
Input
794
Total output
615
Reasoning within output
tag_management f009
100%
Passed
Tokens
83
Input
293
Total output
276
Reasoning within output
tag_management f010
100%
Passed
Tokens
59
Input
137
Total output
103
Reasoning within output
tag_management f011
100%
Passed
Tokens
63
Input
172
Total output
161
Reasoning within output
tag_management f012
100%
Passed
Tokens
101
Input
935
Total output
644
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
162
Input
172
Total output
156
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
172
Input
626
Total output
490
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
248
Input
289
Total output
243
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
254
Input
133
Total output
128
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
155
Input
304
Total output
259
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
159
Input
377
Total output
320
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
162
Input
631
Total output
607
Reasoning within output
worktree_usage f008
0%
Failed
Tokens
185
Input
4,419
Total output
3,748
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
281
Input
172
Total output
176
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
254
Input
376
Total output
267
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
254
Input
358
Total output
309
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
155
Input
147
Total output
119
Reasoning within output