Google / gemini-3.5-flash
xhigh
96.1%
196 / 204 fixtures · 1 run(s)
40,641 input / 126,765 total output / 121,820 reasoning within output tokens $1.20184650
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-1%
Text: 96.1% →
JSON: 95.1%
+3
Gained
JSON pass / text fail
−5
Lost
Text pass / JSON fail
191
Unchanged Pass
Both pass
5
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| branch_cleanup | 100% | 75% | -25% |
| rebase | 75% | 91.7% | + 16.7% |
| cherry_pick | 91.7% | 83.3% | -8.3% |
| worktree_usage | 83.3% | 91.7% | + 8.3% |
| commit_squash | 100% | 91.7% | -8.3% |
| blame_forensics | 91.7% | 91.7% | + 0% |
| commit_messages | 100% | 100% | + 0% |
| git_bisect | 100% | 100% | + 0% |
| git_clean | 100% | 100% | + 0% |
| git_grep | 100% | 100% | + 0% |
| git_log_format | 100% | 100% | + 0% |
| git_show | 100% | 100% | + 0% |
| merge_conflicts | 91.7% | 91.7% | + 0% |
| reflog | 100% | 100% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
| submodule_usage | 100% | 100% | + 0% |
| tag_management | 100% | 100% | + 0% |
Changed Fixtures (8)
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
264
Input
783
Total output
779
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
254
Input
565
Total output
561
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
287
Input
338
Total output
332
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
212
Input
388
Total output
383
Reasoning within output
blame_forensics f005
0%
Failed
Tokens
272
Input
3,456
Total output
3,451
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
249
Input
1,174
Total output
1,171
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
215
Input
322
Total output
318
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
258
Input
607
Total output
604
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
212
Input
456
Total output
453
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
462
Input
485
Total output
482
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
264
Input
385
Total output
382
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
154
Input
652
Total output
648
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
101
Input
431
Total output
427
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
116
Input
530
Total output
523
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
94
Input
385
Total output
384
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
129
Input
486
Total output
479
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
104
Input
342
Total output
338
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
123
Input
490
Total output
480
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
101
Input
393
Total output
386
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
120
Input
495
Total output
486
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
76
Input
349
Total output
348
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
114
Input
354
Total output
349
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
65
Input
295
Total output
294
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
157
Input
589
Total output
577
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
92
Input
503
Total output
499
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
110
Input
1,121
Total output
1,114
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
85
Input
782
Total output
779
Reasoning within output
cherry_pick f004
100%
Passed
Tokens
165
Input
2,410
Total output
2,384
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
155
Input
558
Total output
538
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
138
Input
587
Total output
565
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
130
Input
398
Total output
388
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
126
Input
366
Total output
360
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
105
Input
337
Total output
332
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
227
Input
1,534
Total output
1,504
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
164
Input
399
Total output
376
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
282
Input
1,450
Total output
1,401
Reasoning within output
commit_messages f001
89.3%
Passed
Tokens
112
Input
686
Total output
676
Reasoning within output
commit_messages f002
82%
Passed
Tokens
245
Input
645
Total output
632
Reasoning within output
commit_messages f003
94.7%
Passed
Tokens
83
Input
346
Total output
336
Reasoning within output
commit_messages f004
93.7%
Passed
Tokens
118
Input
346
Total output
342
Reasoning within output
commit_messages f005
92%
Passed
Tokens
109
Input
372
Total output
367
Reasoning within output
commit_messages f006
86%
Passed
Tokens
151
Input
714
Total output
708
Reasoning within output
commit_messages f007
91.7%
Passed
Tokens
209
Input
345
Total output
339
Reasoning within output
commit_messages f008
91%
Passed
Tokens
76
Input
380
Total output
375
Reasoning within output
commit_messages f009
89.3%
Passed
Tokens
133
Input
538
Total output
531
Reasoning within output
commit_messages f010
93.3%
Passed
Tokens
332
Input
510
Total output
504
Reasoning within output
commit_messages f011
64.3%
Passed
Tokens
153
Input
2,129
Total output
2,117
Reasoning within output
commit_messages f012
90%
Passed
Tokens
130
Input
365
Total output
360
Reasoning within output
commit_squash f001
100%
Passed
Tokens
133
Input
1,971
Total output
1,958
Reasoning within output
commit_squash f002
100%
Passed
Tokens
116
Input
381
Total output
372
Reasoning within output
commit_squash f003
100%
Passed
Tokens
126
Input
1,133
Total output
1,120
Reasoning within output
commit_squash f004
100%
Passed
Tokens
113
Input
478
Total output
467
Reasoning within output
commit_squash f005
100%
Passed
Tokens
109
Input
837
Total output
830
Reasoning within output
commit_squash f006
100%
Passed
Tokens
96
Input
410
Total output
399
Reasoning within output
commit_squash f007
100%
Passed
Tokens
100
Input
495
Total output
483
Reasoning within output
commit_squash f008
100%
Passed
Tokens
112
Input
512
Total output
500
Reasoning within output
commit_squash f009
100%
Passed
Tokens
104
Input
729
Total output
718
Reasoning within output
commit_squash f010
100%
Passed
Tokens
117
Input
474
Total output
459
Reasoning within output
commit_squash f011
100%
Passed
Tokens
123
Input
546
Total output
531
Reasoning within output
commit_squash f012
100%
Passed
Tokens
124
Input
543
Total output
523
Reasoning within output
git_bisect f001
100%
Passed
Tokens
185
Input
190
Total output
168
Reasoning within output
git_bisect f002
100%
Passed
Tokens
211
Input
236
Total output
230
Reasoning within output
git_bisect f003
100%
Passed
Tokens
203
Input
290
Total output
271
Reasoning within output
git_bisect f004
100%
Passed
Tokens
215
Input
290
Total output
283
Reasoning within output
git_bisect f005
100%
Passed
Tokens
213
Input
170
Total output
165
Reasoning within output
git_bisect f006
100%
Passed
Tokens
241
Input
310
Total output
304
Reasoning within output
git_bisect f007
100%
Passed
Tokens
211
Input
140
Total output
134
Reasoning within output
git_bisect f008
100%
Passed
Tokens
211
Input
234
Total output
227
Reasoning within output
git_bisect f009
100%
Passed
Tokens
201
Input
220
Total output
215
Reasoning within output
git_bisect f010
100%
Passed
Tokens
217
Input
202
Total output
195
Reasoning within output
git_bisect f011
100%
Passed
Tokens
217
Input
336
Total output
329
Reasoning within output
git_bisect f012
100%
Passed
Tokens
235
Input
345
Total output
305
Reasoning within output
git_clean f001
100%
Passed
Tokens
50
Input
423
Total output
419
Reasoning within output
git_clean f002
100%
Passed
Tokens
62
Input
399
Total output
395
Reasoning within output
git_clean f003
100%
Passed
Tokens
62
Input
330
Total output
326
Reasoning within output
git_clean f004
100%
Passed
Tokens
62
Input
609
Total output
604
Reasoning within output
git_clean f005
100%
Passed
Tokens
49
Input
360
Total output
356
Reasoning within output
git_clean f006
100%
Passed
Tokens
80
Input
442
Total output
436
Reasoning within output
git_clean f007
100%
Passed
Tokens
69
Input
509
Total output
500
Reasoning within output
git_clean f008
100%
Passed
Tokens
77
Input
294
Total output
290
Reasoning within output
git_clean f009
100%
Passed
Tokens
73
Input
682
Total output
674
Reasoning within output
git_clean f010
100%
Passed
Tokens
64
Input
584
Total output
580
Reasoning within output
git_clean f011
100%
Passed
Tokens
64
Input
522
Total output
517
Reasoning within output
git_clean f012
100%
Passed
Tokens
69
Input
535
Total output
528
Reasoning within output
git_grep f001
100%
Passed
Tokens
48
Input
284
Total output
279
Reasoning within output
git_grep f002
100%
Passed
Tokens
59
Input
237
Total output
236
Reasoning within output
git_grep f003
100%
Passed
Tokens
100
Input
185
Total output
184
Reasoning within output
git_grep f004
100%
Passed
Tokens
76
Input
429
Total output
424
Reasoning within output
git_grep f005
100%
Passed
Tokens
114
Input
264
Total output
263
Reasoning within output
git_grep f006
100%
Passed
Tokens
33
Input
254
Total output
253
Reasoning within output
git_grep f007
100%
Passed
Tokens
176
Input
858
Total output
856
Reasoning within output
git_grep f008
100%
Passed
Tokens
56
Input
246
Total output
245
Reasoning within output
git_grep f009
100%
Passed
Tokens
49
Input
257
Total output
254
Reasoning within output
git_grep f010
100%
Passed
Tokens
63
Input
242
Total output
241
Reasoning within output
git_grep f011
100%
Passed
Tokens
126
Input
291
Total output
290
Reasoning within output
git_grep f012
100%
Passed
Tokens
64
Input
315
Total output
302
Reasoning within output
git_log_format f001
100%
Passed
Tokens
935
Input
292
Total output
291
Reasoning within output
git_log_format f002
100%
Passed
Tokens
901
Input
217
Total output
210
Reasoning within output
git_log_format f003
100%
Passed
Tokens
745
Input
405
Total output
396
Reasoning within output
git_log_format f004
100%
Passed
Tokens
891
Input
410
Total output
409
Reasoning within output
git_log_format f005
100%
Passed
Tokens
746
Input
546
Total output
541
Reasoning within output
git_log_format f006
100%
Passed
Tokens
934
Input
589
Total output
588
Reasoning within output
git_log_format f007
100%
Passed
Tokens
928
Input
334
Total output
327
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,246
Input
227
Total output
226
Reasoning within output
git_log_format f009
100%
Passed
Tokens
744
Input
211
Total output
210
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,399
Input
380
Total output
379
Reasoning within output
git_log_format f011
100%
Passed
Tokens
447
Input
590
Total output
587
Reasoning within output
git_log_format f012
100%
Passed
Tokens
402
Input
318
Total output
317
Reasoning within output
git_show f001
100%
Passed
Tokens
194
Input
225
Total output
220
Reasoning within output
git_show f002
100%
Passed
Tokens
204
Input
728
Total output
725
Reasoning within output
git_show f003
100%
Passed
Tokens
727
Input
433
Total output
427
Reasoning within output
git_show f004
100%
Passed
Tokens
482
Input
326
Total output
319
Reasoning within output
git_show f005
100%
Passed
Tokens
208
Input
410
Total output
408
Reasoning within output
git_show f006
100%
Passed
Tokens
212
Input
241
Total output
240
Reasoning within output
git_show f007
100%
Passed
Tokens
203
Input
338
Total output
333
Reasoning within output
git_show f008
100%
Passed
Tokens
205
Input
355
Total output
323
Reasoning within output
git_show f009
100%
Passed
Tokens
324
Input
398
Total output
397
Reasoning within output
git_show f010
100%
Passed
Tokens
201
Input
491
Total output
490
Reasoning within output
git_show f011
100%
Passed
Tokens
229
Input
337
Total output
334
Reasoning within output
git_show f012
100%
Passed
Tokens
355
Input
322
Total output
321
Reasoning within output
merge_conflicts f001
100%
Passed
Tokens
82
Input
676
Total output
672
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
100
Input
1,125
Total output
1,118
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
73
Input
978
Total output
975
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
150
Input
1,570
Total output
1,544
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
140
Input
486
Total output
466
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
126
Input
764
Total output
742
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
136
Input
469
Total output
459
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
107
Input
381
Total output
375
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
99
Input
592
Total output
587
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
190
Input
1,191
Total output
1,161
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
143
Input
505
Total output
482
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
247
Input
2,950
Total output
2,903
Reasoning within output
rebase f001
100%
Passed
Tokens
96
Input
690
Total output
686
Reasoning within output
rebase f002
100%
Passed
Tokens
120
Input
486
Total output
479
Reasoning within output
rebase f003
100%
Passed
Tokens
99
Input
420
Total output
417
Reasoning within output
rebase f004
100%
Passed
Tokens
165
Input
4,138
Total output
4,112
Reasoning within output
rebase f005
100%
Passed
Tokens
145
Input
837
Total output
821
Reasoning within output
rebase f006
0%
Failed
Tokens
140
Input
1,001
Total output
980
Reasoning within output
rebase f007
100%
Passed
Tokens
136
Input
803
Total output
793
Reasoning within output
rebase f008
100%
Passed
Tokens
122
Input
423
Total output
417
Reasoning within output
rebase f009
100%
Passed
Tokens
101
Input
380
Total output
375
Reasoning within output
rebase f010
50%
Failed
Tokens
223
Input
2,135
Total output
2,103
Reasoning within output
rebase f011
100%
Passed
Tokens
152
Input
1,032
Total output
1,009
Reasoning within output
rebase f012
0%
Failed
Tokens
282
Input
3,523
Total output
3,474
Reasoning within output
reflog f001
100%
Passed
Tokens
245
Input
761
Total output
725
Reasoning within output
reflog f002
100%
Passed
Tokens
229
Input
853
Total output
705
Reasoning within output
reflog f003
100%
Passed
Tokens
207
Input
661
Total output
523
Reasoning within output
reflog f004
100%
Passed
Tokens
356
Input
1,730
Total output
1,386
Reasoning within output
reflog f005
100%
Passed
Tokens
400
Input
773
Total output
636
Reasoning within output
reflog f006
100%
Passed
Tokens
249
Input
1,064
Total output
809
Reasoning within output
reflog f007
100%
Passed
Tokens
299
Input
864
Total output
643
Reasoning within output
reflog f008
100%
Passed
Tokens
386
Input
1,206
Total output
956
Reasoning within output
reflog f009
100%
Passed
Tokens
341
Input
1,893
Total output
1,626
Reasoning within output
reflog f010
100%
Passed
Tokens
283
Input
1,287
Total output
996
Reasoning within output
reflog f011
100%
Passed
Tokens
356
Input
2,355
Total output
1,902
Reasoning within output
reflog f012
100%
Passed
Tokens
340
Input
1,262
Total output
975
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
134
Input
241
Total output
219
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
316
Input
459
Total output
333
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
127
Input
218
Total output
201
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
120
Input
235
Total output
215
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
131
Input
218
Total output
188
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
196
Input
237
Total output
220
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
216
Input
254
Total output
238
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
192
Input
300
Total output
283
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
222
Input
389
Total output
338
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
128
Input
256
Total output
239
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
121
Input
381
Total output
356
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
143
Input
329
Total output
312
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
36
Input
210
Total output
202
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
101
Input
1,053
Total output
1,046
Reasoning within output
submodule_usage f003
100%
Passed
Tokens
106
Input
781
Total output
758
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
97
Input
194
Total output
191
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
43
Input
854
Total output
825
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
95
Input
696
Total output
693
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
43
Input
392
Total output
375
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
97
Input
570
Total output
565
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
93
Input
266
Total output
263
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
97
Input
656
Total output
651
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
108
Input
243
Total output
239
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
41
Input
285
Total output
274
Reasoning within output
tag_management f001
100%
Passed
Tokens
55
Input
356
Total output
350
Reasoning within output
tag_management f002
100%
Passed
Tokens
52
Input
446
Total output
429
Reasoning within output
tag_management f003
100%
Passed
Tokens
59
Input
302
Total output
292
Reasoning within output
tag_management f004
100%
Passed
Tokens
67
Input
203
Total output
201
Reasoning within output
tag_management f005
100%
Passed
Tokens
80
Input
294
Total output
287
Reasoning within output
tag_management f006
100%
Passed
Tokens
65
Input
356
Total output
342
Reasoning within output
tag_management f007
100%
Passed
Tokens
60
Input
324
Total output
313
Reasoning within output
tag_management f008
100%
Passed
Tokens
63
Input
872
Total output
849
Reasoning within output
tag_management f009
100%
Passed
Tokens
66
Input
315
Total output
299
Reasoning within output
tag_management f010
100%
Passed
Tokens
45
Input
225
Total output
218
Reasoning within output
tag_management f011
100%
Passed
Tokens
45
Input
253
Total output
248
Reasoning within output
tag_management f012
100%
Passed
Tokens
86
Input
917
Total output
908
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
151
Input
378
Total output
369
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
156
Input
504
Total output
481
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
234
Input
330
Total output
322
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
226
Input
796
Total output
789
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
142
Input
426
Total output
413
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
142
Input
543
Total output
530
Reasoning within output
worktree_usage f007
0%
Failed
Tokens
150
Input
1,445
Total output
1,418
Reasoning within output
worktree_usage f008
0%
Failed
Tokens
170
Input
1,258
Total output
1,204
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
269
Input
317
Total output
313
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
243
Input
636
Total output
621
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
234
Input
308
Total output
300
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
140
Input
263
Total output
250
Reasoning within output