Deepseek / deepseek-v4-pro
low
92.7%
189 / 204 fixtures · 1 run(s)
34,737 input / 68,770 total output / 64,685 reasoning within output tokens $0.22237005
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-15.2%
Text: 92.7% →
JSON: 77.5%
+8
Gained
JSON pass / text fail
−39
Lost
Text pass / JSON fail
150
Unchanged Pass
Both pass
7
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| commit_squash | 100% | 58.3% | -41.7% |
| git_grep | 100% | 58.3% | -41.7% |
| blame_forensics | 91.7% | 58.3% | -33.3% |
| branch_cleanup | 100% | 66.7% | -33.3% |
| merge_conflicts | 83.3% | 50% | -33.3% |
| git_log_format | 100% | 75% | -25% |
| cherry_pick | 83.3% | 66.7% | -16.7% |
| worktree_usage | 83.3% | 66.7% | -16.7% |
| git_clean | 91.7% | 83.3% | -8.3% |
| git_bisect | 100% | 91.7% | -8.3% |
| git_show | 100% | 91.7% | -8.3% |
| rebase | 66.7% | 75% | + 8.3% |
| commit_messages | 100% | 100% | + 0% |
| reflog | 100% | 100% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
| submodule_usage | 91.7% | 91.7% | + 0% |
| tag_management | 83.3% | 83.3% | + 0% |
Changed Fixtures (47)
f002 -loss f007 -loss f011 -loss f012 -loss f003 -loss f004 -loss f009 -loss f011 -loss f006 -loss f010 -loss f001 -loss f002 -loss f003 -loss f004 -loss f008 -loss f012 -loss f004 +gain f005 -loss f012 -loss f003 -loss f006 -loss f007 -loss f009 -loss f011 -loss f003 -loss f010 -loss f012 -loss f007 -loss f001 -loss f002 -loss f003 -loss f007 +gain f010 -loss f011 -loss f001 -loss f004 +gain f011 +gain f003 +gain f011 -loss f002 +gain f004 +gain f007 -loss f008 -loss f001 -loss f002 -loss f009 -loss f010 +gain
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
218
Input
510
Total output
505
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
192
Input
461
Total output
460
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
231
Input
316
Total output
308
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
161
Input
6
Total output
0
Reasoning within output
blame_forensics f005
0%
Failed
Tokens
231
Input
5,357
Total output
5,349
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
214
Input
527
Total output
523
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
183
Input
124
Total output
109
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
200
Input
341
Total output
336
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
174
Input
231
Total output
227
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
365
Input
304
Total output
300
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
209
Input
413
Total output
391
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
126
Input
316
Total output
310
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
98
Input
238
Total output
232
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
115
Input
387
Total output
380
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
89
Input
158
Total output
156
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
120
Input
254
Total output
262
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
105
Input
302
Total output
298
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
114
Input
235
Total output
226
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
100
Input
310
Total output
304
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
119
Input
326
Total output
314
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
82
Input
325
Total output
322
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
112
Input
185
Total output
178
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
69
Input
116
Total output
113
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
153
Input
13
Total output
0
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
92
Input
178
Total output
173
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
120
Input
563
Total output
536
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
87
Input
665
Total output
659
Reasoning within output
cherry_pick f004
0%
Failed
Tokens
147
Input
901
Total output
880
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
148
Input
185
Total output
167
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
136
Input
299
Total output
263
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
129
Input
278
Total output
267
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
121
Input
232
Total output
225
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
105
Input
131
Total output
125
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
213
Input
500
Total output
475
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
161
Input
200
Total output
178
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
264
Input
1,021
Total output
1,011
Reasoning within output
commit_messages f001
96.3%
Passed
Tokens
100
Input
228
Total output
224
Reasoning within output
commit_messages f002
89%
Passed
Tokens
196
Input
205
Total output
196
Reasoning within output
commit_messages f003
93.3%
Passed
Tokens
81
Input
80
Total output
69
Reasoning within output
commit_messages f004
93.3%
Passed
Tokens
108
Input
217
Total output
213
Reasoning within output
commit_messages f005
88.3%
Passed
Tokens
96
Input
177
Total output
181
Reasoning within output
commit_messages f006
83.3%
Passed
Tokens
131
Input
232
Total output
225
Reasoning within output
commit_messages f007
90%
Passed
Tokens
180
Input
353
Total output
345
Reasoning within output
commit_messages f008
92.7%
Passed
Tokens
66
Input
214
Total output
215
Reasoning within output
commit_messages f009
88.3%
Passed
Tokens
119
Input
229
Total output
221
Reasoning within output
commit_messages f010
90%
Passed
Tokens
266
Input
394
Total output
385
Reasoning within output
commit_messages f011
62.3%
Passed
Tokens
137
Input
1,003
Total output
703
Reasoning within output
commit_messages f012
89.7%
Passed
Tokens
120
Input
208
Total output
227
Reasoning within output
commit_squash f001
100%
Passed
Tokens
129
Input
851
Total output
867
Reasoning within output
commit_squash f002
100%
Passed
Tokens
116
Input
244
Total output
234
Reasoning within output
commit_squash f003
100%
Passed
Tokens
123
Input
255
Total output
231
Reasoning within output
commit_squash f004
100%
Passed
Tokens
111
Input
285
Total output
272
Reasoning within output
commit_squash f005
100%
Passed
Tokens
106
Input
393
Total output
385
Reasoning within output
commit_squash f006
100%
Passed
Tokens
95
Input
202
Total output
189
Reasoning within output
commit_squash f007
100%
Passed
Tokens
95
Input
178
Total output
164
Reasoning within output
commit_squash f008
100%
Passed
Tokens
108
Input
207
Total output
193
Reasoning within output
commit_squash f009
100%
Passed
Tokens
98
Input
175
Total output
162
Reasoning within output
commit_squash f010
100%
Passed
Tokens
116
Input
424
Total output
408
Reasoning within output
commit_squash f011
100%
Passed
Tokens
123
Input
313
Total output
297
Reasoning within output
commit_squash f012
100%
Passed
Tokens
119
Input
386
Total output
364
Reasoning within output
git_bisect f001
100%
Passed
Tokens
164
Input
211
Total output
206
Reasoning within output
git_bisect f002
100%
Passed
Tokens
193
Input
273
Total output
266
Reasoning within output
git_bisect f003
100%
Passed
Tokens
191
Input
272
Total output
266
Reasoning within output
git_bisect f004
100%
Passed
Tokens
183
Input
146
Total output
140
Reasoning within output
git_bisect f005
100%
Passed
Tokens
187
Input
245
Total output
227
Reasoning within output
git_bisect f006
100%
Passed
Tokens
216
Input
144
Total output
137
Reasoning within output
git_bisect f007
100%
Passed
Tokens
189
Input
165
Total output
159
Reasoning within output
git_bisect f008
100%
Passed
Tokens
189
Input
64
Total output
0
Reasoning within output
git_bisect f009
100%
Passed
Tokens
185
Input
176
Total output
170
Reasoning within output
git_bisect f010
100%
Passed
Tokens
185
Input
70
Total output
0
Reasoning within output
git_bisect f011
100%
Passed
Tokens
193
Input
305
Total output
297
Reasoning within output
git_bisect f012
100%
Passed
Tokens
210
Input
261
Total output
255
Reasoning within output
git_clean f001
100%
Passed
Tokens
49
Input
451
Total output
441
Reasoning within output
git_clean f002
100%
Passed
Tokens
59
Input
119
Total output
114
Reasoning within output
git_clean f003
100%
Passed
Tokens
57
Input
210
Total output
205
Reasoning within output
git_clean f004
33.3%
Failed
Tokens
57
Input
461
Total output
455
Reasoning within output
git_clean f005
100%
Passed
Tokens
47
Input
329
Total output
321
Reasoning within output
git_clean f006
100%
Passed
Tokens
74
Input
382
Total output
374
Reasoning within output
git_clean f007
100%
Passed
Tokens
69
Input
181
Total output
170
Reasoning within output
git_clean f008
100%
Passed
Tokens
72
Input
71
Total output
66
Reasoning within output
git_clean f009
100%
Passed
Tokens
67
Input
348
Total output
338
Reasoning within output
git_clean f010
100%
Passed
Tokens
63
Input
5
Total output
0
Reasoning within output
git_clean f011
100%
Passed
Tokens
60
Input
368
Total output
370
Reasoning within output
git_clean f012
100%
Passed
Tokens
63
Input
8
Total output
0
Reasoning within output
git_grep f001
100%
Passed
Tokens
52
Input
127
Total output
122
Reasoning within output
git_grep f002
100%
Passed
Tokens
61
Input
92
Total output
90
Reasoning within output
git_grep f003
100%
Passed
Tokens
84
Input
78
Total output
76
Reasoning within output
git_grep f004
100%
Passed
Tokens
72
Input
6
Total output
0
Reasoning within output
git_grep f005
100%
Passed
Tokens
107
Input
169
Total output
166
Reasoning within output
git_grep f006
100%
Passed
Tokens
37
Input
133
Total output
130
Reasoning within output
git_grep f007
100%
Passed
Tokens
156
Input
547
Total output
544
Reasoning within output
git_grep f008
100%
Passed
Tokens
57
Input
97
Total output
94
Reasoning within output
git_grep f009
100%
Passed
Tokens
50
Input
82
Total output
78
Reasoning within output
git_grep f010
100%
Passed
Tokens
60
Input
284
Total output
281
Reasoning within output
git_grep f011
100%
Passed
Tokens
114
Input
108
Total output
99
Reasoning within output
git_grep f012
100%
Passed
Tokens
64
Input
128
Total output
117
Reasoning within output
git_log_format f001
100%
Passed
Tokens
714
Input
2
Total output
0
Reasoning within output
git_log_format f002
100%
Passed
Tokens
716
Input
147
Total output
138
Reasoning within output
git_log_format f003
100%
Passed
Tokens
573
Input
123
Total output
121
Reasoning within output
git_log_format f004
100%
Passed
Tokens
690
Input
133
Total output
130
Reasoning within output
git_log_format f005
100%
Passed
Tokens
573
Input
140
Total output
133
Reasoning within output
git_log_format f006
100%
Passed
Tokens
703
Input
348
Total output
345
Reasoning within output
git_log_format f007
100%
Passed
Tokens
719
Input
51
Total output
45
Reasoning within output
git_log_format f008
100%
Passed
Tokens
944
Input
37
Total output
34
Reasoning within output
git_log_format f009
100%
Passed
Tokens
565
Input
114
Total output
111
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,071
Input
2
Total output
0
Reasoning within output
git_log_format f011
100%
Passed
Tokens
355
Input
3
Total output
0
Reasoning within output
git_log_format f012
100%
Passed
Tokens
318
Input
120
Total output
107
Reasoning within output
git_show f001
100%
Passed
Tokens
170
Input
6
Total output
0
Reasoning within output
git_show f002
100%
Passed
Tokens
168
Input
3
Total output
0
Reasoning within output
git_show f003
100%
Passed
Tokens
559
Input
72
Total output
64
Reasoning within output
git_show f004
100%
Passed
Tokens
390
Input
176
Total output
148
Reasoning within output
git_show f005
100%
Passed
Tokens
172
Input
3
Total output
0
Reasoning within output
git_show f006
100%
Passed
Tokens
183
Input
33
Total output
30
Reasoning within output
git_show f007
100%
Passed
Tokens
162
Input
97
Total output
91
Reasoning within output
git_show f008
100%
Passed
Tokens
176
Input
110
Total output
87
Reasoning within output
git_show f009
100%
Passed
Tokens
257
Input
144
Total output
142
Reasoning within output
git_show f010
100%
Passed
Tokens
171
Input
257
Total output
255
Reasoning within output
git_show f011
100%
Passed
Tokens
197
Input
95
Total output
99
Reasoning within output
git_show f012
100%
Passed
Tokens
295
Input
147
Total output
142
Reasoning within output
merge_conflicts f001
100%
Passed
Tokens
84
Input
971
Total output
965
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
109
Input
593
Total output
557
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
79
Input
643
Total output
704
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
134
Input
1,436
Total output
1,412
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
136
Input
169
Total output
151
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
123
Input
341
Total output
320
Reasoning within output
merge_conflicts f007
0%
Failed
Tokens
130
Input
54
Total output
0
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
106
Input
218
Total output
210
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
104
Input
196
Total output
189
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
178
Input
573
Total output
548
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
142
Input
310
Total output
287
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
231
Input
860
Total output
795
Reasoning within output
rebase f001
100%
Passed
Tokens
93
Input
517
Total output
512
Reasoning within output
rebase f002
100%
Passed
Tokens
127
Input
175
Total output
165
Reasoning within output
rebase f003
100%
Passed
Tokens
104
Input
93
Total output
89
Reasoning within output
rebase f004
0%
Failed
Tokens
144
Input
1,845
Total output
1,931
Reasoning within output
rebase f005
100%
Passed
Tokens
140
Input
428
Total output
414
Reasoning within output
rebase f006
100%
Passed
Tokens
134
Input
828
Total output
807
Reasoning within output
rebase f007
100%
Passed
Tokens
131
Input
279
Total output
267
Reasoning within output
rebase f008
100%
Passed
Tokens
118
Input
7
Total output
0
Reasoning within output
rebase f009
100%
Passed
Tokens
103
Input
263
Total output
265
Reasoning within output
rebase f010
50%
Failed
Tokens
201
Input
2,021
Total output
2,242
Reasoning within output
rebase f011
0%
Failed
Tokens
152
Input
230
Total output
198
Reasoning within output
rebase f012
0%
Failed
Tokens
260
Input
1,186
Total output
1,145
Reasoning within output
reflog f001
100%
Passed
Tokens
197
Input
390
Total output
363
Reasoning within output
reflog f002
100%
Passed
Tokens
199
Input
125
Total output
0
Reasoning within output
reflog f003
100%
Passed
Tokens
175
Input
267
Total output
203
Reasoning within output
reflog f004
100%
Passed
Tokens
281
Input
1,232
Total output
842
Reasoning within output
reflog f005
100%
Passed
Tokens
324
Input
3,022
Total output
2,972
Reasoning within output
reflog f006
100%
Passed
Tokens
220
Input
267
Total output
135
Reasoning within output
reflog f007
100%
Passed
Tokens
253
Input
88
Total output
0
Reasoning within output
reflog f008
100%
Passed
Tokens
323
Input
1,298
Total output
1,185
Reasoning within output
reflog f009
100%
Passed
Tokens
260
Input
1,444
Total output
1,207
Reasoning within output
reflog f010
100%
Passed
Tokens
225
Input
455
Total output
315
Reasoning within output
reflog f011
100%
Passed
Tokens
268
Input
920
Total output
803
Reasoning within output
reflog f012
100%
Passed
Tokens
288
Input
365
Total output
0
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
122
Input
206
Total output
199
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
263
Input
331
Total output
226
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
111
Input
116
Total output
109
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
110
Input
236
Total output
219
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
119
Input
249
Total output
230
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
167
Input
94
Total output
87
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
194
Input
169
Total output
153
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
163
Input
68
Total output
0
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
198
Input
168
Total output
147
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
118
Input
80
Total output
73
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
107
Input
404
Total output
383
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
127
Input
125
Total output
109
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
40
Input
10
Total output
0
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
95
Input
351
Total output
338
Reasoning within output
submodule_usage f003
75%
Failed
Tokens
96
Input
940
Total output
901
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
84
Input
88
Total output
82
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
47
Input
347
Total output
332
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
75
Input
583
Total output
577
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
48
Input
217
Total output
197
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
88
Input
9
Total output
0
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
85
Input
165
Total output
160
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
90
Input
370
Total output
361
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
89
Input
116
Total output
111
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
46
Input
282
Total output
268
Reasoning within output
tag_management f001
100%
Passed
Tokens
54
Input
70
Total output
62
Reasoning within output
tag_management f002
0%
Failed
Tokens
51
Input
94
Total output
69
Reasoning within output
tag_management f003
100%
Passed
Tokens
55
Input
164
Total output
135
Reasoning within output
tag_management f004
75%
Failed
Tokens
63
Input
170
Total output
160
Reasoning within output
tag_management f005
100%
Passed
Tokens
74
Input
139
Total output
129
Reasoning within output
tag_management f006
100%
Passed
Tokens
55
Input
19
Total output
0
Reasoning within output
tag_management f007
100%
Passed
Tokens
61
Input
615
Total output
603
Reasoning within output
tag_management f008
100%
Passed
Tokens
67
Input
357
Total output
332
Reasoning within output
tag_management f009
100%
Passed
Tokens
67
Input
194
Total output
171
Reasoning within output
tag_management f010
100%
Passed
Tokens
45
Input
73
Total output
65
Reasoning within output
tag_management f011
100%
Passed
Tokens
44
Input
6
Total output
0
Reasoning within output
tag_management f012
100%
Passed
Tokens
76
Input
211
Total output
207
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
124
Input
364
Total output
353
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
134
Input
440
Total output
417
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
179
Input
131
Total output
122
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
185
Input
209
Total output
204
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
119
Input
69
Total output
54
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
118
Input
305
Total output
288
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
126
Input
381
Total output
360
Reasoning within output
worktree_usage f008
0%
Failed
Tokens
143
Input
603
Total output
550
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
221
Input
93
Total output
87
Reasoning within output
worktree_usage f010
0%
Failed
Tokens
191
Input
302
Total output
264
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
189
Input
9
Total output
0
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
119
Input
79
Total output
65
Reasoning within output