Deepseek / deepseek-v4-flash
medium
92.7%
189 / 204 fixtures · 1 run(s)
34,692 input / 61,069 total output / 58,117 reasoning within output tokens $0.01903746
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-20.6%
Text: 92.7% →
JSON: 72.1%
+4
Gained
JSON pass / text fail
−46
Lost
Text pass / JSON fail
143
Unchanged Pass
Both pass
11
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| branch_cleanup | 100% | 58.3% | -41.7% |
| merge_conflicts | 91.7% | 58.3% | -33.3% |
| commit_messages | 100% | 66.7% | -33.3% |
| commit_squash | 83.3% | 58.3% | -25% |
| git_bisect | 100% | 75% | -25% |
| git_clean | 83.3% | 58.3% | -25% |
| git_log_format | 100% | 75% | -25% |
| stash_recovery | 100% | 75% | -25% |
| tag_management | 91.7% | 66.7% | -25% |
| blame_forensics | 100% | 83.3% | -16.7% |
| reflog | 100% | 83.3% | -16.7% |
| rebase | 83.3% | 66.7% | -16.7% |
| git_grep | 91.7% | 83.3% | -8.3% |
| git_show | 91.7% | 83.3% | -8.3% |
| submodule_usage | 91.7% | 83.3% | -8.3% |
| worktree_usage | 91.7% | 83.3% | -8.3% |
| cherry_pick | 75% | 66.7% | -8.3% |
Changed Fixtures (50)
f005 -loss f008 -loss f003 -loss f007 -loss f009 -loss f010 -loss f011 -loss f011 -loss f001 -loss f003 -loss f008 -loss f012 -loss f001 -loss f002 -loss f003 -loss f005 +gain f010 -loss f002 -loss f005 -loss f011 -loss f007 -loss f011 -loss f012 -loss f006 -loss f007 +gain f009 -loss f004 -loss f007 -loss f009 -loss f001 -loss f001 -loss f003 -loss f010 -loss f011 -loss f001 +gain f008 -loss f009 -loss f011 -loss f003 -loss f004 -loss f008 -loss f010 -loss f011 -loss f009 -loss f001 -loss f007 -loss f012 -loss f005 -loss f008 +gain f011 -loss
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
221
Input
948
Total output
1,005
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
198
Input
235
Total output
229
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
231
Input
113
Total output
105
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
160
Input
196
Total output
190
Reasoning within output
blame_forensics f005
100%
Passed
Tokens
225
Input
146
Total output
141
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
202
Input
533
Total output
528
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
187
Input
184
Total output
177
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
207
Input
213
Total output
207
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
174
Input
323
Total output
317
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
363
Input
229
Total output
227
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
206
Input
180
Total output
163
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
126
Input
199
Total output
193
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
98
Input
100
Total output
96
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
110
Input
127
Total output
120
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
91
Input
158
Total output
155
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
124
Input
150
Total output
153
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
99
Input
219
Total output
214
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
114
Input
104
Total output
95
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
101
Input
62
Total output
55
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
117
Input
231
Total output
218
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
75
Input
200
Total output
197
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
115
Input
91
Total output
84
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
69
Input
40
Total output
47
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
152
Input
137
Total output
125
Reasoning within output
cherry_pick f001
0%
Failed
Tokens
92
Input
561
Total output
555
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
119
Input
405
Total output
395
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
87
Input
575
Total output
652
Reasoning within output
cherry_pick f004
0%
Failed
Tokens
147
Input
1,242
Total output
1,221
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
148
Input
91
Total output
88
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
133
Input
394
Total output
394
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
128
Input
144
Total output
133
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
123
Input
102
Total output
111
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
105
Input
67
Total output
71
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
209
Input
261
Total output
270
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
160
Input
229
Total output
206
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
262
Input
615
Total output
602
Reasoning within output
commit_messages f001
93.3%
Passed
Tokens
100
Input
123
Total output
116
Reasoning within output
commit_messages f002
93%
Passed
Tokens
196
Input
259
Total output
252
Reasoning within output
commit_messages f003
95%
Passed
Tokens
81
Input
158
Total output
148
Reasoning within output
commit_messages f004
96%
Passed
Tokens
108
Input
141
Total output
136
Reasoning within output
commit_messages f005
91.7%
Passed
Tokens
96
Input
206
Total output
199
Reasoning within output
commit_messages f006
91%
Passed
Tokens
131
Input
571
Total output
563
Reasoning within output
commit_messages f007
95%
Passed
Tokens
197
Input
105
Total output
97
Reasoning within output
commit_messages f008
90%
Passed
Tokens
66
Input
452
Total output
446
Reasoning within output
commit_messages f009
89.3%
Passed
Tokens
119
Input
241
Total output
232
Reasoning within output
commit_messages f010
93.7%
Passed
Tokens
266
Input
156
Total output
154
Reasoning within output
commit_messages f011
64.3%
Passed
Tokens
137
Input
1,745
Total output
1,730
Reasoning within output
commit_messages f012
95.7%
Passed
Tokens
120
Input
128
Total output
121
Reasoning within output
commit_squash f001
100%
Passed
Tokens
132
Input
665
Total output
651
Reasoning within output
commit_squash f002
100%
Passed
Tokens
110
Input
252
Total output
261
Reasoning within output
commit_squash f003
100%
Passed
Tokens
119
Input
570
Total output
556
Reasoning within output
commit_squash f004
100%
Failed
Tokens
110
Input
163
Total output
146
Reasoning within output
commit_squash f005
0%
Failed
Tokens
107
Input
279
Total output
283
Reasoning within output
commit_squash f006
100%
Passed
Tokens
93
Input
70
Total output
61
Reasoning within output
commit_squash f007
100%
Passed
Tokens
93
Input
109
Total output
96
Reasoning within output
commit_squash f008
100%
Passed
Tokens
105
Input
109
Total output
89
Reasoning within output
commit_squash f009
100%
Passed
Tokens
98
Input
371
Total output
359
Reasoning within output
commit_squash f010
100%
Passed
Tokens
116
Input
134
Total output
112
Reasoning within output
commit_squash f011
100%
Passed
Tokens
123
Input
267
Total output
251
Reasoning within output
commit_squash f012
100%
Passed
Tokens
120
Input
147
Total output
125
Reasoning within output
git_bisect f001
100%
Passed
Tokens
172
Input
97
Total output
89
Reasoning within output
git_bisect f002
100%
Passed
Tokens
195
Input
88
Total output
81
Reasoning within output
git_bisect f003
100%
Passed
Tokens
179
Input
114
Total output
108
Reasoning within output
git_bisect f004
100%
Passed
Tokens
189
Input
85
Total output
83
Reasoning within output
git_bisect f005
100%
Passed
Tokens
183
Input
117
Total output
114
Reasoning within output
git_bisect f006
100%
Passed
Tokens
202
Input
104
Total output
99
Reasoning within output
git_bisect f007
100%
Passed
Tokens
189
Input
128
Total output
121
Reasoning within output
git_bisect f008
100%
Passed
Tokens
191
Input
153
Total output
146
Reasoning within output
git_bisect f009
100%
Passed
Tokens
187
Input
96
Total output
92
Reasoning within output
git_bisect f010
100%
Passed
Tokens
191
Input
186
Total output
179
Reasoning within output
git_bisect f011
100%
Passed
Tokens
189
Input
112
Total output
107
Reasoning within output
git_bisect f012
100%
Passed
Tokens
214
Input
84
Total output
78
Reasoning within output
git_clean f001
100%
Passed
Tokens
49
Input
309
Total output
299
Reasoning within output
git_clean f002
100%
Passed
Tokens
59
Input
109
Total output
103
Reasoning within output
git_clean f003
100%
Passed
Tokens
57
Input
93
Total output
87
Reasoning within output
git_clean f004
66.7%
Failed
Tokens
57
Input
119
Total output
112
Reasoning within output
git_clean f005
50%
Failed
Tokens
47
Input
223
Total output
213
Reasoning within output
git_clean f006
100%
Passed
Tokens
74
Input
441
Total output
433
Reasoning within output
git_clean f007
100%
Passed
Tokens
69
Input
300
Total output
288
Reasoning within output
git_clean f008
100%
Passed
Tokens
72
Input
244
Total output
239
Reasoning within output
git_clean f009
100%
Passed
Tokens
67
Input
178
Total output
188
Reasoning within output
git_clean f010
100%
Passed
Tokens
63
Input
279
Total output
275
Reasoning within output
git_clean f011
100%
Passed
Tokens
60
Input
375
Total output
350
Reasoning within output
git_clean f012
100%
Passed
Tokens
63
Input
438
Total output
447
Reasoning within output
git_grep f001
100%
Passed
Tokens
52
Input
106
Total output
109
Reasoning within output
git_grep f002
100%
Passed
Tokens
61
Input
60
Total output
57
Reasoning within output
git_grep f003
100%
Passed
Tokens
84
Input
108
Total output
115
Reasoning within output
git_grep f004
100%
Passed
Tokens
72
Input
122
Total output
115
Reasoning within output
git_grep f005
100%
Passed
Tokens
107
Input
144
Total output
143
Reasoning within output
git_grep f006
100%
Passed
Tokens
37
Input
203
Total output
222
Reasoning within output
git_grep f007
0%
Failed
Tokens
156
Input
4,043
Total output
4,040
Reasoning within output
git_grep f008
100%
Passed
Tokens
57
Input
58
Total output
55
Reasoning within output
git_grep f009
100%
Passed
Tokens
50
Input
53
Total output
48
Reasoning within output
git_grep f010
100%
Passed
Tokens
60
Input
350
Total output
347
Reasoning within output
git_grep f011
100%
Passed
Tokens
114
Input
150
Total output
140
Reasoning within output
git_grep f012
100%
Passed
Tokens
64
Input
102
Total output
91
Reasoning within output
git_log_format f001
100%
Passed
Tokens
731
Input
60
Total output
50
Reasoning within output
git_log_format f002
100%
Passed
Tokens
702
Input
113
Total output
106
Reasoning within output
git_log_format f003
100%
Passed
Tokens
567
Input
68
Total output
57
Reasoning within output
git_log_format f004
100%
Passed
Tokens
719
Input
56
Total output
53
Reasoning within output
git_log_format f005
100%
Passed
Tokens
573
Input
153
Total output
125
Reasoning within output
git_log_format f006
100%
Passed
Tokens
703
Input
96
Total output
93
Reasoning within output
git_log_format f007
100%
Passed
Tokens
704
Input
50
Total output
46
Reasoning within output
git_log_format f008
100%
Passed
Tokens
958
Input
37
Total output
36
Reasoning within output
git_log_format f009
100%
Passed
Tokens
549
Input
122
Total output
119
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,092
Input
60
Total output
57
Reasoning within output
git_log_format f011
100%
Passed
Tokens
346
Input
101
Total output
94
Reasoning within output
git_log_format f012
100%
Passed
Tokens
330
Input
70
Total output
68
Reasoning within output
git_show f001
100%
Passed
Tokens
172
Input
69
Total output
62
Reasoning within output
git_show f002
0%
Failed
Tokens
166
Input
1,093
Total output
1,090
Reasoning within output
git_show f003
100%
Passed
Tokens
565
Input
122
Total output
96
Reasoning within output
git_show f004
100%
Passed
Tokens
388
Input
71
Total output
63
Reasoning within output
git_show f005
100%
Passed
Tokens
176
Input
137
Total output
134
Reasoning within output
git_show f006
100%
Passed
Tokens
180
Input
79
Total output
76
Reasoning within output
git_show f007
100%
Passed
Tokens
162
Input
71
Total output
66
Reasoning within output
git_show f008
100%
Passed
Tokens
178
Input
114
Total output
79
Reasoning within output
git_show f009
100%
Passed
Tokens
260
Input
128
Total output
100
Reasoning within output
git_show f010
100%
Passed
Tokens
169
Input
54
Total output
51
Reasoning within output
git_show f011
100%
Passed
Tokens
199
Input
84
Total output
79
Reasoning within output
git_show f012
100%
Passed
Tokens
295
Input
88
Total output
85
Reasoning within output
merge_conflicts f001
100%
Passed
Tokens
84
Input
665
Total output
659
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
109
Input
313
Total output
282
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
79
Input
349
Total output
344
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
134
Input
506
Total output
482
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
136
Input
121
Total output
109
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
123
Input
352
Total output
331
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
130
Input
348
Total output
337
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
106
Input
109
Total output
101
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
104
Input
241
Total output
261
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
178
Input
574
Total output
548
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
142
Input
183
Total output
134
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
231
Input
485
Total output
443
Reasoning within output
rebase f001
0%
Failed
Tokens
95
Input
774
Total output
834
Reasoning within output
rebase f002
100%
Passed
Tokens
127
Input
100
Total output
71
Reasoning within output
rebase f003
100%
Passed
Tokens
100
Input
257
Total output
277
Reasoning within output
rebase f004
100%
Passed
Tokens
144
Input
1,533
Total output
1,509
Reasoning within output
rebase f005
100%
Passed
Tokens
140
Input
160
Total output
147
Reasoning within output
rebase f006
100%
Passed
Tokens
135
Input
361
Total output
340
Reasoning within output
rebase f007
100%
Passed
Tokens
131
Input
211
Total output
199
Reasoning within output
rebase f008
100%
Passed
Tokens
118
Input
67
Total output
64
Reasoning within output
rebase f009
100%
Passed
Tokens
102
Input
165
Total output
158
Reasoning within output
rebase f010
100%
Passed
Tokens
201
Input
642
Total output
618
Reasoning within output
rebase f011
100%
Passed
Tokens
152
Input
275
Total output
252
Reasoning within output
rebase f012
0%
Failed
Tokens
258
Input
2,372
Total output
2,330
Reasoning within output
reflog f001
100%
Passed
Tokens
199
Input
154
Total output
109
Reasoning within output
reflog f002
100%
Passed
Tokens
197
Input
99
Total output
49
Reasoning within output
reflog f003
100%
Passed
Tokens
175
Input
197
Total output
141
Reasoning within output
reflog f004
100%
Passed
Tokens
277
Input
416
Total output
241
Reasoning within output
reflog f005
100%
Passed
Tokens
329
Input
4,546
Total output
4,488
Reasoning within output
reflog f006
100%
Passed
Tokens
202
Input
132
Total output
78
Reasoning within output
reflog f007
100%
Passed
Tokens
235
Input
534
Total output
473
Reasoning within output
reflog f008
100%
Passed
Tokens
305
Input
738
Total output
654
Reasoning within output
reflog f009
100%
Passed
Tokens
254
Input
677
Total output
557
Reasoning within output
reflog f010
100%
Passed
Tokens
222
Input
268
Total output
229
Reasoning within output
reflog f011
100%
Passed
Tokens
279
Input
1,733
Total output
1,582
Reasoning within output
reflog f012
100%
Passed
Tokens
258
Input
1,421
Total output
989
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
120
Input
82
Total output
76
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
269
Input
196
Total output
178
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
111
Input
101
Total output
91
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
110
Input
127
Total output
115
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
120
Input
105
Total output
86
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
167
Input
95
Total output
82
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
192
Input
142
Total output
125
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
163
Input
172
Total output
143
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
198
Input
98
Total output
79
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
118
Input
113
Total output
92
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
108
Input
249
Total output
243
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
127
Input
118
Total output
84
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
40
Input
106
Total output
96
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
87
Input
169
Total output
143
Reasoning within output
submodule_usage f003
75%
Failed
Tokens
90
Input
365
Total output
382
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
81
Input
49
Total output
43
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
47
Input
144
Total output
119
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
77
Input
175
Total output
170
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
48
Input
112
Total output
91
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
85
Input
888
Total output
879
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
82
Input
57
Total output
51
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
94
Input
157
Total output
163
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
96
Input
72
Total output
66
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
46
Input
87
Total output
76
Reasoning within output
tag_management f001
100%
Passed
Tokens
53
Input
75
Total output
68
Reasoning within output
tag_management f002
100%
Passed
Tokens
54
Input
94
Total output
74
Reasoning within output
tag_management f003
100%
Passed
Tokens
55
Input
81
Total output
66
Reasoning within output
tag_management f004
100%
Passed
Tokens
58
Input
98
Total output
94
Reasoning within output
tag_management f005
100%
Passed
Tokens
74
Input
131
Total output
121
Reasoning within output
tag_management f006
100%
Passed
Tokens
59
Input
100
Total output
88
Reasoning within output
tag_management f007
100%
Passed
Tokens
60
Input
155
Total output
142
Reasoning within output
tag_management f008
100%
Passed
Tokens
63
Input
388
Total output
363
Reasoning within output
tag_management f009
100%
Passed
Tokens
67
Input
163
Total output
145
Reasoning within output
tag_management f010
100%
Passed
Tokens
46
Input
49
Total output
40
Reasoning within output
tag_management f011
0%
Failed
Tokens
45
Input
98
Total output
91
Reasoning within output
tag_management f012
100%
Passed
Tokens
75
Input
280
Total output
269
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
126
Input
147
Total output
138
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
131
Input
299
Total output
298
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
202
Input
236
Total output
226
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
185
Input
55
Total output
49
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
122
Input
257
Total output
242
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
120
Input
118
Total output
105
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
120
Input
628
Total output
609
Reasoning within output
worktree_usage f008
0%
Failed
Tokens
143
Input
664
Total output
621
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
217
Input
206
Total output
199
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
191
Input
90
Total output
72
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
190
Input
207
Total output
197
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
132
Input
101
Total output
86
Reasoning within output