Minimax / minimax-m3
medium
94.1%
192 / 204 fixtures · 1 run(s)
69,390 input / 45,078 total output / 37,547 reasoning within output tokens $0.07603296
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-10.8%
Text: 94.1% →
JSON: 83.3%
+3
Gained
JSON pass / text fail
−25
Lost
Text pass / JSON fail
167
Unchanged Pass
Both pass
9
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|---|---|---|
| f004 | 100% (1/1) | 0% (0/1) | +100% |
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| commit_squash | 100% | 25% | -75% |
| branch_cleanup | 100% | 50% | -50% |
| git_grep | 91.7% | 75% | -16.7% |
| git_clean | 83.3% | 66.7% | -16.7% |
| git_show | 91.7% | 83.3% | -8.3% |
| commit_messages | 100% | 91.7% | -8.3% |
| reflog | 100% | 91.7% | -8.3% |
| submodule_usage | 100% | 91.7% | -8.3% |
| tag_management | 91.7% | 100% | + 8.3% |
| blame_forensics | 100% | 100% | + 0% |
| cherry_pick | 83.3% | 83.3% | + 0% |
| git_bisect | 100% | 100% | + 0% |
| git_log_format | 100% | 100% | + 0% |
| merge_conflicts | 83.3% | 83.3% | + 0% |
| rebase | 83.3% | 83.3% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
| worktree_usage | 91.7% | 91.7% | + 0% |
Changed Fixtures (28)
f002 -loss f003 -loss f004 -loss f009 -loss f010 -loss f011 -loss f001 -loss f011 +gain f004 -loss f001 -loss f002 -loss f003 -loss f004 -loss f005 -loss f006 -loss f008 -loss f009 -loss f010 -loss f005 -loss f007 -loss f011 +gain f012 -loss f006 -loss f011 -loss f009 -loss f004 -loss f005 -loss f011 +gain
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
390
Input
354
Total output
397
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
367
Input
235
Total output
224
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
397
Input
159
Total output
141
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
338
Input
160
Total output
192
Reasoning within output
blame_forensics f005
100%
Passed
Tokens
398
Input
93
Total output
79
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
372
Input
285
Total output
279
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
352
Input
156
Total output
148
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
376
Input
526
Total output
545
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
346
Input
256
Total output
259
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
533
Input
190
Total output
197
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
375
Input
122
Total output
107
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
299
Input
207
Total output
198
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
271
Input
472
Total output
484
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
284
Input
260
Total output
269
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
262
Input
158
Total output
166
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
294
Input
224
Total output
247
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
270
Input
189
Total output
188
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
287
Input
103
Total output
106
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
273
Input
89
Total output
101
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
287
Input
168
Total output
155
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
249
Input
638
Total output
751
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
280
Input
162
Total output
157
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
238
Input
104
Total output
125
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
317
Input
183
Total output
164
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
261
Input
207
Total output
248
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
283
Input
633
Total output
644
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
256
Input
404
Total output
470
Reasoning within output
cherry_pick f004
100%
Passed
Tokens
312
Input
489
Total output
509
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
315
Input
189
Total output
185
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
299
Input
505
Total output
504
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
292
Input
171
Total output
172
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
292
Input
69
Total output
64
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
275
Input
98
Total output
99
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
378
Input
218
Total output
202
Reasoning within output
cherry_pick f011
0%
Failed
Tokens
330
Input
236
Total output
204
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
427
Input
566
Total output
576
Reasoning within output
commit_messages f001
86.7%
Passed
Tokens
271
Input
116
Total output
114
Reasoning within output
commit_messages f002
61.3%
Passed
Tokens
365
Input
50
Total output
46
Reasoning within output
commit_messages f003
93.3%
Passed
Tokens
199
Input
53
Total output
43
Reasoning within output
commit_messages f004
91%
Passed
Tokens
272
Input
30
Total output
32
Reasoning within output
commit_messages f005
87.7%
Passed
Tokens
267
Input
129
Total output
134
Reasoning within output
commit_messages f006
96%
Passed
Tokens
301
Input
142
Total output
154
Reasoning within output
commit_messages f007
91.3%
Passed
Tokens
298
Input
153
Total output
145
Reasoning within output
commit_messages f008
91.7%
Passed
Tokens
238
Input
49
Total output
46
Reasoning within output
commit_messages f009
87.7%
Passed
Tokens
290
Input
105
Total output
120
Reasoning within output
commit_messages f010
90%
Passed
Tokens
426
Input
153
Total output
157
Reasoning within output
commit_messages f011
80%
Passed
Tokens
502
Input
104
Total output
102
Reasoning within output
commit_messages f012
93.3%
Passed
Tokens
235
Input
119
Total output
112
Reasoning within output
commit_squash f001
100%
Passed
Tokens
301
Input
287
Total output
263
Reasoning within output
commit_squash f002
100%
Passed
Tokens
281
Input
159
Total output
146
Reasoning within output
commit_squash f003
100%
Passed
Tokens
292
Input
141
Total output
114
Reasoning within output
commit_squash f004
100%
Passed
Tokens
230
Input
510
Total output
497
Reasoning within output
commit_squash f005
100%
Passed
Tokens
279
Input
159
Total output
170
Reasoning within output
commit_squash f006
100%
Passed
Tokens
262
Input
168
Total output
173
Reasoning within output
commit_squash f007
100%
Passed
Tokens
265
Input
164
Total output
146
Reasoning within output
commit_squash f008
100%
Passed
Tokens
278
Input
130
Total output
120
Reasoning within output
commit_squash f009
100%
Passed
Tokens
272
Input
171
Total output
144
Reasoning within output
commit_squash f010
100%
Passed
Tokens
289
Input
131
Total output
108
Reasoning within output
commit_squash f011
100%
Passed
Tokens
295
Input
175
Total output
134
Reasoning within output
commit_squash f012
100%
Passed
Tokens
294
Input
160
Total output
123
Reasoning within output
git_bisect f001
100%
Passed
Tokens
342
Input
94
Total output
75
Reasoning within output
git_bisect f002
100%
Passed
Tokens
359
Input
50
Total output
54
Reasoning within output
git_bisect f003
100%
Passed
Tokens
357
Input
117
Total output
110
Reasoning within output
git_bisect f004
100%
Passed
Tokens
357
Input
171
Total output
108
Reasoning within output
git_bisect f005
100%
Passed
Tokens
359
Input
260
Total output
186
Reasoning within output
git_bisect f006
100%
Passed
Tokens
388
Input
121
Total output
113
Reasoning within output
git_bisect f007
100%
Passed
Tokens
352
Input
199
Total output
192
Reasoning within output
git_bisect f008
100%
Passed
Tokens
367
Input
185
Total output
134
Reasoning within output
git_bisect f009
100%
Passed
Tokens
361
Input
144
Total output
127
Reasoning within output
git_bisect f010
100%
Passed
Tokens
359
Input
190
Total output
137
Reasoning within output
git_bisect f011
100%
Passed
Tokens
369
Input
186
Total output
112
Reasoning within output
git_bisect f012
100%
Passed
Tokens
650
Input
384
Total output
222
Reasoning within output
git_clean f001
100%
Passed
Tokens
216
Input
203
Total output
213
Reasoning within output
git_clean f002
100%
Passed
Tokens
227
Input
50
Total output
45
Reasoning within output
git_clean f003
100%
Passed
Tokens
224
Input
74
Total output
74
Reasoning within output
git_clean f004
66.7%
Failed
Tokens
224
Input
114
Total output
120
Reasoning within output
git_clean f005
100%
Passed
Tokens
216
Input
257
Total output
258
Reasoning within output
git_clean f006
100%
Passed
Tokens
242
Input
137
Total output
130
Reasoning within output
git_clean f007
100%
Passed
Tokens
238
Input
82
Total output
78
Reasoning within output
git_clean f008
100%
Passed
Tokens
240
Input
35
Total output
36
Reasoning within output
git_clean f009
100%
Passed
Tokens
233
Input
132
Total output
140
Reasoning within output
git_clean f010
100%
Passed
Tokens
230
Input
57
Total output
56
Reasoning within output
git_clean f011
60%
Failed
Tokens
229
Input
122
Total output
123
Reasoning within output
git_clean f012
100%
Passed
Tokens
227
Input
460
Total output
469
Reasoning within output
git_grep f001
100%
Passed
Tokens
221
Input
59
Total output
67
Reasoning within output
git_grep f002
100%
Passed
Tokens
233
Input
77
Total output
75
Reasoning within output
git_grep f003
100%
Passed
Tokens
255
Input
78
Total output
78
Reasoning within output
git_grep f004
0%
Failed
Tokens
244
Input
48
Total output
37
Reasoning within output
git_grep f005
100%
Passed
Tokens
274
Input
127
Total output
127
Reasoning within output
git_grep f006
100%
Passed
Tokens
209
Input
111
Total output
136
Reasoning within output
git_grep f007
100%
Passed
Tokens
325
Input
167
Total output
149
Reasoning within output
git_grep f008
100%
Passed
Tokens
228
Input
73
Total output
77
Reasoning within output
git_grep f009
100%
Passed
Tokens
219
Input
48
Total output
44
Reasoning within output
git_grep f010
100%
Passed
Tokens
229
Input
63
Total output
60
Reasoning within output
git_grep f011
100%
Passed
Tokens
224
Input
105
Total output
102
Reasoning within output
git_grep f012
100%
Passed
Tokens
233
Input
56
Total output
51
Reasoning within output
git_log_format f001
100%
Passed
Tokens
880
Input
101
Total output
91
Reasoning within output
git_log_format f002
100%
Passed
Tokens
870
Input
111
Total output
109
Reasoning within output
git_log_format f003
100%
Passed
Tokens
741
Input
126
Total output
104
Reasoning within output
git_log_format f004
100%
Passed
Tokens
848
Input
75
Total output
59
Reasoning within output
git_log_format f005
100%
Passed
Tokens
737
Input
130
Total output
102
Reasoning within output
git_log_format f006
100%
Passed
Tokens
871
Input
140
Total output
94
Reasoning within output
git_log_format f007
100%
Passed
Tokens
868
Input
38
Total output
33
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,115
Input
99
Total output
81
Reasoning within output
git_log_format f009
100%
Passed
Tokens
725
Input
52
Total output
59
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,212
Input
75
Total output
67
Reasoning within output
git_log_format f011
100%
Passed
Tokens
511
Input
61
Total output
59
Reasoning within output
git_log_format f012
100%
Passed
Tokens
494
Input
126
Total output
104
Reasoning within output
git_show f001
100%
Passed
Tokens
331
Input
43
Total output
49
Reasoning within output
git_show f002
0%
Failed
Tokens
335
Input
231
Total output
219
Reasoning within output
git_show f003
100%
Passed
Tokens
681
Input
94
Total output
86
Reasoning within output
git_show f004
100%
Passed
Tokens
553
Input
62
Total output
53
Reasoning within output
git_show f005
100%
Passed
Tokens
339
Input
74
Total output
73
Reasoning within output
git_show f006
100%
Passed
Tokens
352
Input
67
Total output
66
Reasoning within output
git_show f007
100%
Passed
Tokens
333
Input
106
Total output
86
Reasoning within output
git_show f008
100%
Passed
Tokens
349
Input
60
Total output
44
Reasoning within output
git_show f009
100%
Passed
Tokens
423
Input
47
Total output
30
Reasoning within output
git_show f010
100%
Passed
Tokens
334
Input
164
Total output
179
Reasoning within output
git_show f011
100%
Passed
Tokens
363
Input
38
Total output
40
Reasoning within output
git_show f012
100%
Passed
Tokens
794
Input
96
Total output
87
Reasoning within output
merge_conflicts f001
0%
Failed
Tokens
253
Input
221
Total output
254
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
273
Input
358
Total output
370
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
248
Input
264
Total output
301
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
296
Input
652
Total output
680
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
304
Input
384
Total output
401
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
288
Input
393
Total output
320
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
293
Input
337
Total output
347
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
275
Input
79
Total output
82
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
273
Input
96
Total output
96
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
343
Input
329
Total output
354
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
311
Input
275
Total output
223
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
394
Input
847
Total output
796
Reasoning within output
rebase f001
0%
Failed
Tokens
262
Input
513
Total output
615
Reasoning within output
rebase f002
100%
Passed
Tokens
293
Input
185
Total output
172
Reasoning within output
rebase f003
100%
Passed
Tokens
272
Input
152
Total output
159
Reasoning within output
rebase f004
100%
Passed
Tokens
309
Input
629
Total output
683
Reasoning within output
rebase f005
100%
Passed
Tokens
307
Input
160
Total output
158
Reasoning within output
rebase f006
100%
Passed
Tokens
300
Input
534
Total output
475
Reasoning within output
rebase f007
100%
Passed
Tokens
293
Input
235
Total output
242
Reasoning within output
rebase f008
100%
Passed
Tokens
287
Input
42
Total output
38
Reasoning within output
rebase f009
100%
Passed
Tokens
270
Input
39
Total output
35
Reasoning within output
rebase f010
100%
Passed
Tokens
368
Input
466
Total output
470
Reasoning within output
rebase f011
100%
Passed
Tokens
318
Input
213
Total output
149
Reasoning within output
rebase f012
0%
Failed
Tokens
421
Input
230
Total output
175
Reasoning within output
reflog f001
100%
Passed
Tokens
370
Input
190
Total output
154
Reasoning within output
reflog f002
100%
Passed
Tokens
365
Input
622
Total output
189
Reasoning within output
reflog f003
100%
Passed
Tokens
343
Input
879
Total output
210
Reasoning within output
reflog f004
100%
Passed
Tokens
434
Input
1,405
Total output
1,003
Reasoning within output
reflog f005
100%
Passed
Tokens
479
Input
1,299
Total output
784
Reasoning within output
reflog f006
100%
Passed
Tokens
388
Input
653
Total output
193
Reasoning within output
reflog f007
100%
Passed
Tokens
407
Input
1,059
Total output
547
Reasoning within output
reflog f008
100%
Passed
Tokens
475
Input
704
Total output
367
Reasoning within output
reflog f009
100%
Passed
Tokens
433
Input
697
Total output
223
Reasoning within output
reflog f010
100%
Passed
Tokens
394
Input
607
Total output
196
Reasoning within output
reflog f011
100%
Passed
Tokens
451
Input
1,008
Total output
557
Reasoning within output
reflog f012
100%
Passed
Tokens
451
Input
1,165
Total output
708
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
290
Input
174
Total output
102
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
425
Input
233
Total output
124
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
277
Input
120
Total output
74
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
277
Input
138
Total output
49
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
287
Input
357
Total output
125
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
334
Input
189
Total output
82
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
359
Input
192
Total output
103
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
329
Input
271
Total output
121
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
357
Input
295
Total output
123
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
286
Input
174
Total output
122
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
276
Input
476
Total output
339
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
287
Input
243
Total output
120
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
208
Input
67
Total output
66
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
257
Input
42
Total output
37
Reasoning within output
submodule_usage f003
100%
Passed
Tokens
262
Input
1,201
Total output
1,282
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
195
Input
34
Total output
28
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
215
Input
164
Total output
150
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
241
Input
204
Total output
234
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
166
Input
40
Total output
21
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
255
Input
82
Total output
95
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
250
Input
34
Total output
34
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
255
Input
103
Total output
103
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
263
Input
48
Total output
49
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
216
Input
125
Total output
125
Reasoning within output
tag_management f001
100%
Passed
Tokens
224
Input
44
Total output
33
Reasoning within output
tag_management f002
100%
Passed
Tokens
221
Input
91
Total output
60
Reasoning within output
tag_management f003
100%
Passed
Tokens
229
Input
37
Total output
28
Reasoning within output
tag_management f004
100%
Passed
Tokens
235
Input
51
Total output
44
Reasoning within output
tag_management f005
100%
Passed
Tokens
240
Input
48
Total output
38
Reasoning within output
tag_management f006
100%
Passed
Tokens
231
Input
70
Total output
56
Reasoning within output
tag_management f007
100%
Passed
Tokens
234
Input
47
Total output
36
Reasoning within output
tag_management f008
100%
Passed
Tokens
236
Input
129
Total output
73
Reasoning within output
tag_management f009
100%
Passed
Tokens
239
Input
142
Total output
132
Reasoning within output
tag_management f010
100%
Passed
Tokens
318
Input
35
Total output
24
Reasoning within output
tag_management f011
0%
Failed
Tokens
215
Input
45
Total output
44
Reasoning within output
tag_management f012
100%
Passed
Tokens
245
Input
127
Total output
109
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
292
Input
46
Total output
38
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
295
Input
219
Total output
188
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
348
Input
45
Total output
38
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
343
Input
38
Total output
39
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
285
Input
90
Total output
86
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
284
Input
76
Total output
62
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
286
Input
124
Total output
101
Reasoning within output
worktree_usage f008
0%
Failed
Tokens
315
Input
518
Total output
491
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
381
Input
73
Total output
74
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
356
Input
60
Total output
46
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
349
Input
37
Total output
31
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
285
Input
71
Total output
54
Reasoning within output