Anthropic / claude-haiku-4.5
high
90.7%
185 / 204 fixtures · 1 run(s)
45,497 input / 126,438 total output / 71,097 reasoning within output tokens $0.67768700
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
+2.4%
Text: 90.7% →
JSON: 93.1%
+13
Gained
JSON pass / text fail
−8
Lost
Text pass / JSON fail
177
Unchanged Pass
Both pass
6
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|---|---|---|
| f002 | 0% (0/1) | 100% (1/1) | -100% |
| f004 | 0% (0/1) | 100% (1/1) | -100% |
| f005 | 0% (0/1) | 100% (1/1) | -100% |
| f011 | 0% (0/1) | 100% (1/1) | -100% |
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| commit_messages | 66.7% | 100% | + 33.3% |
| branch_cleanup | 100% | 75% | -25% |
| commit_squash | 83.3% | 91.7% | + 8.3% |
| cherry_pick | 75% | 83.3% | + 8.3% |
| git_clean | 91.7% | 100% | + 8.3% |
| git_show | 91.7% | 100% | + 8.3% |
| blame_forensics | 91.7% | 91.7% | + 0% |
| git_bisect | 100% | 100% | + 0% |
| git_grep | 100% | 100% | + 0% |
| git_log_format | 100% | 100% | + 0% |
| merge_conflicts | 75% | 75% | + 0% |
| rebase | 83.3% | 83.3% | + 0% |
| reflog | 100% | 100% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
| submodule_usage | 83.3% | 83.3% | + 0% |
| tag_management | 100% | 100% | + 0% |
| worktree_usage | 100% | 100% | + 0% |
Changed Fixtures (21)
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
275
Input
939
Total output
657
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
259
Input
551
Total output
467
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
295
Input
331
Total output
243
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
221
Input
741
Total output
589
Reasoning within output
blame_forensics f005
0%
Failed
Tokens
272
Input
10,622
Total output
3,298
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
259
Input
386
Total output
323
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
230
Input
138
Total output
113
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
269
Input
588
Total output
512
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
229
Input
451
Total output
420
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
443
Input
661
Total output
465
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
267
Input
739
Total output
398
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
176
Input
680
Total output
513
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
138
Input
229
Total output
199
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
152
Input
193
Total output
155
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
128
Input
190
Total output
189
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
160
Input
213
Total output
184
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
142
Input
204
Total output
170
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
157
Input
268
Total output
227
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
136
Input
164
Total output
158
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
155
Input
173
Total output
147
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
117
Input
142
Total output
141
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
152
Input
187
Total output
134
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
100
Input
182
Total output
195
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
189
Input
246
Total output
178
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
138
Input
320
Total output
332
Reasoning within output
cherry_pick f002
0%
Failed
Tokens
158
Input
1,789
Total output
822
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
132
Input
713
Total output
566
Reasoning within output
cherry_pick f004
0%
Failed
Tokens
192
Input
1,265
Total output
710
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
198
Input
263
Total output
230
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
174
Input
422
Total output
348
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
173
Input
242
Total output
213
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
167
Input
157
Total output
143
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
150
Input
231
Total output
204
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
265
Input
716
Total output
585
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
211
Input
327
Total output
228
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
317
Input
5,669
Total output
2,136
Reasoning within output
commit_messages f001
91.7%
Passed
Tokens
147
Input
239
Total output
210
Reasoning within output
commit_messages f002
48.3%
Failed
Tokens
267
Input
1,424
Total output
452
Reasoning within output
commit_messages f003
93.3%
Passed
Tokens
123
Input
365
Total output
239
Reasoning within output
commit_messages f004
0%
Failed
Tokens
152
Input
329
Total output
221
Reasoning within output
commit_messages f005
46.7%
Failed
Tokens
146
Input
386
Total output
358
Reasoning within output
commit_messages f006
92.7%
Passed
Tokens
178
Input
332
Total output
314
Reasoning within output
commit_messages f007
89.3%
Passed
Tokens
240
Input
306
Total output
269
Reasoning within output
commit_messages f008
89.7%
Passed
Tokens
108
Input
310
Total output
243
Reasoning within output
commit_messages f009
64.3%
Passed
Tokens
171
Input
734
Total output
519
Reasoning within output
commit_messages f010
61.7%
Passed
Tokens
350
Input
559
Total output
384
Reasoning within output
commit_messages f011
48.3%
Failed
Tokens
194
Input
832
Total output
401
Reasoning within output
commit_messages f012
91.7%
Passed
Tokens
171
Input
281
Total output
239
Reasoning within output
commit_squash f001
100%
Passed
Tokens
171
Input
411
Total output
359
Reasoning within output
commit_squash f002
100%
Passed
Tokens
154
Input
318
Total output
280
Reasoning within output
commit_squash f003
0%
Failed
Tokens
160
Input
662
Total output
529
Reasoning within output
commit_squash f004
100%
Passed
Tokens
146
Input
666
Total output
511
Reasoning within output
commit_squash f005
100%
Passed
Tokens
147
Input
698
Total output
555
Reasoning within output
commit_squash f006
100%
Passed
Tokens
136
Input
377
Total output
353
Reasoning within output
commit_squash f007
0%
Failed
Tokens
134
Input
1,424
Total output
569
Reasoning within output
commit_squash f008
100%
Passed
Tokens
148
Input
333
Total output
283
Reasoning within output
commit_squash f009
100%
Passed
Tokens
138
Input
1,695
Total output
747
Reasoning within output
commit_squash f010
100%
Passed
Tokens
154
Input
289
Total output
245
Reasoning within output
commit_squash f011
100%
Passed
Tokens
166
Input
236
Total output
178
Reasoning within output
commit_squash f012
100%
Passed
Tokens
162
Input
360
Total output
270
Reasoning within output
git_bisect f001
100%
Passed
Tokens
213
Input
439
Total output
275
Reasoning within output
git_bisect f002
100%
Passed
Tokens
242
Input
519
Total output
298
Reasoning within output
git_bisect f003
100%
Passed
Tokens
232
Input
419
Total output
221
Reasoning within output
git_bisect f004
100%
Passed
Tokens
250
Input
474
Total output
254
Reasoning within output
git_bisect f005
100%
Passed
Tokens
246
Input
374
Total output
169
Reasoning within output
git_bisect f006
100%
Passed
Tokens
261
Input
463
Total output
255
Reasoning within output
git_bisect f007
100%
Passed
Tokens
242
Input
437
Total output
233
Reasoning within output
git_bisect f008
100%
Passed
Tokens
242
Input
483
Total output
286
Reasoning within output
git_bisect f009
100%
Passed
Tokens
240
Input
523
Total output
294
Reasoning within output
git_bisect f010
100%
Passed
Tokens
238
Input
531
Total output
311
Reasoning within output
git_bisect f011
100%
Passed
Tokens
232
Input
475
Total output
263
Reasoning within output
git_bisect f012
100%
Passed
Tokens
263
Input
470
Total output
238
Reasoning within output
git_clean f001
100%
Passed
Tokens
90
Input
331
Total output
311
Reasoning within output
git_clean f002
100%
Passed
Tokens
103
Input
156
Total output
144
Reasoning within output
git_clean f003
100%
Passed
Tokens
102
Input
153
Total output
144
Reasoning within output
git_clean f004
66.7%
Failed
Tokens
102
Input
413
Total output
380
Reasoning within output
git_clean f005
100%
Passed
Tokens
88
Input
1,298
Total output
720
Reasoning within output
git_clean f006
100%
Passed
Tokens
122
Input
313
Total output
273
Reasoning within output
git_clean f007
100%
Passed
Tokens
114
Input
240
Total output
207
Reasoning within output
git_clean f008
100%
Passed
Tokens
123
Input
215
Total output
198
Reasoning within output
git_clean f009
100%
Passed
Tokens
114
Input
275
Total output
257
Reasoning within output
git_clean f010
100%
Passed
Tokens
104
Input
202
Total output
191
Reasoning within output
git_clean f011
100%
Passed
Tokens
106
Input
752
Total output
526
Reasoning within output
git_clean f012
100%
Passed
Tokens
108
Input
470
Total output
406
Reasoning within output
git_grep f001
100%
Passed
Tokens
88
Input
119
Total output
107
Reasoning within output
git_grep f002
100%
Passed
Tokens
96
Input
97
Total output
79
Reasoning within output
git_grep f003
100%
Passed
Tokens
138
Input
130
Total output
98
Reasoning within output
git_grep f004
100%
Passed
Tokens
115
Input
118
Total output
99
Reasoning within output
git_grep f005
100%
Passed
Tokens
154
Input
154
Total output
121
Reasoning within output
git_grep f006
100%
Passed
Tokens
72
Input
254
Total output
271
Reasoning within output
git_grep f007
100%
Passed
Tokens
215
Input
726
Total output
528
Reasoning within output
git_grep f008
100%
Passed
Tokens
93
Input
96
Total output
84
Reasoning within output
git_grep f009
100%
Passed
Tokens
87
Input
110
Total output
96
Reasoning within output
git_grep f010
100%
Passed
Tokens
101
Input
114
Total output
92
Reasoning within output
git_grep f011
100%
Passed
Tokens
165
Input
139
Total output
94
Reasoning within output
git_grep f012
100%
Passed
Tokens
103
Input
132
Total output
95
Reasoning within output
git_log_format f001
100%
Passed
Tokens
836
Input
126
Total output
95
Reasoning within output
git_log_format f002
100%
Passed
Tokens
823
Input
197
Total output
153
Reasoning within output
git_log_format f003
100%
Passed
Tokens
678
Input
298
Total output
211
Reasoning within output
git_log_format f004
100%
Passed
Tokens
817
Input
372
Total output
227
Reasoning within output
git_log_format f005
100%
Passed
Tokens
676
Input
313
Total output
197
Reasoning within output
git_log_format f006
100%
Passed
Tokens
828
Input
315
Total output
182
Reasoning within output
git_log_format f007
100%
Passed
Tokens
851
Input
246
Total output
190
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,102
Input
142
Total output
110
Reasoning within output
git_log_format f009
100%
Passed
Tokens
674
Input
213
Total output
185
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,247
Input
237
Total output
157
Reasoning within output
git_log_format f011
100%
Passed
Tokens
433
Input
241
Total output
198
Reasoning within output
git_log_format f012
100%
Passed
Tokens
396
Input
168
Total output
112
Reasoning within output
git_show f001
100%
Passed
Tokens
216
Input
107
Total output
90
Reasoning within output
git_show f002
0%
Failed
Tokens
231
Input
1,048
Total output
666
Reasoning within output
git_show f003
100%
Passed
Tokens
693
Input
120
Total output
88
Reasoning within output
git_show f004
100%
Passed
Tokens
471
Input
168
Total output
131
Reasoning within output
git_show f005
100%
Passed
Tokens
235
Input
223
Total output
210
Reasoning within output
git_show f006
100%
Passed
Tokens
245
Input
144
Total output
115
Reasoning within output
git_show f007
100%
Passed
Tokens
220
Input
115
Total output
90
Reasoning within output
git_show f008
100%
Passed
Tokens
238
Input
170
Total output
112
Reasoning within output
git_show f009
100%
Passed
Tokens
325
Input
397
Total output
313
Reasoning within output
git_show f010
100%
Passed
Tokens
228
Input
220
Total output
191
Reasoning within output
git_show f011
100%
Passed
Tokens
256
Input
191
Total output
142
Reasoning within output
git_show f012
100%
Passed
Tokens
381
Input
192
Total output
152
Reasoning within output
merge_conflicts f001
100%
Passed
Tokens
127
Input
419
Total output
438
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
145
Input
583
Total output
498
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
120
Input
973
Total output
647
Reasoning within output
merge_conflicts f004
0%
Failed
Tokens
177
Input
442
Total output
348
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
185
Input
411
Total output
351
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
162
Input
332
Total output
273
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
175
Input
260
Total output
218
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
149
Input
175
Total output
174
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
149
Input
183
Total output
153
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
232
Input
214
Total output
169
Reasoning within output
merge_conflicts f011
0%
Failed
Tokens
188
Input
486
Total output
315
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
284
Input
654
Total output
502
Reasoning within output
rebase f001
100%
Passed
Tokens
140
Input
2,704
Total output
1,453
Reasoning within output
rebase f002
100%
Passed
Tokens
166
Input
328
Total output
280
Reasoning within output
rebase f003
100%
Passed
Tokens
145
Input
954
Total output
671
Reasoning within output
rebase f004
0%
Failed
Tokens
191
Input
1,863
Total output
880
Reasoning within output
rebase f005
100%
Passed
Tokens
191
Input
788
Total output
548
Reasoning within output
rebase f006
100%
Passed
Tokens
175
Input
278
Total output
215
Reasoning within output
rebase f007
100%
Passed
Tokens
175
Input
345
Total output
276
Reasoning within output
rebase f008
100%
Passed
Tokens
161
Input
199
Total output
188
Reasoning within output
rebase f009
100%
Passed
Tokens
147
Input
473
Total output
424
Reasoning within output
rebase f010
100%
Passed
Tokens
259
Input
280
Total output
224
Reasoning within output
rebase f011
100%
Passed
Tokens
201
Input
464
Total output
336
Reasoning within output
rebase f012
0%
Failed
Tokens
321
Input
6,949
Total output
2,874
Reasoning within output
reflog f001
100%
Passed
Tokens
234
Input
452
Total output
309
Reasoning within output
reflog f002
100%
Passed
Tokens
229
Input
451
Total output
203
Reasoning within output
reflog f003
100%
Passed
Tokens
222
Input
702
Total output
238
Reasoning within output
reflog f004
100%
Passed
Tokens
326
Input
1,448
Total output
656
Reasoning within output
reflog f005
100%
Passed
Tokens
367
Input
754
Total output
291
Reasoning within output
reflog f006
100%
Passed
Tokens
254
Input
447
Total output
140
Reasoning within output
reflog f007
100%
Passed
Tokens
279
Input
1,155
Total output
406
Reasoning within output
reflog f008
100%
Passed
Tokens
357
Input
650
Total output
346
Reasoning within output
reflog f009
100%
Passed
Tokens
310
Input
941
Total output
476
Reasoning within output
reflog f010
100%
Passed
Tokens
277
Input
766
Total output
312
Reasoning within output
reflog f011
100%
Passed
Tokens
332
Input
5,512
Total output
1,971
Reasoning within output
reflog f012
100%
Passed
Tokens
311
Input
1,180
Total output
531
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
187
Input
258
Total output
144
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
360
Input
338
Total output
195
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
169
Input
222
Total output
132
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
167
Input
279
Total output
161
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
181
Input
305
Total output
160
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
238
Input
310
Total output
145
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
269
Input
351
Total output
221
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
231
Input
325
Total output
173
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
273
Input
286
Total output
158
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
179
Input
207
Total output
115
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
166
Input
203
Total output
114
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
188
Input
349
Total output
204
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
77
Input
130
Total output
105
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
135
Input
498
Total output
496
Reasoning within output
submodule_usage f003
100%
Passed
Tokens
143
Input
6,769
Total output
2,719
Reasoning within output
submodule_usage f004
0%
Failed
Tokens
129
Input
126
Total output
113
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
84
Input
562
Total output
479
Reasoning within output
submodule_usage f006
0%
Failed
Tokens
124
Input
8,755
Total output
2,827
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
87
Input
160
Total output
113
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
136
Input
867
Total output
605
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
130
Input
289
Total output
257
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
133
Input
616
Total output
572
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
143
Input
111
Total output
94
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
85
Input
176
Total output
137
Reasoning within output
tag_management f001
100%
Passed
Tokens
92
Input
193
Total output
174
Reasoning within output
tag_management f002
100%
Passed
Tokens
90
Input
194
Total output
143
Reasoning within output
tag_management f003
100%
Passed
Tokens
94
Input
225
Total output
213
Reasoning within output
tag_management f004
100%
Passed
Tokens
99
Input
93
Total output
80
Reasoning within output
tag_management f005
100%
Passed
Tokens
112
Input
156
Total output
110
Reasoning within output
tag_management f006
100%
Passed
Tokens
100
Input
142
Total output
101
Reasoning within output
tag_management f007
100%
Passed
Tokens
96
Input
246
Total output
208
Reasoning within output
tag_management f008
100%
Passed
Tokens
101
Input
623
Total output
416
Reasoning within output
tag_management f009
100%
Passed
Tokens
109
Input
410
Total output
358
Reasoning within output
tag_management f010
100%
Passed
Tokens
81
Input
136
Total output
110
Reasoning within output
tag_management f011
100%
Passed
Tokens
80
Input
112
Total output
102
Reasoning within output
tag_management f012
100%
Passed
Tokens
115
Input
272
Total output
210
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
176
Input
143
Total output
110
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
178
Input
245
Total output
169
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
239
Input
128
Total output
100
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
238
Input
97
Total output
83
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
170
Input
190
Total output
146
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
169
Input
285
Total output
221
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
171
Input
219
Total output
163
Reasoning within output
worktree_usage f008
100%
Passed
Tokens
195
Input
412
Total output
277
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
286
Input
150
Total output
136
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
251
Input
283
Total output
229
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
251
Input
183
Total output
130
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
164
Input
119
Total output
80
Reasoning within output