Anthropic / claude-haiku-4.5
low
92.2%
188 / 204 fixtures · 1 run(s)
45,475 input / 86,726 total output / 58,222 reasoning within output tokens $0.47910500
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-0.5%
Text: 92.2% →
JSON: 91.7%
+9
Gained
JSON pass / text fail
−10
Lost
Text pass / JSON fail
178
Unchanged Pass
Both pass
7
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|---|---|---|
| f001 | 0% (0/1) | 100% (1/1) | -100% |
| f004 | 0% (0/1) | 100% (1/1) | -100% |
| f010 | 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% |
| merge_conflicts | 91.7% | 75% | -16.7% |
| commit_squash | 83.3% | 91.7% | + 8.3% |
| blame_forensics | 100% | 91.7% | -8.3% |
| cherry_pick | 83.3% | 75% | -8.3% |
| git_clean | 100% | 91.7% | -8.3% |
| git_grep | 91.7% | 100% | + 8.3% |
| submodule_usage | 75% | 83.3% | + 8.3% |
| git_bisect | 100% | 100% | + 0% |
| git_log_format | 100% | 100% | + 0% |
| git_show | 91.7% | 91.7% | + 0% |
| rebase | 83.3% | 83.3% | + 0% |
| reflog | 100% | 100% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
| tag_management | 100% | 100% | + 0% |
| worktree_usage | 100% | 100% | + 0% |
Changed Fixtures (19)
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
271
Input
2,078
Total output
984
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
261
Input
638
Total output
423
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
295
Input
212
Total output
145
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
223
Input
587
Total output
534
Reasoning within output
blame_forensics f005
100%
Passed
Tokens
275
Input
4,413
Total output
1,734
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
262
Input
380
Total output
333
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
229
Input
413
Total output
321
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
263
Input
293
Total output
270
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
226
Input
356
Total output
346
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
447
Input
768
Total output
486
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
264
Input
269
Total output
242
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
182
Input
450
Total output
411
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
136
Input
201
Total output
162
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
147
Input
195
Total output
158
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
126
Input
155
Total output
156
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
166
Input
259
Total output
218
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
136
Input
197
Total output
172
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
153
Input
177
Total output
148
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
141
Input
168
Total output
130
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
152
Input
214
Total output
178
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
118
Input
186
Total output
193
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
150
Input
157
Total output
114
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
102
Input
134
Total output
136
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
190
Input
192
Total output
161
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
139
Input
546
Total output
593
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
158
Input
705
Total output
616
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
131
Input
485
Total output
523
Reasoning within output
cherry_pick f004
0%
Failed
Tokens
191
Input
722
Total output
651
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
200
Input
262
Total output
234
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
178
Input
323
Total output
262
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
171
Input
198
Total output
172
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
167
Input
160
Total output
146
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
151
Input
204
Total output
173
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
267
Input
530
Total output
485
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
209
Input
376
Total output
282
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
321
Input
684
Total output
535
Reasoning within output
commit_messages f001
32.7%
Failed
Tokens
147
Input
449
Total output
341
Reasoning within output
commit_messages f002
87.7%
Passed
Tokens
267
Input
351
Total output
235
Reasoning within output
commit_messages f003
93.7%
Passed
Tokens
123
Input
372
Total output
258
Reasoning within output
commit_messages f004
35%
Failed
Tokens
152
Input
303
Total output
260
Reasoning within output
commit_messages f005
91%
Passed
Tokens
146
Input
273
Total output
221
Reasoning within output
commit_messages f006
80.3%
Passed
Tokens
178
Input
478
Total output
482
Reasoning within output
commit_messages f007
91.7%
Passed
Tokens
240
Input
315
Total output
235
Reasoning within output
commit_messages f008
90%
Passed
Tokens
108
Input
211
Total output
184
Reasoning within output
commit_messages f009
57.7%
Passed
Tokens
171
Input
1,414
Total output
537
Reasoning within output
commit_messages f010
41%
Failed
Tokens
350
Input
612
Total output
466
Reasoning within output
commit_messages f011
33.3%
Failed
Tokens
194
Input
1,041
Total output
533
Reasoning within output
commit_messages f012
92%
Passed
Tokens
171
Input
205
Total output
192
Reasoning within output
commit_squash f001
100%
Passed
Tokens
170
Input
724
Total output
490
Reasoning within output
commit_squash f002
100%
Passed
Tokens
147
Input
264
Total output
221
Reasoning within output
commit_squash f003
0%
Failed
Tokens
165
Input
2,838
Total output
1,411
Reasoning within output
commit_squash f004
100%
Passed
Tokens
149
Input
560
Total output
500
Reasoning within output
commit_squash f005
100%
Passed
Tokens
145
Input
389
Total output
377
Reasoning within output
commit_squash f006
100%
Passed
Tokens
132
Input
235
Total output
223
Reasoning within output
commit_squash f007
100%
Passed
Tokens
133
Input
752
Total output
578
Reasoning within output
commit_squash f008
100%
Passed
Tokens
148
Input
345
Total output
292
Reasoning within output
commit_squash f009
100%
Passed
Tokens
140
Input
539
Total output
444
Reasoning within output
commit_squash f010
0%
Failed
Tokens
154
Input
272
Total output
217
Reasoning within output
commit_squash f011
100%
Passed
Tokens
164
Input
539
Total output
471
Reasoning within output
commit_squash f012
100%
Passed
Tokens
159
Input
275
Total output
234
Reasoning within output
git_bisect f001
100%
Passed
Tokens
217
Input
333
Total output
229
Reasoning within output
git_bisect f002
100%
Passed
Tokens
240
Input
419
Total output
208
Reasoning within output
git_bisect f003
100%
Passed
Tokens
242
Input
318
Total output
164
Reasoning within output
git_bisect f004
100%
Passed
Tokens
250
Input
401
Total output
201
Reasoning within output
git_bisect f005
100%
Passed
Tokens
244
Input
459
Total output
224
Reasoning within output
git_bisect f006
100%
Passed
Tokens
277
Input
465
Total output
208
Reasoning within output
git_bisect f007
100%
Passed
Tokens
244
Input
520
Total output
262
Reasoning within output
git_bisect f008
100%
Passed
Tokens
240
Input
459
Total output
253
Reasoning within output
git_bisect f009
100%
Passed
Tokens
236
Input
423
Total output
238
Reasoning within output
git_bisect f010
100%
Passed
Tokens
234
Input
401
Total output
209
Reasoning within output
git_bisect f011
100%
Passed
Tokens
242
Input
436
Total output
253
Reasoning within output
git_bisect f012
100%
Passed
Tokens
263
Input
445
Total output
223
Reasoning within output
git_clean f001
100%
Passed
Tokens
90
Input
203
Total output
174
Reasoning within output
git_clean f002
100%
Passed
Tokens
103
Input
142
Total output
139
Reasoning within output
git_clean f003
100%
Passed
Tokens
102
Input
182
Total output
174
Reasoning within output
git_clean f004
100%
Passed
Tokens
102
Input
174
Total output
150
Reasoning within output
git_clean f005
100%
Passed
Tokens
88
Input
347
Total output
319
Reasoning within output
git_clean f006
100%
Passed
Tokens
122
Input
271
Total output
241
Reasoning within output
git_clean f007
100%
Passed
Tokens
114
Input
264
Total output
217
Reasoning within output
git_clean f008
100%
Passed
Tokens
123
Input
358
Total output
323
Reasoning within output
git_clean f009
100%
Passed
Tokens
114
Input
272
Total output
262
Reasoning within output
git_clean f010
100%
Passed
Tokens
104
Input
524
Total output
518
Reasoning within output
git_clean f011
100%
Passed
Tokens
106
Input
229
Total output
212
Reasoning within output
git_clean f012
100%
Passed
Tokens
108
Input
373
Total output
316
Reasoning within output
git_grep f001
100%
Passed
Tokens
88
Input
113
Total output
100
Reasoning within output
git_grep f002
100%
Passed
Tokens
95
Input
99
Total output
81
Reasoning within output
git_grep f003
100%
Passed
Tokens
138
Input
126
Total output
94
Reasoning within output
git_grep f004
100%
Passed
Tokens
115
Input
128
Total output
101
Reasoning within output
git_grep f005
100%
Passed
Tokens
154
Input
190
Total output
148
Reasoning within output
git_grep f006
100%
Passed
Tokens
72
Input
174
Total output
193
Reasoning within output
git_grep f007
0%
Failed
Tokens
215
Input
415
Total output
324
Reasoning within output
git_grep f008
100%
Passed
Tokens
93
Input
115
Total output
104
Reasoning within output
git_grep f009
100%
Passed
Tokens
87
Input
110
Total output
91
Reasoning within output
git_grep f010
100%
Passed
Tokens
101
Input
103
Total output
86
Reasoning within output
git_grep f011
100%
Passed
Tokens
165
Input
159
Total output
115
Reasoning within output
git_grep f012
100%
Passed
Tokens
103
Input
122
Total output
84
Reasoning within output
git_log_format f001
100%
Passed
Tokens
830
Input
152
Total output
104
Reasoning within output
git_log_format f002
100%
Passed
Tokens
830
Input
176
Total output
144
Reasoning within output
git_log_format f003
100%
Passed
Tokens
671
Input
381
Total output
268
Reasoning within output
git_log_format f004
100%
Passed
Tokens
819
Input
351
Total output
222
Reasoning within output
git_log_format f005
100%
Passed
Tokens
676
Input
274
Total output
188
Reasoning within output
git_log_format f006
100%
Passed
Tokens
828
Input
285
Total output
172
Reasoning within output
git_log_format f007
100%
Passed
Tokens
844
Input
135
Total output
115
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,088
Input
142
Total output
111
Reasoning within output
git_log_format f009
100%
Passed
Tokens
650
Input
196
Total output
173
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,249
Input
238
Total output
165
Reasoning within output
git_log_format f011
100%
Passed
Tokens
436
Input
174
Total output
128
Reasoning within output
git_log_format f012
100%
Passed
Tokens
385
Input
192
Total output
138
Reasoning within output
git_show f001
100%
Passed
Tokens
217
Input
119
Total output
90
Reasoning within output
git_show f002
0%
Failed
Tokens
227
Input
558
Total output
535
Reasoning within output
git_show f003
100%
Passed
Tokens
691
Input
157
Total output
110
Reasoning within output
git_show f004
100%
Passed
Tokens
469
Input
133
Total output
99
Reasoning within output
git_show f005
100%
Passed
Tokens
232
Input
388
Total output
368
Reasoning within output
git_show f006
100%
Passed
Tokens
242
Input
247
Total output
183
Reasoning within output
git_show f007
100%
Passed
Tokens
220
Input
171
Total output
130
Reasoning within output
git_show f008
100%
Passed
Tokens
239
Input
180
Total output
107
Reasoning within output
git_show f009
100%
Passed
Tokens
333
Input
309
Total output
244
Reasoning within output
git_show f010
100%
Passed
Tokens
229
Input
111
Total output
94
Reasoning within output
git_show f011
100%
Passed
Tokens
258
Input
108
Total output
88
Reasoning within output
git_show f012
100%
Passed
Tokens
382
Input
184
Total output
143
Reasoning within output
merge_conflicts f001
100%
Passed
Tokens
127
Input
600
Total output
592
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
145
Input
385
Total output
375
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
120
Input
876
Total output
596
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
177
Input
424
Total output
384
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
185
Input
259
Total output
213
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
162
Input
341
Total output
304
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
175
Input
196
Total output
160
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
149
Input
158
Total output
143
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
149
Input
251
Total output
218
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
232
Input
477
Total output
428
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
188
Input
316
Total output
244
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
284
Input
1,474
Total output
831
Reasoning within output
rebase f001
100%
Passed
Tokens
140
Input
752
Total output
686
Reasoning within output
rebase f002
100%
Passed
Tokens
165
Input
266
Total output
223
Reasoning within output
rebase f003
100%
Passed
Tokens
143
Input
867
Total output
632
Reasoning within output
rebase f004
0%
Failed
Tokens
191
Input
921
Total output
575
Reasoning within output
rebase f005
100%
Passed
Tokens
191
Input
736
Total output
550
Reasoning within output
rebase f006
100%
Passed
Tokens
176
Input
962
Total output
564
Reasoning within output
rebase f007
100%
Passed
Tokens
176
Input
246
Total output
221
Reasoning within output
rebase f008
100%
Passed
Tokens
161
Input
225
Total output
199
Reasoning within output
rebase f009
100%
Passed
Tokens
148
Input
274
Total output
241
Reasoning within output
rebase f010
100%
Passed
Tokens
263
Input
226
Total output
176
Reasoning within output
rebase f011
100%
Passed
Tokens
201
Input
356
Total output
228
Reasoning within output
rebase f012
0%
Failed
Tokens
319
Input
1,480
Total output
790
Reasoning within output
reflog f001
100%
Passed
Tokens
238
Input
562
Total output
407
Reasoning within output
reflog f002
100%
Passed
Tokens
236
Input
413
Total output
183
Reasoning within output
reflog f003
100%
Passed
Tokens
213
Input
599
Total output
183
Reasoning within output
reflog f004
100%
Passed
Tokens
322
Input
1,507
Total output
805
Reasoning within output
reflog f005
100%
Passed
Tokens
375
Input
1,364
Total output
600
Reasoning within output
reflog f006
100%
Passed
Tokens
260
Input
612
Total output
231
Reasoning within output
reflog f007
100%
Passed
Tokens
283
Input
867
Total output
431
Reasoning within output
reflog f008
100%
Passed
Tokens
360
Input
888
Total output
399
Reasoning within output
reflog f009
100%
Passed
Tokens
318
Input
993
Total output
414
Reasoning within output
reflog f010
100%
Passed
Tokens
276
Input
644
Total output
314
Reasoning within output
reflog f011
100%
Passed
Tokens
318
Input
1,834
Total output
666
Reasoning within output
reflog f012
100%
Passed
Tokens
304
Input
881
Total output
341
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
186
Input
228
Total output
146
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
360
Input
264
Total output
119
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
169
Input
279
Total output
151
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
167
Input
314
Total output
149
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
181
Input
307
Total output
160
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
238
Input
281
Total output
138
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
271
Input
229
Total output
131
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
231
Input
370
Total output
210
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
273
Input
382
Total output
200
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
179
Input
182
Total output
87
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
166
Input
279
Total output
168
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
188
Input
346
Total output
179
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
77
Input
154
Total output
114
Reasoning within output
submodule_usage f002
0%
Failed
Tokens
139
Input
426
Total output
394
Reasoning within output
submodule_usage f003
75%
Failed
Tokens
141
Input
2,260
Total output
911
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
123
Input
106
Total output
102
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
84
Input
327
Total output
240
Reasoning within output
submodule_usage f006
0%
Failed
Tokens
124
Input
956
Total output
659
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
87
Input
166
Total output
110
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
134
Input
489
Total output
472
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
131
Input
312
Total output
253
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
136
Input
595
Total output
598
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
142
Input
97
Total output
83
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
85
Input
203
Total output
164
Reasoning within output
tag_management f001
100%
Passed
Tokens
90
Input
143
Total output
125
Reasoning within output
tag_management f002
100%
Passed
Tokens
92
Input
148
Total output
99
Reasoning within output
tag_management f003
100%
Passed
Tokens
97
Input
242
Total output
211
Reasoning within output
tag_management f004
100%
Passed
Tokens
101
Input
96
Total output
91
Reasoning within output
tag_management f005
100%
Passed
Tokens
111
Input
137
Total output
109
Reasoning within output
tag_management f006
100%
Passed
Tokens
98
Input
158
Total output
111
Reasoning within output
tag_management f007
100%
Passed
Tokens
95
Input
518
Total output
425
Reasoning within output
tag_management f008
100%
Passed
Tokens
103
Input
503
Total output
409
Reasoning within output
tag_management f009
100%
Passed
Tokens
109
Input
191
Total output
152
Reasoning within output
tag_management f010
100%
Passed
Tokens
81
Input
146
Total output
115
Reasoning within output
tag_management f011
100%
Passed
Tokens
84
Input
157
Total output
147
Reasoning within output
tag_management f012
100%
Passed
Tokens
116
Input
262
Total output
219
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
174
Input
144
Total output
112
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
183
Input
259
Total output
185
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
242
Input
135
Total output
109
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
241
Input
104
Total output
94
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
166
Input
217
Total output
172
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
164
Input
186
Total output
146
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
177
Input
173
Total output
99
Reasoning within output
worktree_usage f008
100%
Passed
Tokens
197
Input
735
Total output
549
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
282
Input
148
Total output
131
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
250
Input
180
Total output
139
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
245
Input
138
Total output
109
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
168
Input
146
Total output
93
Reasoning within output