Prism-ml / ternary-bonsai-2-27b
high
93.6%
191 / 204 fixtures · 1 run(s)
41,347 input / 66,042 total output / 56,816 reasoning within output tokens $0.03612203
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
+0.5%
Text: 93.6% →
JSON: 94.1%
+7
Gained
JSON pass / text fail
−6
Lost
Text pass / JSON fail
185
Unchanged Pass
Both pass
6
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| branch_cleanup | 100% | 75% | -25% |
| git_clean | 66.7% | 91.7% | + 25% |
| rebase | 91.7% | 83.3% | -8.3% |
| worktree_usage | 83.3% | 91.7% | + 8.3% |
| blame_forensics | 91.7% | 100% | + 8.3% |
| commit_squash | 100% | 91.7% | -8.3% |
| git_show | 91.7% | 100% | + 8.3% |
| cherry_pick | 91.7% | 91.7% | + 0% |
| commit_messages | 100% | 100% | + 0% |
| git_bisect | 100% | 100% | + 0% |
| git_grep | 100% | 100% | + 0% |
| git_log_format | 100% | 100% | + 0% |
| merge_conflicts | 83.3% | 83.3% | + 0% |
| reflog | 100% | 100% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
| submodule_usage | 91.7% | 91.7% | + 0% |
| tag_management | 100% | 100% | + 0% |
Changed Fixtures (13)
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
258
Input
395
Total output
387
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
254
Input
318
Total output
311
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
287
Input
241
Total output
232
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
207
Input
194
Total output
186
Reasoning within output
blame_forensics f005
0%
Failed
Tokens
270
Input
1,134
Total output
1,127
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
247
Input
409
Total output
403
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
226
Input
200
Total output
192
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
262
Input
1,168
Total output
1,161
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
212
Input
336
Total output
329
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
445
Input
528
Total output
522
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
260
Input
269
Total output
263
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
156
Input
367
Total output
359
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
108
Input
178
Total output
172
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
127
Input
211
Total output
203
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
102
Input
169
Total output
165
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
136
Input
201
Total output
193
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
115
Input
160
Total output
154
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
134
Input
183
Total output
171
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
112
Input
143
Total output
135
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
135
Input
195
Total output
183
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
89
Input
220
Total output
216
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
128
Input
219
Total output
211
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
75
Input
134
Total output
130
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
164
Input
244
Total output
231
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
101
Input
276
Total output
269
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
120
Input
478
Total output
468
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
94
Input
287
Total output
280
Reasoning within output
cherry_pick f004
100%
Passed
Tokens
162
Input
1,533
Total output
1,501
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
156
Input
345
Total output
319
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
148
Input
319
Total output
289
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
139
Input
244
Total output
231
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
128
Input
116
Total output
107
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
112
Input
115
Total output
107
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
222
Input
307
Total output
276
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
172
Input
315
Total output
288
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
279
Input
588
Total output
542
Reasoning within output
commit_messages f001
82.5%
Passed
Tokens
117
Input
130
Total output
118
Reasoning within output
commit_messages f002
82.5%
Passed
Tokens
233
Input
173
Total output
160
Reasoning within output
commit_messages f003
92.5%
Passed
Tokens
88
Input
152
Total output
141
Reasoning within output
commit_messages f004
91.5%
Passed
Tokens
125
Input
112
Total output
106
Reasoning within output
commit_messages f005
92.5%
Passed
Tokens
109
Input
127
Total output
120
Reasoning within output
commit_messages f006
92.5%
Passed
Tokens
156
Input
171
Total output
163
Reasoning within output
commit_messages f007
94%
Passed
Tokens
197
Input
178
Total output
169
Reasoning within output
commit_messages f008
90%
Passed
Tokens
81
Input
160
Total output
153
Reasoning within output
commit_messages f009
79%
Passed
Tokens
135
Input
201
Total output
191
Reasoning within output
commit_messages f010
90%
Passed
Tokens
305
Input
476
Total output
464
Reasoning within output
commit_messages f011
89%
Passed
Tokens
154
Input
731
Total output
718
Reasoning within output
commit_messages f012
94%
Passed
Tokens
135
Input
160
Total output
152
Reasoning within output
commit_squash f001
100%
Passed
Tokens
145
Input
336
Total output
321
Reasoning within output
commit_squash f002
100%
Passed
Tokens
121
Input
218
Total output
207
Reasoning within output
commit_squash f003
100%
Passed
Tokens
136
Input
223
Total output
208
Reasoning within output
commit_squash f004
100%
Passed
Tokens
125
Input
267
Total output
253
Reasoning within output
commit_squash f005
100%
Passed
Tokens
121
Input
241
Total output
231
Reasoning within output
commit_squash f006
100%
Passed
Tokens
105
Input
356
Total output
342
Reasoning within output
commit_squash f007
100%
Passed
Tokens
112
Input
240
Total output
225
Reasoning within output
commit_squash f008
100%
Passed
Tokens
121
Input
223
Total output
208
Reasoning within output
commit_squash f009
100%
Passed
Tokens
116
Input
237
Total output
223
Reasoning within output
commit_squash f010
100%
Passed
Tokens
129
Input
197
Total output
179
Reasoning within output
commit_squash f011
100%
Passed
Tokens
137
Input
210
Total output
192
Reasoning within output
commit_squash f012
100%
Passed
Tokens
140
Input
244
Total output
221
Reasoning within output
git_bisect f001
100%
Passed
Tokens
188
Input
242
Total output
157
Reasoning within output
git_bisect f002
100%
Passed
Tokens
220
Input
218
Total output
118
Reasoning within output
git_bisect f003
100%
Passed
Tokens
220
Input
352
Total output
190
Reasoning within output
git_bisect f004
100%
Passed
Tokens
220
Input
316
Total output
172
Reasoning within output
git_bisect f005
100%
Passed
Tokens
216
Input
327
Total output
243
Reasoning within output
git_bisect f006
100%
Passed
Tokens
248
Input
531
Total output
329
Reasoning within output
git_bisect f007
100%
Passed
Tokens
218
Input
199
Total output
82
Reasoning within output
git_bisect f008
100%
Passed
Tokens
220
Input
239
Total output
131
Reasoning within output
git_bisect f009
100%
Passed
Tokens
216
Input
339
Total output
139
Reasoning within output
git_bisect f010
100%
Passed
Tokens
222
Input
395
Total output
208
Reasoning within output
git_bisect f011
100%
Passed
Tokens
224
Input
291
Total output
182
Reasoning within output
git_bisect f012
100%
Passed
Tokens
246
Input
253
Total output
161
Reasoning within output
git_clean f001
100%
Passed
Tokens
55
Input
400
Total output
393
Reasoning within output
git_clean f002
100%
Passed
Tokens
66
Input
87
Total output
80
Reasoning within output
git_clean f003
100%
Passed
Tokens
65
Input
120
Total output
113
Reasoning within output
git_clean f004
33.3%
Failed
Tokens
62
Input
548
Total output
539
Reasoning within output
git_clean f005
75%
Failed
Tokens
54
Input
380
Total output
373
Reasoning within output
git_clean f006
100%
Passed
Tokens
79
Input
353
Total output
343
Reasoning within output
git_clean f007
100%
Passed
Tokens
70
Input
742
Total output
729
Reasoning within output
git_clean f008
100%
Passed
Tokens
82
Input
73
Total output
66
Reasoning within output
git_clean f009
100%
Passed
Tokens
73
Input
437
Total output
427
Reasoning within output
git_clean f010
100%
Passed
Tokens
68
Input
109
Total output
102
Reasoning within output
git_clean f011
40%
Failed
Tokens
66
Input
2,380
Total output
2,360
Reasoning within output
git_clean f012
66.7%
Failed
Tokens
68
Input
252
Total output
242
Reasoning within output
git_grep f001
100%
Passed
Tokens
54
Input
91
Total output
85
Reasoning within output
git_grep f002
100%
Passed
Tokens
71
Input
94
Total output
90
Reasoning within output
git_grep f003
100%
Passed
Tokens
89
Input
82
Total output
78
Reasoning within output
git_grep f004
100%
Passed
Tokens
78
Input
116
Total output
108
Reasoning within output
git_grep f005
100%
Passed
Tokens
119
Input
150
Total output
146
Reasoning within output
git_grep f006
100%
Passed
Tokens
43
Input
214
Total output
210
Reasoning within output
git_grep f007
100%
Passed
Tokens
171
Input
193
Total output
188
Reasoning within output
git_grep f008
100%
Passed
Tokens
60
Input
94
Total output
90
Reasoning within output
git_grep f009
100%
Passed
Tokens
56
Input
51
Total output
45
Reasoning within output
git_grep f010
100%
Passed
Tokens
66
Input
313
Total output
309
Reasoning within output
git_grep f011
100%
Passed
Tokens
114
Input
117
Total output
113
Reasoning within output
git_grep f012
100%
Passed
Tokens
67
Input
79
Total output
68
Reasoning within output
git_log_format f001
100%
Passed
Tokens
910
Input
118
Total output
114
Reasoning within output
git_log_format f002
100%
Passed
Tokens
893
Input
124
Total output
114
Reasoning within output
git_log_format f003
100%
Passed
Tokens
733
Input
73
Total output
61
Reasoning within output
git_log_format f004
100%
Passed
Tokens
896
Input
76
Total output
72
Reasoning within output
git_log_format f005
100%
Passed
Tokens
739
Input
199
Total output
191
Reasoning within output
git_log_format f006
100%
Passed
Tokens
913
Input
303
Total output
299
Reasoning within output
git_log_format f007
100%
Passed
Tokens
913
Input
45
Total output
37
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,220
Input
128
Total output
124
Reasoning within output
git_log_format f009
100%
Passed
Tokens
751
Input
70
Total output
66
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,392
Input
83
Total output
79
Reasoning within output
git_log_format f011
100%
Passed
Tokens
440
Input
88
Total output
83
Reasoning within output
git_log_format f012
100%
Passed
Tokens
404
Input
108
Total output
104
Reasoning within output
git_show f001
100%
Passed
Tokens
201
Input
84
Total output
78
Reasoning within output
git_show f002
0%
Failed
Tokens
200
Input
397
Total output
360
Reasoning within output
git_show f003
100%
Passed
Tokens
683
Input
132
Total output
123
Reasoning within output
git_show f004
100%
Passed
Tokens
484
Input
85
Total output
76
Reasoning within output
git_show f005
100%
Passed
Tokens
208
Input
251
Total output
246
Reasoning within output
git_show f006
100%
Passed
Tokens
216
Input
63
Total output
59
Reasoning within output
git_show f007
100%
Passed
Tokens
204
Input
156
Total output
149
Reasoning within output
git_show f008
100%
Passed
Tokens
213
Input
123
Total output
84
Reasoning within output
git_show f009
100%
Passed
Tokens
328
Input
220
Total output
216
Reasoning within output
git_show f010
100%
Passed
Tokens
201
Input
123
Total output
119
Reasoning within output
git_show f011
100%
Passed
Tokens
229
Input
85
Total output
80
Reasoning within output
git_show f012
100%
Passed
Tokens
345
Input
110
Total output
106
Reasoning within output
merge_conflicts f001
0%
Failed
Tokens
88
Input
915
Total output
908
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
106
Input
521
Total output
511
Reasoning within output
merge_conflicts f003
100%
Passed
Tokens
81
Input
476
Total output
469
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
148
Input
362
Total output
334
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
140
Input
441
Total output
415
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
132
Input
322
Total output
292
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
136
Input
301
Total output
288
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
108
Input
157
Total output
148
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
105
Input
192
Total output
184
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
185
Input
424
Total output
393
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
149
Input
379
Total output
347
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
240
Input
1,030
Total output
976
Reasoning within output
rebase f001
100%
Passed
Tokens
102
Input
457
Total output
450
Reasoning within output
rebase f002
100%
Passed
Tokens
127
Input
185
Total output
175
Reasoning within output
rebase f003
100%
Passed
Tokens
108
Input
475
Total output
469
Reasoning within output
rebase f004
100%
Passed
Tokens
162
Input
441
Total output
413
Reasoning within output
rebase f005
100%
Passed
Tokens
146
Input
430
Total output
407
Reasoning within output
rebase f006
100%
Passed
Tokens
147
Input
247
Total output
222
Reasoning within output
rebase f007
100%
Passed
Tokens
139
Input
167
Total output
154
Reasoning within output
rebase f008
100%
Passed
Tokens
122
Input
127
Total output
118
Reasoning within output
rebase f009
100%
Passed
Tokens
106
Input
200
Total output
192
Reasoning within output
rebase f010
100%
Passed
Tokens
218
Input
451
Total output
420
Reasoning within output
rebase f011
100%
Passed
Tokens
161
Input
1,225
Total output
1,198
Reasoning within output
rebase f012
0%
Failed
Tokens
277
Input
606
Total output
560
Reasoning within output
reflog f001
100%
Passed
Tokens
266
Input
468
Total output
426
Reasoning within output
reflog f002
100%
Passed
Tokens
252
Input
811
Total output
409
Reasoning within output
reflog f003
100%
Passed
Tokens
224
Input
765
Total output
289
Reasoning within output
reflog f004
100%
Passed
Tokens
357
Input
1,658
Total output
1,184
Reasoning within output
reflog f005
100%
Passed
Tokens
423
Input
2,053
Total output
1,807
Reasoning within output
reflog f006
100%
Passed
Tokens
276
Input
568
Total output
340
Reasoning within output
reflog f007
100%
Passed
Tokens
296
Input
525
Total output
363
Reasoning within output
reflog f008
100%
Passed
Tokens
403
Input
1,125
Total output
664
Reasoning within output
reflog f009
100%
Passed
Tokens
363
Input
1,548
Total output
1,091
Reasoning within output
reflog f010
100%
Passed
Tokens
307
Input
860
Total output
465
Reasoning within output
reflog f011
100%
Passed
Tokens
356
Input
1,724
Total output
1,323
Reasoning within output
reflog f012
100%
Passed
Tokens
341
Input
1,109
Total output
516
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
142
Input
143
Total output
88
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
311
Input
351
Total output
201
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
129
Input
150
Total output
80
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
124
Input
255
Total output
135
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
136
Input
277
Total output
109
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
188
Input
176
Total output
93
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
219
Input
168
Total output
108
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
186
Input
234
Total output
111
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
216
Input
209
Total output
133
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
133
Input
160
Total output
68
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
127
Input
205
Total output
127
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
139
Input
229
Total output
109
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
46
Input
90
Total output
79
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
107
Input
179
Total output
169
Reasoning within output
submodule_usage f003
75%
Failed
Tokens
114
Input
720
Total output
702
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
100
Input
46
Total output
40
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
53
Input
144
Total output
119
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
95
Input
248
Total output
242
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
53
Input
146
Total output
126
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
103
Input
589
Total output
581
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
100
Input
318
Total output
312
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
106
Input
389
Total output
378
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
110
Input
64
Total output
57
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
51
Input
135
Total output
121
Reasoning within output
tag_management f001
100%
Passed
Tokens
64
Input
120
Total output
111
Reasoning within output
tag_management f002
100%
Passed
Tokens
63
Input
113
Total output
93
Reasoning within output
tag_management f003
100%
Passed
Tokens
70
Input
57
Total output
45
Reasoning within output
tag_management f004
100%
Passed
Tokens
77
Input
79
Total output
74
Reasoning within output
tag_management f005
100%
Passed
Tokens
88
Input
99
Total output
89
Reasoning within output
tag_management f006
100%
Passed
Tokens
74
Input
74
Total output
57
Reasoning within output
tag_management f007
100%
Passed
Tokens
70
Input
106
Total output
92
Reasoning within output
tag_management f008
100%
Passed
Tokens
73
Input
410
Total output
383
Reasoning within output
tag_management f009
100%
Passed
Tokens
76
Input
184
Total output
166
Reasoning within output
tag_management f010
100%
Passed
Tokens
54
Input
67
Total output
57
Reasoning within output
tag_management f011
100%
Passed
Tokens
56
Input
58
Total output
50
Reasoning within output
tag_management f012
100%
Passed
Tokens
98
Input
601
Total output
590
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
148
Input
112
Total output
100
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
151
Input
188
Total output
163
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
229
Input
153
Total output
141
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
232
Input
40
Total output
33
Reasoning within output
worktree_usage f005
100%
Passed
Tokens
142
Input
146
Total output
132
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
146
Input
125
Total output
110
Reasoning within output
worktree_usage f007
100%
Passed
Tokens
147
Input
140
Total output
118
Reasoning within output
worktree_usage f008
0%
Failed
Tokens
174
Input
503
Total output
455
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
264
Input
75
Total output
68
Reasoning within output
worktree_usage f010
0%
Failed
Tokens
237
Input
105
Total output
87
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
232
Input
54
Total output
43
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
142
Input
140
Total output
124
Reasoning within output