Prism-ml / ternary-bonsai-2-27b
low
93.1%
190 / 204 fixtures · 1 run(s)
41,310 input / 63,091 total output / 53,613 reasoning within output tokens $0.03464375
Reliability by Benchmark (Text)
Loading reliability summary…
Text vs JSON Schema Comparison
Pass Rate Delta
-3.4%
Text: 93.1% →
JSON: 89.7%
+5
Gained
JSON pass / text fail
−12
Lost
Text pass / JSON fail
178
Unchanged Pass
Both pass
9
Unchanged Fail
Both fail
Fixture Reliability Delta
| Fixture | Text | JSON | Delta |
|---|
Benchmark Deltas
| Benchmark | Text | JSON | Delta |
|---|---|---|---|
| branch_cleanup | 100% | 75% | -25% |
| cherry_pick | 91.7% | 75% | -16.7% |
| commit_squash | 100% | 83.3% | -16.7% |
| merge_conflicts | 75% | 91.7% | + 16.7% |
| worktree_usage | 75% | 91.7% | + 16.7% |
| submodule_usage | 91.7% | 83.3% | -8.3% |
| git_clean | 66.7% | 75% | + 8.3% |
| git_grep | 100% | 91.7% | -8.3% |
| git_show | 100% | 91.7% | -8.3% |
| reflog | 100% | 91.7% | -8.3% |
| tag_management | 100% | 91.7% | -8.3% |
| blame_forensics | 100% | 100% | + 0% |
| commit_messages | 100% | 100% | + 0% |
| git_bisect | 100% | 100% | + 0% |
| git_log_format | 100% | 100% | + 0% |
| rebase | 83.3% | 83.3% | + 0% |
| stash_recovery | 100% | 100% | + 0% |
Changed Fixtures (17)
Fixture Gallery (204)
blame_forensics f001
100%
Passed
Tokens
260
Input
675
Total output
667
Reasoning within output
blame_forensics f002
100%
Passed
Tokens
256
Input
346
Total output
339
Reasoning within output
blame_forensics f003
100%
Passed
Tokens
287
Input
190
Total output
181
Reasoning within output
blame_forensics f004
100%
Passed
Tokens
208
Input
258
Total output
250
Reasoning within output
blame_forensics f005
100%
Passed
Tokens
263
Input
903
Total output
897
Reasoning within output
blame_forensics f006
100%
Passed
Tokens
246
Input
422
Total output
416
Reasoning within output
blame_forensics f007
100%
Passed
Tokens
226
Input
145
Total output
137
Reasoning within output
blame_forensics f008
100%
Passed
Tokens
264
Input
322
Total output
315
Reasoning within output
blame_forensics f009
100%
Passed
Tokens
215
Input
383
Total output
376
Reasoning within output
blame_forensics f010
100%
Passed
Tokens
450
Input
501
Total output
495
Reasoning within output
blame_forensics f011
100%
Passed
Tokens
258
Input
250
Total output
244
Reasoning within output
blame_forensics f012
100%
Passed
Tokens
155
Input
371
Total output
363
Reasoning within output
branch_cleanup f001
100%
Passed
Tokens
111
Input
208
Total output
202
Reasoning within output
branch_cleanup f002
100%
Passed
Tokens
128
Input
155
Total output
147
Reasoning within output
branch_cleanup f003
100%
Passed
Tokens
102
Input
190
Total output
186
Reasoning within output
branch_cleanup f004
100%
Passed
Tokens
139
Input
132
Total output
124
Reasoning within output
branch_cleanup f005
100%
Passed
Tokens
112
Input
219
Total output
213
Reasoning within output
branch_cleanup f006
100%
Passed
Tokens
139
Input
211
Total output
199
Reasoning within output
branch_cleanup f007
100%
Passed
Tokens
112
Input
154
Total output
146
Reasoning within output
branch_cleanup f008
100%
Passed
Tokens
133
Input
122
Total output
110
Reasoning within output
branch_cleanup f009
100%
Passed
Tokens
87
Input
210
Total output
206
Reasoning within output
branch_cleanup f010
100%
Passed
Tokens
120
Input
156
Total output
148
Reasoning within output
branch_cleanup f011
100%
Passed
Tokens
75
Input
154
Total output
150
Reasoning within output
branch_cleanup f012
100%
Passed
Tokens
169
Input
221
Total output
208
Reasoning within output
cherry_pick f001
100%
Passed
Tokens
101
Input
252
Total output
245
Reasoning within output
cherry_pick f002
100%
Passed
Tokens
120
Input
289
Total output
279
Reasoning within output
cherry_pick f003
100%
Passed
Tokens
94
Input
310
Total output
303
Reasoning within output
cherry_pick f004
100%
Passed
Tokens
163
Input
655
Total output
627
Reasoning within output
cherry_pick f005
100%
Passed
Tokens
153
Input
229
Total output
203
Reasoning within output
cherry_pick f006
100%
Passed
Tokens
147
Input
286
Total output
256
Reasoning within output
cherry_pick f007
100%
Passed
Tokens
139
Input
232
Total output
219
Reasoning within output
cherry_pick f008
100%
Passed
Tokens
127
Input
120
Total output
111
Reasoning within output
cherry_pick f009
100%
Passed
Tokens
110
Input
116
Total output
108
Reasoning within output
cherry_pick f010
100%
Passed
Tokens
222
Input
322
Total output
291
Reasoning within output
cherry_pick f011
100%
Passed
Tokens
171
Input
435
Total output
408
Reasoning within output
cherry_pick f012
0%
Failed
Tokens
279
Input
592
Total output
532
Reasoning within output
commit_messages f001
95%
Passed
Tokens
117
Input
126
Total output
120
Reasoning within output
commit_messages f002
83.5%
Passed
Tokens
233
Input
171
Total output
158
Reasoning within output
commit_messages f003
98%
Passed
Tokens
88
Input
165
Total output
154
Reasoning within output
commit_messages f004
92%
Passed
Tokens
125
Input
133
Total output
127
Reasoning within output
commit_messages f005
92%
Passed
Tokens
109
Input
168
Total output
161
Reasoning within output
commit_messages f006
87%
Passed
Tokens
156
Input
186
Total output
177
Reasoning within output
commit_messages f007
67.5%
Passed
Tokens
197
Input
148
Total output
139
Reasoning within output
commit_messages f008
91%
Passed
Tokens
81
Input
146
Total output
139
Reasoning within output
commit_messages f009
82.5%
Passed
Tokens
135
Input
152
Total output
142
Reasoning within output
commit_messages f010
91.5%
Passed
Tokens
305
Input
296
Total output
284
Reasoning within output
commit_messages f011
82.5%
Passed
Tokens
154
Input
238
Total output
222
Reasoning within output
commit_messages f012
92.5%
Passed
Tokens
135
Input
146
Total output
138
Reasoning within output
commit_squash f001
100%
Passed
Tokens
144
Input
334
Total output
319
Reasoning within output
commit_squash f002
100%
Passed
Tokens
123
Input
263
Total output
252
Reasoning within output
commit_squash f003
100%
Passed
Tokens
138
Input
264
Total output
249
Reasoning within output
commit_squash f004
100%
Passed
Tokens
126
Input
354
Total output
340
Reasoning within output
commit_squash f005
100%
Passed
Tokens
123
Input
411
Total output
401
Reasoning within output
commit_squash f006
100%
Passed
Tokens
106
Input
272
Total output
258
Reasoning within output
commit_squash f007
100%
Passed
Tokens
112
Input
270
Total output
255
Reasoning within output
commit_squash f008
100%
Passed
Tokens
123
Input
175
Total output
160
Reasoning within output
commit_squash f009
100%
Passed
Tokens
114
Input
240
Total output
226
Reasoning within output
commit_squash f010
100%
Passed
Tokens
130
Input
251
Total output
233
Reasoning within output
commit_squash f011
100%
Passed
Tokens
135
Input
221
Total output
203
Reasoning within output
commit_squash f012
100%
Passed
Tokens
137
Input
264
Total output
241
Reasoning within output
git_bisect f001
100%
Passed
Tokens
194
Input
246
Total output
171
Reasoning within output
git_bisect f002
100%
Passed
Tokens
214
Input
264
Total output
131
Reasoning within output
git_bisect f003
100%
Passed
Tokens
218
Input
237
Total output
157
Reasoning within output
git_bisect f004
100%
Passed
Tokens
224
Input
373
Total output
230
Reasoning within output
git_bisect f005
100%
Passed
Tokens
216
Input
335
Total output
196
Reasoning within output
git_bisect f006
100%
Passed
Tokens
246
Input
241
Total output
166
Reasoning within output
git_bisect f007
100%
Passed
Tokens
216
Input
353
Total output
165
Reasoning within output
git_bisect f008
100%
Passed
Tokens
226
Input
308
Total output
137
Reasoning within output
git_bisect f009
100%
Passed
Tokens
216
Input
258
Total output
154
Reasoning within output
git_bisect f010
100%
Passed
Tokens
220
Input
278
Total output
154
Reasoning within output
git_bisect f011
100%
Passed
Tokens
218
Input
297
Total output
134
Reasoning within output
git_bisect f012
100%
Passed
Tokens
248
Input
259
Total output
160
Reasoning within output
git_clean f001
100%
Passed
Tokens
55
Input
248
Total output
241
Reasoning within output
git_clean f002
100%
Passed
Tokens
66
Input
140
Total output
133
Reasoning within output
git_clean f003
100%
Passed
Tokens
65
Input
114
Total output
107
Reasoning within output
git_clean f004
33.3%
Failed
Tokens
62
Input
1,527
Total output
1,519
Reasoning within output
git_clean f005
100%
Passed
Tokens
54
Input
314
Total output
307
Reasoning within output
git_clean f006
100%
Passed
Tokens
79
Input
120
Total output
110
Reasoning within output
git_clean f007
100%
Passed
Tokens
70
Input
108
Total output
91
Reasoning within output
git_clean f008
75%
Failed
Tokens
82
Input
126
Total output
119
Reasoning within output
git_clean f009
100%
Passed
Tokens
73
Input
114
Total output
104
Reasoning within output
git_clean f010
100%
Passed
Tokens
68
Input
99
Total output
92
Reasoning within output
git_clean f011
40%
Failed
Tokens
66
Input
2,200
Total output
2,191
Reasoning within output
git_clean f012
66.7%
Failed
Tokens
68
Input
270
Total output
260
Reasoning within output
git_grep f001
100%
Passed
Tokens
54
Input
94
Total output
88
Reasoning within output
git_grep f002
100%
Passed
Tokens
70
Input
96
Total output
92
Reasoning within output
git_grep f003
100%
Passed
Tokens
89
Input
90
Total output
86
Reasoning within output
git_grep f004
100%
Passed
Tokens
78
Input
110
Total output
102
Reasoning within output
git_grep f005
100%
Passed
Tokens
119
Input
146
Total output
142
Reasoning within output
git_grep f006
100%
Passed
Tokens
43
Input
162
Total output
158
Reasoning within output
git_grep f007
100%
Passed
Tokens
171
Input
1,360
Total output
1,355
Reasoning within output
git_grep f008
100%
Passed
Tokens
60
Input
81
Total output
77
Reasoning within output
git_grep f009
100%
Passed
Tokens
56
Input
62
Total output
56
Reasoning within output
git_grep f010
100%
Passed
Tokens
66
Input
241
Total output
237
Reasoning within output
git_grep f011
100%
Passed
Tokens
114
Input
115
Total output
111
Reasoning within output
git_grep f012
100%
Passed
Tokens
67
Input
114
Total output
103
Reasoning within output
git_log_format f001
100%
Passed
Tokens
911
Input
95
Total output
91
Reasoning within output
git_log_format f002
100%
Passed
Tokens
891
Input
138
Total output
128
Reasoning within output
git_log_format f003
100%
Passed
Tokens
731
Input
94
Total output
82
Reasoning within output
git_log_format f004
100%
Passed
Tokens
901
Input
93
Total output
89
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
240
Total output
236
Reasoning within output
git_log_format f007
100%
Passed
Tokens
916
Input
130
Total output
122
Reasoning within output
git_log_format f008
100%
Passed
Tokens
1,227
Input
122
Total output
118
Reasoning within output
git_log_format f009
100%
Passed
Tokens
735
Input
58
Total output
54
Reasoning within output
git_log_format f010
100%
Passed
Tokens
1,400
Input
111
Total output
107
Reasoning within output
git_log_format f011
100%
Passed
Tokens
441
Input
106
Total output
101
Reasoning within output
git_log_format f012
100%
Passed
Tokens
409
Input
114
Total output
110
Reasoning within output
git_show f001
100%
Passed
Tokens
201
Input
96
Total output
90
Reasoning within output
git_show f002
100%
Passed
Tokens
201
Input
791
Total output
786
Reasoning within output
git_show f003
100%
Passed
Tokens
693
Input
159
Total output
150
Reasoning within output
git_show f004
100%
Passed
Tokens
476
Input
85
Total output
76
Reasoning within output
git_show f005
100%
Passed
Tokens
209
Input
181
Total output
176
Reasoning within output
git_show f006
100%
Passed
Tokens
214
Input
77
Total output
73
Reasoning within output
git_show f007
100%
Passed
Tokens
204
Input
83
Total output
76
Reasoning within output
git_show f008
100%
Passed
Tokens
211
Input
120
Total output
83
Reasoning within output
git_show f009
100%
Passed
Tokens
327
Input
339
Total output
335
Reasoning within output
git_show f010
100%
Passed
Tokens
201
Input
115
Total output
111
Reasoning within output
git_show f011
100%
Passed
Tokens
229
Input
67
Total output
62
Reasoning within output
git_show f012
100%
Passed
Tokens
340
Input
129
Total output
125
Reasoning within output
merge_conflicts f001
0%
Failed
Tokens
88
Input
596
Total output
589
Reasoning within output
merge_conflicts f002
100%
Passed
Tokens
106
Input
325
Total output
315
Reasoning within output
merge_conflicts f003
0%
Failed
Tokens
81
Input
449
Total output
443
Reasoning within output
merge_conflicts f004
100%
Passed
Tokens
148
Input
412
Total output
384
Reasoning within output
merge_conflicts f005
100%
Passed
Tokens
140
Input
374
Total output
348
Reasoning within output
merge_conflicts f006
100%
Passed
Tokens
132
Input
350
Total output
321
Reasoning within output
merge_conflicts f007
100%
Passed
Tokens
136
Input
187
Total output
174
Reasoning within output
merge_conflicts f008
100%
Passed
Tokens
108
Input
133
Total output
124
Reasoning within output
merge_conflicts f009
100%
Passed
Tokens
105
Input
234
Total output
226
Reasoning within output
merge_conflicts f010
100%
Passed
Tokens
185
Input
329
Total output
298
Reasoning within output
merge_conflicts f011
100%
Passed
Tokens
149
Input
318
Total output
291
Reasoning within output
merge_conflicts f012
0%
Failed
Tokens
240
Input
439
Total output
385
Reasoning within output
rebase f001
0%
Failed
Tokens
102
Input
715
Total output
708
Reasoning within output
rebase f002
100%
Passed
Tokens
129
Input
255
Total output
245
Reasoning within output
rebase f003
100%
Passed
Tokens
108
Input
296
Total output
290
Reasoning within output
rebase f004
100%
Passed
Tokens
161
Input
456
Total output
428
Reasoning within output
rebase f005
100%
Passed
Tokens
148
Input
281
Total output
258
Reasoning within output
rebase f006
100%
Passed
Tokens
147
Input
374
Total output
344
Reasoning within output
rebase f007
100%
Passed
Tokens
140
Input
148
Total output
135
Reasoning within output
rebase f008
100%
Passed
Tokens
123
Input
199
Total output
190
Reasoning within output
rebase f009
100%
Passed
Tokens
106
Input
165
Total output
157
Reasoning within output
rebase f010
100%
Passed
Tokens
218
Input
366
Total output
335
Reasoning within output
rebase f011
100%
Passed
Tokens
161
Input
316
Total output
289
Reasoning within output
rebase f012
0%
Failed
Tokens
277
Input
915
Total output
861
Reasoning within output
reflog f001
100%
Passed
Tokens
248
Input
532
Total output
496
Reasoning within output
reflog f002
100%
Passed
Tokens
252
Input
609
Total output
202
Reasoning within output
reflog f003
100%
Passed
Tokens
218
Input
802
Total output
328
Reasoning within output
reflog f004
100%
Passed
Tokens
353
Input
1,618
Total output
1,147
Reasoning within output
reflog f005
100%
Passed
Tokens
418
Input
1,127
Total output
845
Reasoning within output
reflog f006
100%
Passed
Tokens
273
Input
821
Total output
298
Reasoning within output
reflog f007
100%
Passed
Tokens
307
Input
625
Total output
426
Reasoning within output
reflog f008
100%
Passed
Tokens
406
Input
1,635
Total output
1,168
Reasoning within output
reflog f009
100%
Passed
Tokens
343
Input
1,585
Total output
1,072
Reasoning within output
reflog f010
100%
Passed
Tokens
303
Input
638
Total output
241
Reasoning within output
reflog f011
100%
Passed
Tokens
357
Input
1,806
Total output
1,421
Reasoning within output
reflog f012
100%
Passed
Tokens
368
Input
1,830
Total output
1,236
Reasoning within output
stash_recovery f001
100%
Passed
Tokens
140
Input
155
Total output
66
Reasoning within output
stash_recovery f002
100%
Passed
Tokens
308
Input
230
Total output
161
Reasoning within output
stash_recovery f003
100%
Passed
Tokens
129
Input
172
Total output
81
Reasoning within output
stash_recovery f004
100%
Passed
Tokens
124
Input
210
Total output
99
Reasoning within output
stash_recovery f005
100%
Passed
Tokens
135
Input
191
Total output
121
Reasoning within output
stash_recovery f006
100%
Passed
Tokens
188
Input
189
Total output
100
Reasoning within output
stash_recovery f007
100%
Passed
Tokens
217
Input
141
Total output
74
Reasoning within output
stash_recovery f008
100%
Passed
Tokens
186
Input
258
Total output
139
Reasoning within output
stash_recovery f009
100%
Passed
Tokens
216
Input
200
Total output
132
Reasoning within output
stash_recovery f010
100%
Passed
Tokens
133
Input
206
Total output
91
Reasoning within output
stash_recovery f011
100%
Passed
Tokens
127
Input
218
Total output
104
Reasoning within output
stash_recovery f012
100%
Passed
Tokens
139
Input
264
Total output
133
Reasoning within output
submodule_usage f001
100%
Passed
Tokens
46
Input
104
Total output
93
Reasoning within output
submodule_usage f002
100%
Passed
Tokens
106
Input
427
Total output
416
Reasoning within output
submodule_usage f003
75%
Failed
Tokens
114
Input
863
Total output
831
Reasoning within output
submodule_usage f004
100%
Passed
Tokens
100
Input
52
Total output
46
Reasoning within output
submodule_usage f005
100%
Passed
Tokens
53
Input
294
Total output
268
Reasoning within output
submodule_usage f006
100%
Passed
Tokens
94
Input
188
Total output
182
Reasoning within output
submodule_usage f007
100%
Passed
Tokens
53
Input
123
Total output
103
Reasoning within output
submodule_usage f008
100%
Passed
Tokens
103
Input
524
Total output
514
Reasoning within output
submodule_usage f009
100%
Passed
Tokens
101
Input
225
Total output
219
Reasoning within output
submodule_usage f010
100%
Passed
Tokens
109
Input
231
Total output
222
Reasoning within output
submodule_usage f011
100%
Passed
Tokens
108
Input
44
Total output
37
Reasoning within output
submodule_usage f012
100%
Passed
Tokens
51
Input
143
Total output
128
Reasoning within output
tag_management f001
100%
Passed
Tokens
67
Input
135
Total output
126
Reasoning within output
tag_management f002
100%
Passed
Tokens
63
Input
172
Total output
151
Reasoning within output
tag_management f003
100%
Passed
Tokens
68
Input
68
Total output
56
Reasoning within output
tag_management f004
100%
Passed
Tokens
78
Input
72
Total output
67
Reasoning within output
tag_management f005
100%
Passed
Tokens
92
Input
83
Total output
73
Reasoning within output
tag_management f006
100%
Passed
Tokens
72
Input
99
Total output
84
Reasoning within output
tag_management f007
100%
Passed
Tokens
70
Input
366
Total output
348
Reasoning within output
tag_management f008
100%
Passed
Tokens
74
Input
290
Total output
263
Reasoning within output
tag_management f009
100%
Passed
Tokens
76
Input
140
Total output
122
Reasoning within output
tag_management f010
100%
Passed
Tokens
54
Input
62
Total output
52
Reasoning within output
tag_management f011
100%
Passed
Tokens
57
Input
64
Total output
56
Reasoning within output
tag_management f012
100%
Passed
Tokens
97
Input
522
Total output
513
Reasoning within output
worktree_usage f001
100%
Passed
Tokens
148
Input
91
Total output
79
Reasoning within output
worktree_usage f002
100%
Passed
Tokens
157
Input
267
Total output
242
Reasoning within output
worktree_usage f003
100%
Passed
Tokens
227
Input
76
Total output
64
Reasoning within output
worktree_usage f004
100%
Passed
Tokens
226
Input
72
Total output
65
Reasoning within output
worktree_usage f005
33.3%
Failed
Tokens
139
Input
122
Total output
110
Reasoning within output
worktree_usage f006
100%
Passed
Tokens
144
Input
168
Total output
152
Reasoning within output
worktree_usage f007
0%
Failed
Tokens
149
Input
267
Total output
241
Reasoning within output
worktree_usage f008
0%
Failed
Tokens
168
Input
289
Total output
235
Reasoning within output
worktree_usage f009
100%
Passed
Tokens
259
Input
71
Total output
64
Reasoning within output
worktree_usage f010
100%
Passed
Tokens
235
Input
123
Total output
105
Reasoning within output
worktree_usage f011
100%
Passed
Tokens
225
Input
92
Total output
81
Reasoning within output
worktree_usage f012
100%
Passed
Tokens
143
Input
96
Total output
79
Reasoning within output