@lucasschirm/devinp 0.1.2 → 0.1.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (4) hide show
  1. package/README.md +3 -2
  2. package/bin/command.js +3 -3
  3. package/data/benchlm-scores.json +69 -41
  4. package/package.json +1 -1
package/README.md CHANGED
@@ -54,8 +54,9 @@ devinp --json # emit normalized rows as JSON
54
54
 
55
55
  - Prices are USD per 1M tokens; columns are heat-colored green → red.
56
56
  - **Value** = `Bench % ÷ output $/1M` — the score you buy per output dollar.
57
- Free models show `∞`; models without a Bench score show `—` and sort last
58
- under `-o=bench` / `-o=value`.
57
+ Free models always show `∞` and sort first under `-o=bench` / `-o=value`,
58
+ even without a Bench score. Paid models without a score show `—` and sort
59
+ last.
59
60
  - Markers: `β` = beta, `●` = new, magenta name = free. `Sidekick` shows the
60
61
  partner-model pricing for Fusion rows.
61
62
  - Colors disable automatically when piping; `NO_COLOR` / `FORCE_COLOR` are
package/bin/command.js CHANGED
@@ -54666,9 +54666,9 @@ function buildRows(data, scores = {}) {
54666
54666
  return [...rows.values()];
54667
54667
  }
54668
54668
  function computeValue(score, output) {
54669
+ if (output === 0) return Number.POSITIVE_INFINITY;
54669
54670
  if (score == null) return null;
54670
54671
  if (output == null) return null;
54671
- if (output === 0) return Number.POSITIVE_INFINITY;
54672
54672
  return score / output;
54673
54673
  }
54674
54674
  var priceAsc = (field) => (a2, b) => (a2[field] ?? Number.POSITIVE_INFINITY) - (b[field] ?? Number.POSITIVE_INFINITY) || (a2.output ?? Number.POSITIVE_INFINITY) - (b.output ?? Number.POSITIVE_INFINITY) || a2.label.localeCompare(b.label);
@@ -54676,7 +54676,7 @@ var SORTS = {
54676
54676
  input: priceAsc("input"),
54677
54677
  cached: priceAsc("cached"),
54678
54678
  output: priceAsc("output"),
54679
- bench: (a2, b) => (b.bench?.score ?? -1) - (a2.bench?.score ?? -1) || (a2.output ?? Number.POSITIVE_INFINITY) - (b.output ?? Number.POSITIVE_INFINITY) || a2.label.localeCompare(b.label),
54679
+ bench: (a2, b) => (b.free ? 1 : 0) - (a2.free ? 1 : 0) || (b.bench?.score ?? -1) - (a2.bench?.score ?? -1) || (a2.output ?? Number.POSITIVE_INFINITY) - (b.output ?? Number.POSITIVE_INFINITY) || a2.label.localeCompare(b.label),
54680
54680
  value: (a2, b) => (b.value ?? -1) - (a2.value ?? -1) || (b.bench?.score ?? -1) - (a2.bench?.score ?? -1) || a2.label.localeCompare(b.label)
54681
54681
  };
54682
54682
  var SORT_IDS = Object.keys(SORTS);
@@ -54946,7 +54946,7 @@ async function main() {
54946
54946
  input: (a, b) => (a.input ?? 1e9) - (b.input ?? 1e9) || (a.output ?? 1e9) - (b.output ?? 1e9) || a.label.localeCompare(b.label),
54947
54947
  cached: (a, b) => (a.cached ?? 1e9) - (b.cached ?? 1e9) || (a.output ?? 1e9) - (b.output ?? 1e9) || a.label.localeCompare(b.label),
54948
54948
  output: (a, b) => (a.output ?? 1e9) - (b.output ?? 1e9) || a.label.localeCompare(b.label),
54949
- bench: (a, b) => (b.bench?.score ?? -1) - (a.bench?.score ?? -1) || (a.output ?? 1e9) - (b.output ?? 1e9) || a.label.localeCompare(b.label),
54949
+ bench: (a, b) => (b.free ? 1 : 0) - (a.free ? 1 : 0) || (b.bench?.score ?? -1) - (a.bench?.score ?? -1) || (a.output ?? 1e9) - (b.output ?? 1e9) || a.label.localeCompare(b.label),
54950
54950
  value: (a, b) => (b.value ?? -1) - (a.value ?? -1) || (b.bench?.score ?? -1) - (a.bench?.score ?? -1) || a.label.localeCompare(b.label),
54951
54951
  };
54952
54952
  const GROUPS = {
package/data/benchlm-scores.json CHANGED
@@ -1,206 +1,231 @@
1
1
  {
2
- "updated": "2026-09-22",
2
+ "updated": "2026-09-29",
3
3
  "source": "https://benchlm.ai/api/data/leaderboard?category=coding&limit=200",
4
4
  "models": {
5
5
  "claude-fable-5": {
6
- "score": 74.47,
6
+ "score": 73.1,
7
7
  "benchName": "Claude Fable 5",
8
8
  "supported": true
9
9
  },
10
10
  "claude-fable-5.1": {
11
- "score": 81.29,
11
+ "score": 79.99,
12
12
  "benchName": "Claude Fable 5.1",
13
13
  "supported": true
14
14
  },
15
15
  "claude-haiku-4.5": {
16
- "score": 19.77,
16
+ "score": 19.61,
17
17
  "benchName": "Claude Haiku 4.5",
18
18
  "supported": true
19
19
  },
20
20
  "claude-opus-4.5": {
21
- "score": 46.1,
21
+ "score": 42.56,
22
22
  "benchName": "Claude Opus 4.5",
23
23
  "supported": false
24
24
  },
25
25
  "claude-opus-4.6": {
26
- "score": 51.32,
26
+ "score": 49.68,
27
27
  "benchName": "Claude Opus 4.6",
28
28
  "supported": true
29
29
  },
30
30
  "claude-opus-4.7": {
31
- "score": 59.77,
31
+ "score": 58.1,
32
32
  "benchName": "Claude Opus 4.7",
33
33
  "supported": true
34
34
  },
35
35
  "claude-opus-4.8": {
36
- "score": 63.58,
36
+ "score": 62.37,
37
37
  "benchName": "Claude Opus 4.8",
38
38
  "supported": true
39
39
  },
40
40
  "claude-opus-5": {
41
- "score": 72.91,
41
+ "score": 72.18,
42
42
  "benchName": "Claude Opus 5",
43
43
  "supported": true
44
44
  },
45
45
  "claude-opus-5.5": {
46
- "score": 86.85,
46
+ "score": 83.6,
47
47
  "benchName": "Claude Opus 5.5",
48
48
  "supported": true
49
49
  },
50
50
  "claude-sonnet-4.6": {
51
- "score": 48.52,
51
+ "score": 47.04,
52
52
  "benchName": "Claude Sonnet 4.6",
53
53
  "supported": true
54
54
  },
55
55
  "claude-sonnet-5": {
56
- "score": 61.05,
56
+ "score": 59.96,
57
57
  "benchName": "Claude Sonnet 5",
58
58
  "supported": true
59
59
  },
60
+ "claude-sonnet-5.5": {
61
+ "score": 85.12,
62
+ "benchName": "Claude Sonnet 5.5",
63
+ "supported": true
64
+ },
65
+ "deepseek-v4-pro": {
66
+ "score": 49.36,
67
+ "benchName": "DeepSeek V4 Pro 0813",
68
+ "supported": true
69
+ },
70
+ "deepseek-v4.1-flash": {
71
+ "score": 49.63,
72
+ "benchName": "DeepSeek V4.1 Flash",
73
+ "supported": false
74
+ },
60
75
  "gemini-3-flash": {
61
- "score": 36.84,
76
+ "score": 36.38,
62
77
  "benchName": "Gemini 3 Flash",
63
78
  "supported": true
64
79
  },
65
80
  "gemini-3.1-pro": {
66
- "score": 43.36,
81
+ "score": 42.44,
67
82
  "benchName": "Gemini 3.1 Pro",
68
83
  "supported": true
69
84
  },
70
85
  "gemini-3.5-flash": {
71
- "score": 53.7,
86
+ "score": 52.43,
72
87
  "benchName": "Gemini 3.5 Flash",
73
88
  "supported": true
74
89
  },
75
90
  "gemini-3.6-flash": {
76
- "score": 54.73,
91
+ "score": 53.78,
77
92
  "benchName": "Gemini 3.6 Flash",
78
93
  "supported": true
79
94
  },
80
95
  "gemini-3.7-flash": {
81
- "score": 59.79,
96
+ "score": 59.48,
82
97
  "benchName": "Gemini 3.7 Flash",
83
98
  "supported": true
84
99
  },
85
100
  "gemini-3.8-flash": {
86
- "score": 64.14,
101
+ "score": 64.16,
87
102
  "benchName": "Gemini 3.8 Flash",
88
103
  "supported": true
89
104
  },
90
105
  "glm-5.2": {
91
- "score": 57.32,
106
+ "score": 56.29,
92
107
  "benchName": "GLM-5.2",
93
108
  "supported": true
94
109
  },
95
110
  "glm-5.3": {
96
- "score": 58.15,
111
+ "score": 56.84,
97
112
  "benchName": "GLM-5.3",
98
113
  "supported": true
99
114
  },
100
115
  "glm-5.3-flash": {
101
- "score": 51.6,
116
+ "score": 52.54,
102
117
  "benchName": "GLM-5.3-Flash",
103
118
  "supported": true
104
119
  },
105
120
  "gpt-5.1": {
106
- "score": 39.77,
121
+ "score": 38.86,
107
122
  "benchName": "GPT-5.1",
108
123
  "supported": true
109
124
  },
110
125
  "gpt-5.2": {
111
- "score": 41.57,
126
+ "score": 39.7,
112
127
  "benchName": "GPT-5.2",
113
128
  "supported": true
114
129
  },
115
130
  "gpt-5.3-codex": {
116
- "score": 56.8,
131
+ "score": 55.5,
117
132
  "benchName": "GPT-5.3 Codex",
118
133
  "supported": true
119
134
  },
120
135
  "gpt-5.4": {
121
- "score": 50.93,
136
+ "score": 49.83,
122
137
  "benchName": "GPT-5.4",
123
138
  "supported": true
124
139
  },
125
140
  "gpt-5.4-mini": {
126
- "score": 38.61,
141
+ "score": 36.92,
127
142
  "benchName": "GPT-5.4 mini",
128
143
  "supported": true
129
144
  },
130
145
  "gpt-5.5": {
131
- "score": 64.47,
146
+ "score": 62.83,
132
147
  "benchName": "GPT-5.5",
133
148
  "supported": true
134
149
  },
135
150
  "gpt-5.6-luna": {
136
- "score": 64.72,
151
+ "score": 63.3,
137
152
  "benchName": "GPT-5.6 Luna",
138
153
  "supported": true
139
154
  },
140
155
  "gpt-5.6-sol": {
141
- "score": 71.77,
156
+ "score": 70.53,
142
157
  "benchName": "GPT-5.6 Sol",
143
158
  "supported": true
144
159
  },
145
160
  "gpt-5.6-terra": {
146
- "score": 65.05,
161
+ "score": 64.41,
147
162
  "benchName": "GPT-5.6 Terra",
148
163
  "supported": true
149
164
  },
150
165
  "gpt-6-astra": {
151
- "score": 74.43,
166
+ "score": 74.18,
152
167
  "benchName": "GPT-6 Astra",
153
168
  "supported": true
154
169
  },
155
170
  "gpt-6-luna": {
156
- "score": 59.93,
171
+ "score": 51.37,
157
172
  "benchName": "GPT-6 Luna",
158
173
  "supported": true
159
174
  },
160
175
  "gpt-6-sol": {
161
- "score": 65.48,
176
+ "score": 62.99,
162
177
  "benchName": "GPT-6 Sol",
163
178
  "supported": true
164
179
  },
180
+ "gpt-6.1-sol": {
181
+ "score": 66.97,
182
+ "benchName": "GPT-6.1 Sol",
183
+ "supported": true
184
+ },
165
185
  "grok-4.5": {
166
- "score": 58.55,
186
+ "score": 58.07,
167
187
  "benchName": "Grok 4.5",
168
188
  "supported": true
169
189
  },
170
190
  "grok-4.6": {
171
- "score": 62.55,
191
+ "score": 61.62,
172
192
  "benchName": "Grok 4.6",
173
193
  "supported": true
174
194
  },
175
195
  "inkling": {
176
- "score": 37.23,
196
+ "score": 36.56,
177
197
  "benchName": "Inkling",
178
198
  "supported": true
179
199
  },
180
200
  "kimi-k2.6": {
181
- "score": 47.36,
201
+ "score": 46.36,
182
202
  "benchName": "Kimi K2.6",
183
203
  "supported": true
184
204
  },
205
+ "kimi-k2.7": {
206
+ "score": 44.45,
207
+ "benchName": "Kimi K2.7 Code",
208
+ "supported": true
209
+ },
185
210
  "kimi-k3": {
186
- "score": 63.31,
211
+ "score": 61.59,
187
212
  "benchName": "Kimi K3",
188
213
  "supported": true
189
214
  },
190
215
  "nemotron-3-ultra": {
191
- "score": 23.53,
216
+ "score": 22.78,
192
217
  "benchName": "Nemotron 3 Ultra",
193
218
  "supported": true
194
219
  }
195
220
  },
196
221
  "unmatched": [
197
222
  "Apodex 1.1",
223
+ "Celeris-1",
198
224
  "Claude 3 Opus",
199
225
  "Command A+",
200
226
  "DeepSeek V3",
201
227
  "DeepSeek V3.2",
202
228
  "DeepSeek V3.2 (Thinking)",
203
- "DeepSeek V4 Pro 0813",
204
229
  "GLM-4.6",
205
230
  "GLM-4.7",
206
231
  "GLM-5",
@@ -241,7 +266,6 @@
241
266
  "Inkling-Small",
242
267
  "K-Exaone",
243
268
  "Kimi K2.5",
244
- "Kimi K2.7 Code",
245
269
  "LFM2.5-2.6B",
246
270
  "Laguna M.1",
247
271
  "Laguna XS.2",
@@ -255,6 +279,8 @@
255
279
  "MiMo-V2-Pro",
256
280
  "MiMo-V2.5",
257
281
  "MiMo-V2.5-Pro",
282
+ "MiMo-V2.6-Flash",
283
+ "MiMo-V2.6-Pro",
258
284
  "MiniMax M2.5",
259
285
  "MiniMax M2.7",
260
286
  "MiniMax M3",
@@ -272,6 +298,7 @@
272
298
  "Nemotron 3.5 Lightning 30B A3B NVFP4",
273
299
  "Ornith-1.5-35B-A3B",
274
300
  "Ornith-1.5-397B",
301
+ "Quasar 438B",
275
302
  "Qwen 3.6 Max (preview)",
276
303
  "Qwen3.5 Flash",
277
304
  "Qwen3.5-122B-A10B",
@@ -286,6 +313,7 @@
286
313
  "Qwen3.8-27B",
287
314
  "Qwen3.8-Flash-Next",
288
315
  "Step 3.7 Flash",
316
+ "Step 5 Preview",
289
317
  "Trinity-Large-Thinking",
290
318
  "o1",
291
319
  "o1-preview"
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@lucasschirm/devinp",
3
- "version": "0.1.2",
3
+ "version": "0.1.4",
4
4
  "description": "Pretty pricing table for Devin models, powered by `devin models list`",
5
5
  "type": "module",
6
6
  "bin": {