pi-free 2.2.4 → 2.2.6

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 (35) hide show
  1. package/CHANGELOG.md +24 -0
  2. package/README.md +1 -1
  3. package/banner.svg +1 -1
  4. package/config.ts +58 -29
  5. package/constants.ts +1 -4
  6. package/index.ts +30 -24
  7. package/lib/built-in-toggle.ts +185 -18
  8. package/lib/provider-cache.ts +22 -0
  9. package/package.json +74 -74
  10. package/provider-failover/benchmark-lookup.ts +4 -1
  11. package/provider-failover/benchmarks-chunk-0.ts +570 -570
  12. package/provider-failover/benchmarks-chunk-1.ts +676 -676
  13. package/provider-failover/benchmarks-chunk-2.ts +673 -673
  14. package/provider-failover/benchmarks-chunk-3.ts +680 -680
  15. package/provider-failover/benchmarks-chunk-4.ts +683 -683
  16. package/provider-failover/benchmarks-chunk-5.ts +816 -474
  17. package/provider-helper.ts +64 -0
  18. package/providers/bai/bai.ts +8 -2
  19. package/providers/crofai/crofai.ts +8 -2
  20. package/providers/deepinfra/deepinfra.ts +8 -2
  21. package/providers/dynamic-built-in/index.ts +36 -26
  22. package/providers/kilo/kilo.ts +41 -22
  23. package/providers/novita/novita.ts +8 -2
  24. package/providers/openmodel/openmodel.ts +8 -2
  25. package/providers/qoder/models.ts +63 -175
  26. package/providers/qoder/qoder.ts +49 -84
  27. package/providers/qoder/stream.ts +182 -274
  28. package/providers/qoder/transform.ts +5 -2
  29. package/providers/routeway/routeway.ts +8 -2
  30. package/providers/sambanova/sambanova.ts +12 -6
  31. package/providers/together/together.ts +8 -2
  32. package/providers/tokenrouter/tokenrouter.ts +8 -2
  33. package/providers/zenmux/zenmux.ts +8 -2
  34. package/providers/codestral/codestral.ts +0 -128
  35. package/providers/qoder/encoding.ts +0 -48
package/provider-failover/benchmarks-chunk-2.ts CHANGED
@@ -1,20 +1,20 @@
1
1
  // Auto-generated benchmark data chunk 2
2
- // Models: trinity-large-thinking .. o3-mini (90 entries)
3
- // Last updated: 2026-06-01
2
+ // Models: k2-think-v2 .. gpt-5-nano-high (90 entries)
3
+ // Last updated: 2026-07-01
4
4
  // DO NOT EDIT MANUALLY — generated by scripts/update-benchmarks.ts
5
5
 
6
6
  import type { HardcodedBenchmark } from "./hardcoded-benchmarks.ts";
7
7
 
8
8
  export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
9
- "trinity-large-thinking": {
9
+ "k2-think-v2": {
10
10
  // AA specific benchmarks
11
- codingIndex: 27.2,
11
+ codingIndex: 21,
12
12
  mathIndex: undefined,
13
13
 
14
14
  // Academic benchmarks
15
15
  mmluPro: undefined,
16
- gpqa: 0.752,
17
- hle: 0.147,
16
+ gpqa: 0.713,
17
+ hle: 0.095,
18
18
 
19
19
  // Capabilities
20
20
  contextWindow: 8192,
@@ -22,18 +22,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
22
22
  supportsVision: false,
23
23
 
24
24
  // Metadata
25
- lastUpdated: "2026-06-01",
26
- originalModel: "Trinity Large Thinking",
25
+ lastUpdated: "2026-07-01",
26
+ originalModel: "K2 Think V2",
27
27
  },
28
- "jt-mini": {
28
+ "k2-v2-low": {
29
29
  // AA specific benchmarks
30
- codingIndex: 21.2,
31
- mathIndex: undefined,
30
+ codingIndex: undefined,
31
+ mathIndex: 35.3,
32
32
 
33
33
  // Academic benchmarks
34
- mmluPro: undefined,
35
- gpqa: 0.676,
36
- hle: 0.066,
34
+ mmluPro: 0.713,
35
+ gpqa: 0.541,
36
+ hle: 0.039,
37
37
 
38
38
  // Capabilities
39
39
  contextWindow: 8192,
@@ -41,18 +41,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
41
41
  supportsVision: false,
42
42
 
43
43
  // Metadata
44
- lastUpdated: "2026-06-01",
45
- originalModel: "JT-MINI",
44
+ lastUpdated: "2026-07-01",
45
+ originalModel: "K2-V2 (low)",
46
46
  },
47
- "jt-35b-flash": {
47
+ "mi-dm-k-2.5-pro": {
48
48
  // AA specific benchmarks
49
- codingIndex: 28.9,
50
- mathIndex: undefined,
49
+ codingIndex: undefined,
50
+ mathIndex: 76.7,
51
51
 
52
52
  // Academic benchmarks
53
- mmluPro: undefined,
54
- gpqa: 0.829,
55
- hle: 0.061,
53
+ mmluPro: 0.809,
54
+ gpqa: 0.701,
55
+ hle: 0.077,
56
56
 
57
57
  // Capabilities
58
58
  contextWindow: 8192,
@@ -60,18 +60,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
60
60
  supportsVision: false,
61
61
 
62
62
  // Metadata
63
- lastUpdated: "2026-06-01",
64
- originalModel: "JT-35B-Flash",
63
+ lastUpdated: "2026-07-01",
64
+ originalModel: "Mi:dm K 2.5 Pro",
65
65
  },
66
- "glm-5.1-reasoning": {
66
+ "hyperclova-x-seed-think-32b": {
67
67
  // AA specific benchmarks
68
- codingIndex: 43.4,
69
- mathIndex: undefined,
68
+ codingIndex: undefined,
69
+ mathIndex: 59,
70
70
 
71
71
  // Academic benchmarks
72
- mmluPro: undefined,
73
- gpqa: 0.868,
74
- hle: 0.28,
72
+ mmluPro: 0.785,
73
+ gpqa: 0.615,
74
+ hle: 0.055,
75
75
 
76
76
  // Capabilities
77
77
  contextWindow: 8192,
@@ -79,18 +79,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
79
79
  supportsVision: false,
80
80
 
81
81
  // Metadata
82
- lastUpdated: "2026-06-01",
83
- originalModel: "GLM-5.1 (Reasoning)",
82
+ lastUpdated: "2026-07-01",
83
+ originalModel: "HyperCLOVA X SEED Think (32B)",
84
84
  },
85
- "glm-5.1-non-reasoning": {
85
+ "longcat-flash-lite": {
86
86
  // AA specific benchmarks
87
- codingIndex: 35.8,
87
+ codingIndex: undefined,
88
88
  mathIndex: undefined,
89
89
 
90
90
  // Academic benchmarks
91
91
  mmluPro: undefined,
92
- gpqa: 0.839,
93
- hle: 0.256,
92
+ gpqa: 0.636,
93
+ hle: 0.06,
94
94
 
95
95
  // Capabilities
96
96
  contextWindow: 8192,
@@ -98,18 +98,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
98
98
  supportsVision: false,
99
99
 
100
100
  // Metadata
101
- lastUpdated: "2026-06-01",
102
- originalModel: "GLM-5.1 (Non-reasoning)",
101
+ lastUpdated: "2026-07-01",
102
+ originalModel: "LongCat Flash Lite",
103
103
  },
104
- "glm-5v-turbo-reasoning": {
104
+ "tri-21b-think-preview": {
105
105
  // AA specific benchmarks
106
- codingIndex: 36.2,
106
+ codingIndex: undefined,
107
107
  mathIndex: undefined,
108
108
 
109
109
  // Academic benchmarks
110
110
  mmluPro: undefined,
111
- gpqa: 0.809,
112
- hle: 0.158,
111
+ gpqa: 0.538,
112
+ hle: 0.057,
113
113
 
114
114
  // Capabilities
115
115
  contextWindow: 8192,
@@ -117,18 +117,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
117
117
  supportsVision: false,
118
118
 
119
119
  // Metadata
120
- lastUpdated: "2026-06-01",
121
- originalModel: "GLM 5V Turbo (Reasoning)",
120
+ lastUpdated: "2026-07-01",
121
+ originalModel: "Tri-21B-think Preview",
122
122
  },
123
- "glm-5-turbo": {
123
+ "tri-21b-think": {
124
124
  // AA specific benchmarks
125
- codingIndex: 36.8,
125
+ codingIndex: undefined,
126
126
  mathIndex: undefined,
127
127
 
128
128
  // Academic benchmarks
129
129
  mmluPro: undefined,
130
- gpqa: 0.847,
131
- hle: 0.254,
130
+ gpqa: 0.601,
131
+ hle: 0.061,
132
132
 
133
133
  // Capabilities
134
134
  contextWindow: 8192,
@@ -136,18 +136,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
136
136
  supportsVision: false,
137
137
 
138
138
  // Metadata
139
- lastUpdated: "2026-06-01",
140
- originalModel: "GLM-5-Turbo",
139
+ lastUpdated: "2026-07-01",
140
+ originalModel: "Tri-21B-Think",
141
141
  },
142
- "command-a": {
142
+ "nanbeige4.1-3b": {
143
143
  // AA specific benchmarks
144
- codingIndex: 9.9,
145
- mathIndex: 13,
144
+ codingIndex: undefined,
145
+ mathIndex: undefined,
146
146
 
147
147
  // Academic benchmarks
148
- mmluPro: 0.712,
149
- gpqa: 0.527,
150
- hle: 0.046,
148
+ mmluPro: undefined,
149
+ gpqa: 0.849,
150
+ hle: 0.1,
151
151
 
152
152
  // Capabilities
153
153
  contextWindow: 8192,
@@ -155,18 +155,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
155
155
  supportsVision: false,
156
156
 
157
157
  // Metadata
158
- lastUpdated: "2026-06-01",
159
- originalModel: "Command A",
158
+ lastUpdated: "2026-07-01",
159
+ originalModel: "Nanbeige4.1-3B",
160
160
  },
161
- "tiny-aya-global": {
161
+ "apertus-8b-instruct": {
162
162
  // AA specific benchmarks
163
- codingIndex: 1.2,
163
+ codingIndex: undefined,
164
164
  mathIndex: undefined,
165
165
 
166
166
  // Academic benchmarks
167
167
  mmluPro: undefined,
168
- gpqa: 0.305,
169
- hle: 0.052,
168
+ gpqa: 0.256,
169
+ hle: 0.05,
170
170
 
171
171
  // Capabilities
172
172
  contextWindow: 8192,
@@ -174,18 +174,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
174
174
  supportsVision: false,
175
175
 
176
176
  // Metadata
177
- lastUpdated: "2026-06-01",
178
- originalModel: "Tiny Aya Global",
177
+ lastUpdated: "2026-07-01",
178
+ originalModel: "Apertus 8B Instruct",
179
179
  },
180
- "command-a+": {
180
+ "apertus-70b-instruct": {
181
181
  // AA specific benchmarks
182
- codingIndex: 29.3,
182
+ codingIndex: undefined,
183
183
  mathIndex: undefined,
184
184
 
185
185
  // Academic benchmarks
186
186
  mmluPro: undefined,
187
- gpqa: 0.761,
188
- hle: 0.114,
187
+ gpqa: 0.272,
188
+ hle: 0.055,
189
189
 
190
190
  // Capabilities
191
191
  contextWindow: 8192,
@@ -193,18 +193,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
193
193
  supportsVision: false,
194
194
 
195
195
  // Metadata
196
- lastUpdated: "2026-06-01",
197
- originalModel: "Command A+",
196
+ lastUpdated: "2026-07-01",
197
+ originalModel: "Apertus 70B Instruct",
198
198
  },
199
- "apriel-v1.6-15b-thinker": {
199
+ "minicpm5-1b-non-reasoning": {
200
200
  // AA specific benchmarks
201
- codingIndex: 22,
202
- mathIndex: 88,
201
+ codingIndex: undefined,
202
+ mathIndex: undefined,
203
203
 
204
204
  // Academic benchmarks
205
- mmluPro: 0.79,
206
- gpqa: 0.733,
207
- hle: 0.098,
205
+ mmluPro: undefined,
206
+ gpqa: 0.269,
207
+ hle: 0.046,
208
208
 
209
209
  // Capabilities
210
210
  contextWindow: 8192,
@@ -212,18 +212,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
212
212
  supportsVision: false,
213
213
 
214
214
  // Metadata
215
- lastUpdated: "2026-06-01",
216
- originalModel: "Apriel-v1.6-15B-Thinker",
215
+ lastUpdated: "2026-07-01",
216
+ originalModel: "MiniCPM5-1B (Non-reasoning)",
217
217
  },
218
- "jamba-1.7-large": {
218
+ "minicpm-v-4.6-1.3b": {
219
219
  // AA specific benchmarks
220
- codingIndex: 7.8,
221
- mathIndex: 2.3,
220
+ codingIndex: 0.7,
221
+ mathIndex: undefined,
222
222
 
223
223
  // Academic benchmarks
224
- mmluPro: 0.577,
225
- gpqa: 0.39,
226
- hle: 0.038,
224
+ mmluPro: undefined,
225
+ gpqa: 0.305,
226
+ hle: 0.049,
227
227
 
228
228
  // Capabilities
229
229
  contextWindow: 8192,
@@ -231,18 +231,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
231
231
  supportsVision: false,
232
232
 
233
233
  // Metadata
234
- lastUpdated: "2026-06-01",
235
- originalModel: "Jamba 1.7 Large",
234
+ lastUpdated: "2026-07-01",
235
+ originalModel: "MiniCPM-V 4.6 1.3B",
236
236
  },
237
- "jamba-1.7-mini": {
237
+ "minicpm5-1b-reasoning": {
238
238
  // AA specific benchmarks
239
- codingIndex: 3.1,
240
- mathIndex: 0.3,
239
+ codingIndex: undefined,
240
+ mathIndex: undefined,
241
241
 
242
242
  // Academic benchmarks
243
- mmluPro: 0.388,
244
- gpqa: 0.322,
245
- hle: 0.045,
243
+ mmluPro: undefined,
244
+ gpqa: 0.278,
245
+ hle: 0.065,
246
246
 
247
247
  // Capabilities
248
248
  contextWindow: 8192,
@@ -250,18 +250,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
250
250
  supportsVision: false,
251
251
 
252
252
  // Metadata
253
- lastUpdated: "2026-06-01",
254
- originalModel: "Jamba 1.7 Mini",
253
+ lastUpdated: "2026-07-01",
254
+ originalModel: "MiniCPM5-1B (Reasoning)",
255
255
  },
256
- "jamba-reasoning-3b": {
256
+ "trinity-large-thinking": {
257
257
  // AA specific benchmarks
258
- codingIndex: 2.5,
259
- mathIndex: 10.7,
258
+ codingIndex: undefined,
259
+ mathIndex: undefined,
260
260
 
261
261
  // Academic benchmarks
262
- mmluPro: 0.577,
263
- gpqa: 0.333,
264
- hle: 0.046,
262
+ mmluPro: undefined,
263
+ gpqa: 0.752,
264
+ hle: 0.147,
265
265
 
266
266
  // Capabilities
267
267
  contextWindow: 8192,
@@ -269,18 +269,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
269
269
  supportsVision: false,
270
270
 
271
271
  // Metadata
272
- lastUpdated: "2026-06-01",
273
- originalModel: "Jamba Reasoning 3B",
272
+ lastUpdated: "2026-07-01",
273
+ originalModel: "Trinity Large Thinking",
274
274
  },
275
- "qwen3-next-80b-a3b-reasoning": {
275
+ "jt-35b-flash": {
276
276
  // AA specific benchmarks
277
- codingIndex: 19.5,
278
- mathIndex: 84.3,
277
+ codingIndex: undefined,
278
+ mathIndex: undefined,
279
279
 
280
280
  // Academic benchmarks
281
- mmluPro: 0.824,
282
- gpqa: 0.759,
283
- hle: 0.117,
281
+ mmluPro: undefined,
282
+ gpqa: 0.829,
283
+ hle: 0.061,
284
284
 
285
285
  // Capabilities
286
286
  contextWindow: 8192,
@@ -288,18 +288,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
288
288
  supportsVision: false,
289
289
 
290
290
  // Metadata
291
- lastUpdated: "2026-06-01",
292
- originalModel: "Qwen3 Next 80B A3B (Reasoning)",
291
+ lastUpdated: "2026-07-01",
292
+ originalModel: "JT-35B-Flash",
293
293
  },
294
- "qwen3.6-35b-a3b-reasoning": {
294
+ "jt-mini": {
295
295
  // AA specific benchmarks
296
- codingIndex: 35.2,
296
+ codingIndex: undefined,
297
297
  mathIndex: undefined,
298
298
 
299
299
  // Academic benchmarks
300
300
  mmluPro: undefined,
301
- gpqa: 0.841,
302
- hle: 0.202,
301
+ gpqa: 0.676,
302
+ hle: 0.066,
303
303
 
304
304
  // Capabilities
305
305
  contextWindow: 8192,
@@ -307,18 +307,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
307
307
  supportsVision: false,
308
308
 
309
309
  // Metadata
310
- lastUpdated: "2026-06-01",
311
- originalModel: "Qwen3.6 35B A3B (Reasoning)",
310
+ lastUpdated: "2026-07-01",
311
+ originalModel: "JT-MINI",
312
312
  },
313
- "qwen3.5-122b-a10b-reasoning": {
313
+ "nex-n2-pro": {
314
314
  // AA specific benchmarks
315
- codingIndex: 34.7,
315
+ codingIndex: 59.1,
316
316
  mathIndex: undefined,
317
317
 
318
318
  // Academic benchmarks
319
319
  mmluPro: undefined,
320
- gpqa: 0.857,
321
- hle: 0.234,
320
+ gpqa: 0.892,
321
+ hle: 0.324,
322
322
 
323
323
  // Capabilities
324
324
  contextWindow: 8192,
@@ -326,18 +326,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
326
326
  supportsVision: false,
327
327
 
328
328
  // Metadata
329
- lastUpdated: "2026-06-01",
330
- originalModel: "Qwen3.5 122B A10B (Reasoning)",
329
+ lastUpdated: "2026-07-01",
330
+ originalModel: "Nex-N2-Pro",
331
331
  },
332
- "qwen3.5-9b-non-reasoning": {
332
+ "glm-5-turbo": {
333
333
  // AA specific benchmarks
334
- codingIndex: 21.3,
334
+ codingIndex: undefined,
335
335
  mathIndex: undefined,
336
336
 
337
337
  // Academic benchmarks
338
338
  mmluPro: undefined,
339
- gpqa: 0.786,
340
- hle: 0.086,
339
+ gpqa: 0.847,
340
+ hle: 0.254,
341
341
 
342
342
  // Capabilities
343
343
  contextWindow: 8192,
@@ -345,18 +345,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
345
345
  supportsVision: false,
346
346
 
347
347
  // Metadata
348
- lastUpdated: "2026-06-01",
349
- originalModel: "Qwen3.5 9B (Non-reasoning)",
348
+ lastUpdated: "2026-07-01",
349
+ originalModel: "GLM-5-Turbo",
350
350
  },
351
- "qwen3.5-9b-reasoning": {
351
+ "glm-5.1-reasoning": {
352
352
  // AA specific benchmarks
353
- codingIndex: 25.3,
353
+ codingIndex: 55.8,
354
354
  mathIndex: undefined,
355
355
 
356
356
  // Academic benchmarks
357
357
  mmluPro: undefined,
358
- gpqa: 0.806,
359
- hle: 0.133,
358
+ gpqa: 0.868,
359
+ hle: 0.28,
360
360
 
361
361
  // Capabilities
362
362
  contextWindow: 8192,
@@ -364,18 +364,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
364
364
  supportsVision: false,
365
365
 
366
366
  // Metadata
367
- lastUpdated: "2026-06-01",
368
- originalModel: "Qwen3.5 9B (Reasoning)",
367
+ lastUpdated: "2026-07-01",
368
+ originalModel: "GLM-5.1 (Reasoning)",
369
369
  },
370
- "qwen3.6-27b-reasoning": {
370
+ "glm-5.2-max": {
371
371
  // AA specific benchmarks
372
- codingIndex: 36.5,
372
+ codingIndex: 68.8,
373
373
  mathIndex: undefined,
374
374
 
375
375
  // Academic benchmarks
376
376
  mmluPro: undefined,
377
- gpqa: 0.842,
378
- hle: 0.216,
377
+ gpqa: 0.895,
378
+ hle: 0.401,
379
379
 
380
380
  // Capabilities
381
381
  contextWindow: 8192,
@@ -383,18 +383,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
383
383
  supportsVision: false,
384
384
 
385
385
  // Metadata
386
- lastUpdated: "2026-06-01",
387
- originalModel: "Qwen3.6 27B (Reasoning)",
386
+ lastUpdated: "2026-07-01",
387
+ originalModel: "GLM-5.2 (max)",
388
388
  },
389
- "qwen3.5-397b-a17b-reasoning": {
389
+ "glm-5.1-non-reasoning": {
390
390
  // AA specific benchmarks
391
- codingIndex: 41.3,
391
+ codingIndex: undefined,
392
392
  mathIndex: undefined,
393
393
 
394
394
  // Academic benchmarks
395
395
  mmluPro: undefined,
396
- gpqa: 0.893,
397
- hle: 0.273,
396
+ gpqa: 0.839,
397
+ hle: 0.256,
398
398
 
399
399
  // Capabilities
400
400
  contextWindow: 8192,
@@ -402,18 +402,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
402
402
  supportsVision: false,
403
403
 
404
404
  // Metadata
405
- lastUpdated: "2026-06-01",
406
- originalModel: "Qwen3.5 397B A17B (Reasoning)",
405
+ lastUpdated: "2026-07-01",
406
+ originalModel: "GLM-5.1 (Non-reasoning)",
407
407
  },
408
- "qwen3.5-397b-a17b-non-reasoning": {
408
+ "glm-5v-turbo-reasoning": {
409
409
  // AA specific benchmarks
410
- codingIndex: 37.4,
410
+ codingIndex: undefined,
411
411
  mathIndex: undefined,
412
412
 
413
413
  // Academic benchmarks
414
414
  mmluPro: undefined,
415
- gpqa: 0.861,
416
- hle: 0.188,
415
+ gpqa: 0.809,
416
+ hle: 0.158,
417
417
 
418
418
  // Capabilities
419
419
  contextWindow: 8192,
@@ -421,18 +421,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
421
421
  supportsVision: false,
422
422
 
423
423
  // Metadata
424
- lastUpdated: "2026-06-01",
425
- originalModel: "Qwen3.5 397B A17B (Non-reasoning)",
424
+ lastUpdated: "2026-07-01",
425
+ originalModel: "GLM 5V Turbo (Reasoning)",
426
426
  },
427
- "qwen3.5-0.8b-reasoning": {
427
+ "north-mini-code": {
428
428
  // AA specific benchmarks
429
- codingIndex: 0,
429
+ codingIndex: undefined,
430
430
  mathIndex: undefined,
431
431
 
432
432
  // Academic benchmarks
433
433
  mmluPro: undefined,
434
- gpqa: 0.111,
435
- hle: 0.012,
434
+ gpqa: 0.757,
435
+ hle: 0.099,
436
436
 
437
437
  // Capabilities
438
438
  contextWindow: 8192,
@@ -440,18 +440,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
440
440
  supportsVision: false,
441
441
 
442
442
  // Metadata
443
- lastUpdated: "2026-06-01",
444
- originalModel: "Qwen3.5 0.8B (Reasoning)",
443
+ lastUpdated: "2026-07-01",
444
+ originalModel: "North Mini Code",
445
445
  },
446
- "qwen3.5-2b-non-reasoning": {
446
+ "command-a": {
447
447
  // AA specific benchmarks
448
- codingIndex: 4.9,
449
- mathIndex: undefined,
448
+ codingIndex: undefined,
449
+ mathIndex: 13,
450
450
 
451
451
  // Academic benchmarks
452
- mmluPro: undefined,
453
- gpqa: 0.438,
454
- hle: 0.049,
452
+ mmluPro: 0.712,
453
+ gpqa: 0.527,
454
+ hle: 0.046,
455
455
 
456
456
  // Capabilities
457
457
  contextWindow: 8192,
@@ -459,18 +459,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
459
459
  supportsVision: false,
460
460
 
461
461
  // Metadata
462
- lastUpdated: "2026-06-01",
463
- originalModel: "Qwen3.5 2B (Non-reasoning)",
462
+ lastUpdated: "2026-07-01",
463
+ originalModel: "Command A",
464
464
  },
465
- "qwen3.6-27b-non-reasoning": {
465
+ "command-a+": {
466
466
  // AA specific benchmarks
467
- codingIndex: 26.6,
467
+ codingIndex: undefined,
468
468
  mathIndex: undefined,
469
469
 
470
470
  // Academic benchmarks
471
471
  mmluPro: undefined,
472
- gpqa: 0.829,
473
- hle: 0.136,
472
+ gpqa: 0.761,
473
+ hle: 0.114,
474
474
 
475
475
  // Capabilities
476
476
  contextWindow: 8192,
@@ -478,18 +478,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
478
478
  supportsVision: false,
479
479
 
480
480
  // Metadata
481
- lastUpdated: "2026-06-01",
482
- originalModel: "Qwen3.6 27B (Non-reasoning)",
481
+ lastUpdated: "2026-07-01",
482
+ originalModel: "Command A+",
483
483
  },
484
- "qwen3.5-122b-a10b-non-reasoning": {
484
+ "tiny-aya-global": {
485
485
  // AA specific benchmarks
486
- codingIndex: 31.6,
486
+ codingIndex: undefined,
487
487
  mathIndex: undefined,
488
488
 
489
489
  // Academic benchmarks
490
490
  mmluPro: undefined,
491
- gpqa: 0.827,
492
- hle: 0.148,
491
+ gpqa: 0.305,
492
+ hle: 0.052,
493
493
 
494
494
  // Capabilities
495
495
  contextWindow: 8192,
@@ -497,18 +497,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
497
497
  supportsVision: false,
498
498
 
499
499
  // Metadata
500
- lastUpdated: "2026-06-01",
501
- originalModel: "Qwen3.5 122B A10B (Non-reasoning)",
500
+ lastUpdated: "2026-07-01",
501
+ originalModel: "Tiny Aya Global",
502
502
  },
503
- "qwen3.6-35b-a3b-non-reasoning": {
503
+ "apriel-v1.6-15b-thinker": {
504
504
  // AA specific benchmarks
505
- codingIndex: 17.6,
506
- mathIndex: undefined,
505
+ codingIndex: undefined,
506
+ mathIndex: 88,
507
507
 
508
508
  // Academic benchmarks
509
- mmluPro: undefined,
510
- gpqa: 0.817,
511
- hle: 0.125,
509
+ mmluPro: 0.79,
510
+ gpqa: 0.733,
511
+ hle: 0.098,
512
512
 
513
513
  // Capabilities
514
514
  contextWindow: 8192,
@@ -516,18 +516,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
516
516
  supportsVision: false,
517
517
 
518
518
  // Metadata
519
- lastUpdated: "2026-06-01",
520
- originalModel: "Qwen3.6 35B A3B (Non-reasoning)",
519
+ lastUpdated: "2026-07-01",
520
+ originalModel: "Apriel-v1.6-15B-Thinker",
521
521
  },
522
- "qwen3.6-plus": {
522
+ "jamba-reasoning-3b": {
523
523
  // AA specific benchmarks
524
- codingIndex: 42.9,
525
- mathIndex: undefined,
524
+ codingIndex: undefined,
525
+ mathIndex: 10.7,
526
526
 
527
527
  // Academic benchmarks
528
- mmluPro: undefined,
529
- gpqa: 0.882,
530
- hle: 0.257,
528
+ mmluPro: 0.577,
529
+ gpqa: 0.333,
530
+ hle: 0.046,
531
531
 
532
532
  // Capabilities
533
533
  contextWindow: 8192,
@@ -535,18 +535,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
535
535
  supportsVision: false,
536
536
 
537
537
  // Metadata
538
- lastUpdated: "2026-06-01",
539
- originalModel: "Qwen3.6 Plus",
538
+ lastUpdated: "2026-07-01",
539
+ originalModel: "Jamba Reasoning 3B",
540
540
  },
541
- "qwen3-omni-30b-a3b-reasoning": {
541
+ "jamba-1.7-large": {
542
542
  // AA specific benchmarks
543
- codingIndex: 12.7,
544
- mathIndex: 74,
543
+ codingIndex: undefined,
544
+ mathIndex: 2.3,
545
545
 
546
546
  // Academic benchmarks
547
- mmluPro: 0.792,
548
- gpqa: 0.726,
549
- hle: 0.073,
547
+ mmluPro: 0.577,
548
+ gpqa: 0.39,
549
+ hle: 0.038,
550
550
 
551
551
  // Capabilities
552
552
  contextWindow: 8192,
@@ -554,18 +554,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
554
554
  supportsVision: false,
555
555
 
556
556
  // Metadata
557
- lastUpdated: "2026-06-01",
558
- originalModel: "Qwen3 Omni 30B A3B (Reasoning)",
557
+ lastUpdated: "2026-07-01",
558
+ originalModel: "Jamba 1.7 Large",
559
559
  },
560
- "qwen3.5-omni-plus": {
560
+ "jamba-1.7-mini": {
561
561
  // AA specific benchmarks
562
- codingIndex: 27.6,
563
- mathIndex: undefined,
562
+ codingIndex: undefined,
563
+ mathIndex: 0.3,
564
564
 
565
565
  // Academic benchmarks
566
- mmluPro: undefined,
567
- gpqa: 0.826,
568
- hle: 0.139,
566
+ mmluPro: 0.388,
567
+ gpqa: 0.322,
568
+ hle: 0.045,
569
569
 
570
570
  // Capabilities
571
571
  contextWindow: 8192,
@@ -573,56 +573,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
573
573
  supportsVision: false,
574
574
 
575
575
  // Metadata
576
- lastUpdated: "2026-06-01",
577
- originalModel: "Qwen3.5 Omni Plus",
578
- },
579
- "qwen3-omni-30b-a3b-instruct": {
580
- // AA specific benchmarks
581
- codingIndex: 7.2,
582
- mathIndex: 52.3,
583
-
584
- // Academic benchmarks
585
- mmluPro: 0.725,
586
- gpqa: 0.62,
587
- hle: 0.051,
588
-
589
- // Capabilities
590
- contextWindow: 8192,
591
- supportsReasoning: false,
592
- supportsVision: false,
593
-
594
- // Metadata
595
- lastUpdated: "2026-06-01",
596
- originalModel: "Qwen3 Omni 30B A3B Instruct",
597
- },
598
- "qwen3-next-80b-a3b-instruct": {
599
- // AA specific benchmarks
600
- codingIndex: 15.3,
601
- mathIndex: 66.3,
602
-
603
- // Academic benchmarks
604
- mmluPro: 0.819,
605
- gpqa: 0.738,
606
- hle: 0.073,
607
-
608
- // Capabilities
609
- contextWindow: 8192,
610
- supportsReasoning: false,
611
- supportsVision: false,
612
-
613
- // Metadata
614
- lastUpdated: "2026-06-01",
615
- originalModel: "Qwen3 Next 80B A3B Instruct",
576
+ lastUpdated: "2026-07-01",
577
+ originalModel: "Jamba 1.7 Mini",
616
578
  },
617
- "qwen3.7-max": {
579
+ "qwen3.6-27b-non-reasoning": {
618
580
  // AA specific benchmarks
619
- codingIndex: 50.1,
581
+ codingIndex: undefined,
620
582
  mathIndex: undefined,
621
583
 
622
584
  // Academic benchmarks
623
585
  mmluPro: undefined,
624
- gpqa: 0.923,
625
- hle: 0.381,
586
+ gpqa: 0.829,
587
+ hle: 0.136,
626
588
 
627
589
  // Capabilities
628
590
  contextWindow: 8192,
@@ -630,18 +592,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
630
592
  supportsVision: false,
631
593
 
632
594
  // Metadata
633
- lastUpdated: "2026-06-01",
634
- originalModel: "Qwen3.7 Max",
595
+ lastUpdated: "2026-07-01",
596
+ originalModel: "Qwen3.6 27B (Non-reasoning)",
635
597
  },
636
- "qwen3.5-omni-flash": {
598
+ "qwen3.5-397b-a17b-reasoning": {
637
599
  // AA specific benchmarks
638
- codingIndex: 14,
600
+ codingIndex: 48.2,
639
601
  mathIndex: undefined,
640
602
 
641
603
  // Academic benchmarks
642
604
  mmluPro: undefined,
643
- gpqa: 0.742,
644
- hle: 0.071,
605
+ gpqa: 0.893,
606
+ hle: 0.273,
645
607
 
646
608
  // Capabilities
647
609
  contextWindow: 8192,
@@ -649,18 +611,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
649
611
  supportsVision: false,
650
612
 
651
613
  // Metadata
652
- lastUpdated: "2026-06-01",
653
- originalModel: "Qwen3.5 Omni Flash",
614
+ lastUpdated: "2026-07-01",
615
+ originalModel: "Qwen3.5 397B A17B (Reasoning)",
654
616
  },
655
- "qwen3.5-35b-a3b-non-reasoning": {
617
+ "qwen3.5-122b-a10b-reasoning": {
656
618
  // AA specific benchmarks
657
- codingIndex: 16.8,
619
+ codingIndex: 45.7,
658
620
  mathIndex: undefined,
659
621
 
660
622
  // Academic benchmarks
661
623
  mmluPro: undefined,
662
- gpqa: 0.819,
663
- hle: 0.128,
624
+ gpqa: 0.857,
625
+ hle: 0.234,
664
626
 
665
627
  // Capabilities
666
628
  contextWindow: 8192,
@@ -668,18 +630,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
668
630
  supportsVision: false,
669
631
 
670
632
  // Metadata
671
- lastUpdated: "2026-06-01",
672
- originalModel: "Qwen3.5 35B A3B (Non-reasoning)",
633
+ lastUpdated: "2026-07-01",
634
+ originalModel: "Qwen3.5 122B A10B (Reasoning)",
673
635
  },
674
- "qwen3-coder-next": {
636
+ "qwen3-next-80b-a3b-instruct": {
675
637
  // AA specific benchmarks
676
- codingIndex: 22.9,
677
- mathIndex: undefined,
638
+ codingIndex: undefined,
639
+ mathIndex: 66.3,
678
640
 
679
641
  // Academic benchmarks
680
- mmluPro: undefined,
681
- gpqa: 0.737,
682
- hle: 0.093,
642
+ mmluPro: 0.819,
643
+ gpqa: 0.738,
644
+ hle: 0.073,
683
645
 
684
646
  // Capabilities
685
647
  contextWindow: 8192,
@@ -687,18 +649,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
687
649
  supportsVision: false,
688
650
 
689
651
  // Metadata
690
- lastUpdated: "2026-06-01",
691
- originalModel: "Qwen3 Coder Next",
652
+ lastUpdated: "2026-07-01",
653
+ originalModel: "Qwen3 Next 80B A3B Instruct",
692
654
  },
693
- "qwen3.5-4b-reasoning": {
655
+ "qwen3.5-397b-a17b-non-reasoning": {
694
656
  // AA specific benchmarks
695
- codingIndex: 17.5,
657
+ codingIndex: undefined,
696
658
  mathIndex: undefined,
697
659
 
698
660
  // Academic benchmarks
699
661
  mmluPro: undefined,
700
- gpqa: 0.771,
701
- hle: 0.078,
662
+ gpqa: 0.861,
663
+ hle: 0.188,
702
664
 
703
665
  // Capabilities
704
666
  contextWindow: 8192,
@@ -706,18 +668,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
706
668
  supportsVision: false,
707
669
 
708
670
  // Metadata
709
- lastUpdated: "2026-06-01",
710
- originalModel: "Qwen3.5 4B (Reasoning)",
671
+ lastUpdated: "2026-07-01",
672
+ originalModel: "Qwen3.5 397B A17B (Non-reasoning)",
711
673
  },
712
- "qwen3.5-2b-reasoning": {
674
+ "qwen3-next-80b-a3b-reasoning": {
713
675
  // AA specific benchmarks
714
- codingIndex: 3.5,
715
- mathIndex: undefined,
676
+ codingIndex: undefined,
677
+ mathIndex: 84.3,
716
678
 
717
679
  // Academic benchmarks
718
- mmluPro: undefined,
719
- gpqa: 0.456,
720
- hle: 0.021,
680
+ mmluPro: 0.824,
681
+ gpqa: 0.759,
682
+ hle: 0.117,
721
683
 
722
684
  // Capabilities
723
685
  contextWindow: 8192,
@@ -725,18 +687,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
725
687
  supportsVision: false,
726
688
 
727
689
  // Metadata
728
- lastUpdated: "2026-06-01",
729
- originalModel: "Qwen3.5 2B (Reasoning)",
690
+ lastUpdated: "2026-07-01",
691
+ originalModel: "Qwen3 Next 80B A3B (Reasoning)",
730
692
  },
731
- "qwen3.5-0.8b-non-reasoning": {
693
+ "qwen3.6-35b-a3b-non-reasoning": {
732
694
  // AA specific benchmarks
733
- codingIndex: 1,
695
+ codingIndex: undefined,
734
696
  mathIndex: undefined,
735
697
 
736
698
  // Academic benchmarks
737
699
  mmluPro: undefined,
738
- gpqa: 0.236,
739
- hle: 0.049,
700
+ gpqa: 0.817,
701
+ hle: 0.125,
740
702
 
741
703
  // Capabilities
742
704
  contextWindow: 8192,
@@ -744,56 +706,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
744
706
  supportsVision: false,
745
707
 
746
708
  // Metadata
747
- lastUpdated: "2026-06-01",
748
- originalModel: "Qwen3.5 0.8B (Non-reasoning)",
709
+ lastUpdated: "2026-07-01",
710
+ originalModel: "Qwen3.6 35B A3B (Non-reasoning)",
749
711
  },
750
- "qwen3.5-4b-non-reasoning": {
712
+ "qwen3.7-max": {
751
713
  // AA specific benchmarks
752
- codingIndex: 13.7,
714
+ codingIndex: 66,
753
715
  mathIndex: undefined,
754
716
 
755
717
  // Academic benchmarks
756
718
  mmluPro: undefined,
757
- gpqa: 0.712,
758
- hle: 0.075,
759
-
760
- // Capabilities
761
- contextWindow: 8192,
762
- supportsReasoning: false,
763
- supportsVision: false,
764
-
765
- // Metadata
766
- lastUpdated: "2026-06-01",
767
- originalModel: "Qwen3.5 4B (Non-reasoning)",
768
- },
769
- "ring-flash-2.0": {
770
- // AA specific benchmarks
771
- codingIndex: 10.6,
772
- mathIndex: 83.7,
773
-
774
- // Academic benchmarks
775
- mmluPro: 0.793,
776
- gpqa: 0.725,
777
- hle: 0.089,
778
-
779
- // Capabilities
780
- contextWindow: 8192,
781
- supportsReasoning: false,
782
- supportsVision: false,
783
-
784
- // Metadata
785
- lastUpdated: "2026-06-01",
786
- originalModel: "Ring-flash-2.0",
787
- },
788
- "ling-mini-2.0": {
789
- // AA specific benchmarks
790
- codingIndex: 5,
791
- mathIndex: 49.3,
792
-
793
- // Academic benchmarks
794
- mmluPro: 0.671,
795
- gpqa: 0.562,
796
- hle: 0.05,
719
+ gpqa: 0.923,
720
+ hle: 0.381,
797
721
 
798
722
  // Capabilities
799
723
  contextWindow: 8192,
@@ -801,18 +725,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
801
725
  supportsVision: false,
802
726
 
803
727
  // Metadata
804
- lastUpdated: "2026-06-01",
805
- originalModel: "Ling-mini-2.0",
728
+ lastUpdated: "2026-07-01",
729
+ originalModel: "Qwen3.7 Max",
806
730
  },
807
- "ling-2.6-1t": {
731
+ "qwen3.6-35b-a3b-reasoning": {
808
732
  // AA specific benchmarks
809
- codingIndex: 33.1,
733
+ codingIndex: 41.9,
810
734
  mathIndex: undefined,
811
735
 
812
736
  // Academic benchmarks
813
737
  mmluPro: undefined,
814
- gpqa: 0.752,
815
- hle: 0.082,
738
+ gpqa: 0.841,
739
+ hle: 0.202,
816
740
 
817
741
  // Capabilities
818
742
  contextWindow: 8192,
@@ -820,18 +744,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
820
744
  supportsVision: false,
821
745
 
822
746
  // Metadata
823
- lastUpdated: "2026-06-01",
824
- originalModel: "Ling-2.6-1T",
747
+ lastUpdated: "2026-07-01",
748
+ originalModel: "Qwen3.6 35B A3B (Reasoning)",
825
749
  },
826
- "ling-2.6-flash": {
750
+ "qwen3.6-27b-reasoning": {
827
751
  // AA specific benchmarks
828
- codingIndex: 23.2,
752
+ codingIndex: 53.7,
829
753
  mathIndex: undefined,
830
754
 
831
755
  // Academic benchmarks
832
756
  mmluPro: undefined,
833
- gpqa: 0.593,
834
- hle: 0.062,
757
+ gpqa: 0.842,
758
+ hle: 0.216,
835
759
 
836
760
  // Capabilities
837
761
  contextWindow: 8192,
@@ -839,18 +763,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
839
763
  supportsVision: false,
840
764
 
841
765
  // Metadata
842
- lastUpdated: "2026-06-01",
843
- originalModel: "Ling 2.6 Flash",
766
+ lastUpdated: "2026-07-01",
767
+ originalModel: "Qwen3.6 27B (Reasoning)",
844
768
  },
845
- "ring-2.6-1t": {
769
+ "qwen3-coder-next": {
846
770
  // AA specific benchmarks
847
- codingIndex: 33.3,
771
+ codingIndex: undefined,
848
772
  mathIndex: undefined,
849
773
 
850
774
  // Academic benchmarks
851
775
  mmluPro: undefined,
852
- gpqa: 0.857,
853
- hle: 0.183,
776
+ gpqa: 0.737,
777
+ hle: 0.093,
854
778
 
855
779
  // Capabilities
856
780
  contextWindow: 8192,
@@ -858,17 +782,17 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
858
782
  supportsVision: false,
859
783
 
860
784
  // Metadata
861
- lastUpdated: "2026-06-01",
862
- originalModel: "Ring-2.6-1T",
785
+ lastUpdated: "2026-07-01",
786
+ originalModel: "Qwen3 Coder Next",
863
787
  },
864
- "doubao-seed-code": {
788
+ "qwen3.5-9b-reasoning": {
865
789
  // AA specific benchmarks
866
- codingIndex: 31.3,
867
- mathIndex: 79.3,
790
+ codingIndex: 28.7,
791
+ mathIndex: undefined,
868
792
 
869
793
  // Academic benchmarks
870
- mmluPro: 0.854,
871
- gpqa: 0.764,
794
+ mmluPro: undefined,
795
+ gpqa: 0.806,
872
796
  hle: 0.133,
873
797
 
874
798
  // Capabilities
@@ -877,18 +801,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
877
801
  supportsVision: false,
878
802
 
879
803
  // Metadata
880
- lastUpdated: "2026-06-01",
881
- originalModel: "Doubao Seed Code",
804
+ lastUpdated: "2026-07-01",
805
+ originalModel: "Qwen3.5 9B (Reasoning)",
882
806
  },
883
- "o1": {
807
+ "qwen3-omni-30b-a3b-instruct": {
884
808
  // AA specific benchmarks
885
- codingIndex: 20.5,
886
- mathIndex: undefined,
809
+ codingIndex: undefined,
810
+ mathIndex: 52.3,
887
811
 
888
812
  // Academic benchmarks
889
- mmluPro: 0.841,
890
- gpqa: 0.747,
891
- hle: 0.077,
813
+ mmluPro: 0.725,
814
+ gpqa: 0.62,
815
+ hle: 0.051,
892
816
 
893
817
  // Capabilities
894
818
  contextWindow: 8192,
@@ -896,18 +820,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
896
820
  supportsVision: false,
897
821
 
898
822
  // Metadata
899
- lastUpdated: "2026-06-01",
900
- originalModel: "o1",
823
+ lastUpdated: "2026-07-01",
824
+ originalModel: "Qwen3 Omni 30B A3B Instruct",
901
825
  },
902
- "o1-preview": {
826
+ "qwen3.5-122b-a10b-non-reasoning": {
903
827
  // AA specific benchmarks
904
- codingIndex: 34,
828
+ codingIndex: undefined,
905
829
  mathIndex: undefined,
906
830
 
907
831
  // Academic benchmarks
908
832
  mmluPro: undefined,
909
- gpqa: undefined,
910
- hle: undefined,
833
+ gpqa: 0.827,
834
+ hle: 0.148,
911
835
 
912
836
  // Capabilities
913
837
  contextWindow: 8192,
@@ -915,18 +839,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
915
839
  supportsVision: false,
916
840
 
917
841
  // Metadata
918
- lastUpdated: "2026-06-01",
919
- originalModel: "o1-preview",
842
+ lastUpdated: "2026-07-01",
843
+ originalModel: "Qwen3.5 122B A10B (Non-reasoning)",
920
844
  },
921
- "o1-mini": {
845
+ "qwen3.5-omni-plus": {
922
846
  // AA specific benchmarks
923
847
  codingIndex: undefined,
924
848
  mathIndex: undefined,
925
849
 
926
850
  // Academic benchmarks
927
- mmluPro: 0.742,
928
- gpqa: 0.603,
929
- hle: 0.049,
851
+ mmluPro: undefined,
852
+ gpqa: 0.826,
853
+ hle: 0.139,
930
854
 
931
855
  // Capabilities
932
856
  contextWindow: 8192,
@@ -934,18 +858,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
934
858
  supportsVision: false,
935
859
 
936
860
  // Metadata
937
- lastUpdated: "2026-06-01",
938
- originalModel: "o1-mini",
861
+ lastUpdated: "2026-07-01",
862
+ originalModel: "Qwen3.5 Omni Plus",
939
863
  },
940
- "gpt-4o-aug-24": {
864
+ "qwen3.6-plus": {
941
865
  // AA specific benchmarks
942
- codingIndex: 16.6,
866
+ codingIndex: 54.5,
943
867
  mathIndex: undefined,
944
868
 
945
869
  // Academic benchmarks
946
870
  mmluPro: undefined,
947
- gpqa: 0.521,
948
- hle: 0.029,
871
+ gpqa: 0.882,
872
+ hle: 0.257,
949
873
 
950
874
  // Capabilities
951
875
  contextWindow: 8192,
@@ -953,18 +877,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
953
877
  supportsVision: false,
954
878
 
955
879
  // Metadata
956
- lastUpdated: "2026-06-01",
957
- originalModel: "GPT-4o (Aug '24)",
880
+ lastUpdated: "2026-07-01",
881
+ originalModel: "Qwen3.6 Plus",
958
882
  },
959
- "gpt-4o-may-24": {
883
+ "qwen3.5-35b-a3b-non-reasoning": {
960
884
  // AA specific benchmarks
961
- codingIndex: 24.2,
885
+ codingIndex: undefined,
962
886
  mathIndex: undefined,
963
887
 
964
888
  // Academic benchmarks
965
- mmluPro: 0.74,
966
- gpqa: 0.526,
967
- hle: 0.028,
889
+ mmluPro: undefined,
890
+ gpqa: 0.819,
891
+ hle: 0.128,
968
892
 
969
893
  // Capabilities
970
894
  contextWindow: 8192,
@@ -972,37 +896,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
972
896
  supportsVision: false,
973
897
 
974
898
  // Metadata
975
- lastUpdated: "2026-06-01",
976
- originalModel: "GPT-4o (May '24)",
899
+ lastUpdated: "2026-07-01",
900
+ originalModel: "Qwen3.5 35B A3B (Non-reasoning)",
977
901
  },
978
- "gpt-4-turbo": {
902
+ "qwen3.5-0.8b-reasoning": {
979
903
  // AA specific benchmarks
980
- codingIndex: 21.5,
904
+ codingIndex: 15,
981
905
  mathIndex: undefined,
982
906
 
983
907
  // Academic benchmarks
984
- mmluPro: 0.694,
985
- gpqa: undefined,
986
- hle: 0.033,
987
-
988
- // Capabilities
989
- contextWindow: 8192,
990
- supportsReasoning: false,
991
- supportsVision: false,
992
-
993
- // Metadata
994
- lastUpdated: "2026-06-01",
995
- originalModel: "GPT-4 Turbo",
996
- },
997
- "gpt-4o-nov-24": {
998
- // AA specific benchmarks
999
- codingIndex: 16.7,
1000
- mathIndex: 6,
1001
-
1002
- // Academic benchmarks
1003
- mmluPro: 0.748,
1004
- gpqa: 0.543,
1005
- hle: 0.033,
908
+ mmluPro: undefined,
909
+ gpqa: 0.111,
910
+ hle: 0.012,
1006
911
 
1007
912
  // Capabilities
1008
913
  contextWindow: 8192,
@@ -1010,18 +915,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1010
915
  supportsVision: false,
1011
916
 
1012
917
  // Metadata
1013
- lastUpdated: "2026-06-01",
1014
- originalModel: "GPT-4o (Nov '24)",
918
+ lastUpdated: "2026-07-01",
919
+ originalModel: "Qwen3.5 0.8B (Reasoning)",
1015
920
  },
1016
- "gpt-4o-mini": {
921
+ "qwen3-omni-30b-a3b-reasoning": {
1017
922
  // AA specific benchmarks
1018
923
  codingIndex: undefined,
1019
- mathIndex: 14.7,
924
+ mathIndex: 74,
1020
925
 
1021
926
  // Academic benchmarks
1022
- mmluPro: 0.648,
1023
- gpqa: 0.426,
1024
- hle: 0.04,
927
+ mmluPro: 0.792,
928
+ gpqa: 0.726,
929
+ hle: 0.073,
1025
930
 
1026
931
  // Capabilities
1027
932
  contextWindow: 8192,
@@ -1029,37 +934,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1029
934
  supportsVision: false,
1030
935
 
1031
936
  // Metadata
1032
- lastUpdated: "2026-06-01",
1033
- originalModel: "GPT-4o mini",
937
+ lastUpdated: "2026-07-01",
938
+ originalModel: "Qwen3 Omni 30B A3B (Reasoning)",
1034
939
  },
1035
- "gpt-3.5-turbo": {
940
+ "qwen3.7-plus": {
1036
941
  // AA specific benchmarks
1037
- codingIndex: 10.7,
942
+ codingIndex: 55.9,
1038
943
  mathIndex: undefined,
1039
944
 
1040
945
  // Academic benchmarks
1041
- mmluPro: 0.462,
1042
- gpqa: 0.297,
1043
- hle: undefined,
1044
-
1045
- // Capabilities
1046
- contextWindow: 8192,
1047
- supportsReasoning: false,
1048
- supportsVision: false,
1049
-
1050
- // Metadata
1051
- lastUpdated: "2026-06-01",
1052
- originalModel: "GPT-3.5 Turbo",
1053
- },
1054
- "gpt-4.1": {
1055
- // AA specific benchmarks
1056
- codingIndex: 21.8,
1057
- mathIndex: 34.7,
1058
-
1059
- // Academic benchmarks
1060
- mmluPro: 0.806,
1061
- gpqa: 0.666,
1062
- hle: 0.046,
946
+ mmluPro: undefined,
947
+ gpqa: 0.9,
948
+ hle: 0.334,
1063
949
 
1064
950
  // Capabilities
1065
951
  contextWindow: 8192,
@@ -1067,18 +953,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1067
953
  supportsVision: false,
1068
954
 
1069
955
  // Metadata
1070
- lastUpdated: "2026-06-01",
1071
- originalModel: "GPT-4.1",
956
+ lastUpdated: "2026-07-01",
957
+ originalModel: "Qwen3.7 Plus",
1072
958
  },
1073
- "gpt-5-mini-high": {
959
+ "qwen3.5-4b-reasoning": {
1074
960
  // AA specific benchmarks
1075
- codingIndex: 35.3,
1076
- mathIndex: 90.7,
961
+ codingIndex: 22.6,
962
+ mathIndex: undefined,
1077
963
 
1078
964
  // Academic benchmarks
1079
- mmluPro: 0.837,
1080
- gpqa: 0.828,
1081
- hle: 0.197,
965
+ mmluPro: undefined,
966
+ gpqa: 0.771,
967
+ hle: 0.078,
1082
968
 
1083
969
  // Capabilities
1084
970
  contextWindow: 8192,
@@ -1086,18 +972,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1086
972
  supportsVision: false,
1087
973
 
1088
974
  // Metadata
1089
- lastUpdated: "2026-06-01",
1090
- originalModel: "GPT-5 mini (high)",
975
+ lastUpdated: "2026-07-01",
976
+ originalModel: "Qwen3.5 4B (Reasoning)",
1091
977
  },
1092
- "gpt-5-high": {
978
+ "qwen3.5-2b-reasoning": {
1093
979
  // AA specific benchmarks
1094
- codingIndex: 36,
1095
- mathIndex: 94.3,
980
+ codingIndex: 19.7,
981
+ mathIndex: undefined,
1096
982
 
1097
983
  // Academic benchmarks
1098
- mmluPro: 0.871,
1099
- gpqa: 0.854,
1100
- hle: 0.265,
984
+ mmluPro: undefined,
985
+ gpqa: 0.456,
986
+ hle: 0.021,
1101
987
 
1102
988
  // Capabilities
1103
989
  contextWindow: 8192,
@@ -1105,18 +991,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1105
991
  supportsVision: false,
1106
992
 
1107
993
  // Metadata
1108
- lastUpdated: "2026-06-01",
1109
- originalModel: "GPT-5 (high)",
994
+ lastUpdated: "2026-07-01",
995
+ originalModel: "Qwen3.5 2B (Reasoning)",
1110
996
  },
1111
- "gpt-5.2-xhigh": {
997
+ "qwen3.5-0.8b-non-reasoning": {
1112
998
  // AA specific benchmarks
1113
- codingIndex: 48.7,
1114
- mathIndex: 99,
999
+ codingIndex: 21.9,
1000
+ mathIndex: undefined,
1115
1001
 
1116
1002
  // Academic benchmarks
1117
- mmluPro: 0.874,
1118
- gpqa: 0.903,
1119
- hle: 0.354,
1003
+ mmluPro: undefined,
1004
+ gpqa: 0.236,
1005
+ hle: 0.049,
1120
1006
 
1121
1007
  // Capabilities
1122
1008
  contextWindow: 8192,
@@ -1124,18 +1010,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1124
1010
  supportsVision: false,
1125
1011
 
1126
1012
  // Metadata
1127
- lastUpdated: "2026-06-01",
1128
- originalModel: "GPT-5.2 (xhigh)",
1013
+ lastUpdated: "2026-07-01",
1014
+ originalModel: "Qwen3.5 0.8B (Non-reasoning)",
1129
1015
  },
1130
- "gpt-5.1-codex-mini-high": {
1016
+ "qwen3.5-4b-non-reasoning": {
1131
1017
  // AA specific benchmarks
1132
- codingIndex: 36.4,
1133
- mathIndex: 91.7,
1018
+ codingIndex: 20.3,
1019
+ mathIndex: undefined,
1134
1020
 
1135
1021
  // Academic benchmarks
1136
- mmluPro: 0.82,
1137
- gpqa: 0.813,
1138
- hle: 0.169,
1022
+ mmluPro: undefined,
1023
+ gpqa: 0.712,
1024
+ hle: 0.075,
1139
1025
 
1140
1026
  // Capabilities
1141
1027
  contextWindow: 8192,
@@ -1143,18 +1029,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1143
1029
  supportsVision: false,
1144
1030
 
1145
1031
  // Metadata
1146
- lastUpdated: "2026-06-01",
1147
- originalModel: "GPT-5.1 Codex mini (high)",
1032
+ lastUpdated: "2026-07-01",
1033
+ originalModel: "Qwen3.5 4B (Non-reasoning)",
1148
1034
  },
1149
- "gpt-5.1-high": {
1035
+ "qwen3.5-9b-non-reasoning": {
1150
1036
  // AA specific benchmarks
1151
- codingIndex: 44.7,
1152
- mathIndex: 94,
1037
+ codingIndex: 23.5,
1038
+ mathIndex: undefined,
1153
1039
 
1154
1040
  // Academic benchmarks
1155
- mmluPro: 0.87,
1156
- gpqa: 0.873,
1157
- hle: 0.265,
1041
+ mmluPro: undefined,
1042
+ gpqa: 0.786,
1043
+ hle: 0.086,
1158
1044
 
1159
1045
  // Capabilities
1160
1046
  contextWindow: 8192,
@@ -1162,18 +1048,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1162
1048
  supportsVision: false,
1163
1049
 
1164
1050
  // Metadata
1165
- lastUpdated: "2026-06-01",
1166
- originalModel: "GPT-5.1 (high)",
1051
+ lastUpdated: "2026-07-01",
1052
+ originalModel: "Qwen3.5 9B (Non-reasoning)",
1167
1053
  },
1168
- "gpt-5-nano-high": {
1054
+ "qwen3.5-2b-non-reasoning": {
1169
1055
  // AA specific benchmarks
1170
- codingIndex: 20.3,
1171
- mathIndex: 83.7,
1056
+ codingIndex: 17.4,
1057
+ mathIndex: undefined,
1172
1058
 
1173
1059
  // Academic benchmarks
1174
- mmluPro: 0.78,
1175
- gpqa: 0.676,
1176
- hle: 0.082,
1060
+ mmluPro: undefined,
1061
+ gpqa: 0.438,
1062
+ hle: 0.049,
1177
1063
 
1178
1064
  // Capabilities
1179
1065
  contextWindow: 8192,
@@ -1181,18 +1067,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1181
1067
  supportsVision: false,
1182
1068
 
1183
1069
  // Metadata
1184
- lastUpdated: "2026-06-01",
1185
- originalModel: "GPT-5 nano (high)",
1070
+ lastUpdated: "2026-07-01",
1071
+ originalModel: "Qwen3.5 2B (Non-reasoning)",
1186
1072
  },
1187
- "gpt-5.1-non-reasoning": {
1073
+ "qwen3.5-omni-flash": {
1188
1074
  // AA specific benchmarks
1189
- codingIndex: 27.3,
1190
- mathIndex: 38,
1075
+ codingIndex: undefined,
1076
+ mathIndex: undefined,
1191
1077
 
1192
1078
  // Academic benchmarks
1193
- mmluPro: 0.801,
1194
- gpqa: 0.643,
1195
- hle: 0.052,
1079
+ mmluPro: undefined,
1080
+ gpqa: 0.742,
1081
+ hle: 0.071,
1196
1082
 
1197
1083
  // Capabilities
1198
1084
  contextWindow: 8192,
@@ -1200,18 +1086,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1200
1086
  supportsVision: false,
1201
1087
 
1202
1088
  // Metadata
1203
- lastUpdated: "2026-06-01",
1204
- originalModel: "GPT-5.1 (Non-reasoning)",
1089
+ lastUpdated: "2026-07-01",
1090
+ originalModel: "Qwen3.5 Omni Flash",
1205
1091
  },
1206
- "o3-pro": {
1092
+ "ling-2.6-flash": {
1207
1093
  // AA specific benchmarks
1208
1094
  codingIndex: undefined,
1209
1095
  mathIndex: undefined,
1210
1096
 
1211
1097
  // Academic benchmarks
1212
1098
  mmluPro: undefined,
1213
- gpqa: 0.845,
1214
- hle: undefined,
1099
+ gpqa: 0.593,
1100
+ hle: 0.062,
1215
1101
 
1216
1102
  // Capabilities
1217
1103
  contextWindow: 8192,
@@ -1219,18 +1105,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1219
1105
  supportsVision: false,
1220
1106
 
1221
1107
  // Metadata
1222
- lastUpdated: "2026-06-01",
1223
- originalModel: "o3-pro",
1108
+ lastUpdated: "2026-07-01",
1109
+ originalModel: "Ling 2.6 Flash",
1224
1110
  },
1225
- "gpt-5-minimal": {
1111
+ "ling-mini-2.0": {
1226
1112
  // AA specific benchmarks
1227
- codingIndex: 25,
1228
- mathIndex: 31.7,
1113
+ codingIndex: undefined,
1114
+ mathIndex: 49.3,
1229
1115
 
1230
1116
  // Academic benchmarks
1231
- mmluPro: 0.806,
1232
- gpqa: 0.673,
1233
- hle: 0.054,
1117
+ mmluPro: 0.671,
1118
+ gpqa: 0.562,
1119
+ hle: 0.05,
1234
1120
 
1235
1121
  // Capabilities
1236
1122
  contextWindow: 8192,
@@ -1238,18 +1124,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1238
1124
  supportsVision: false,
1239
1125
 
1240
1126
  // Metadata
1241
- lastUpdated: "2026-06-01",
1242
- originalModel: "GPT-5 (minimal)",
1127
+ lastUpdated: "2026-07-01",
1128
+ originalModel: "Ling-mini-2.0",
1243
1129
  },
1244
- "gpt-5.2-codex-xhigh": {
1130
+ "ling-2.6-1t": {
1245
1131
  // AA specific benchmarks
1246
- codingIndex: 43,
1132
+ codingIndex: undefined,
1247
1133
  mathIndex: undefined,
1248
1134
 
1249
1135
  // Academic benchmarks
1250
1136
  mmluPro: undefined,
1251
- gpqa: 0.899,
1252
- hle: 0.335,
1137
+ gpqa: 0.752,
1138
+ hle: 0.082,
1253
1139
 
1254
1140
  // Capabilities
1255
1141
  contextWindow: 8192,
@@ -1257,18 +1143,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1257
1143
  supportsVision: false,
1258
1144
 
1259
1145
  // Metadata
1260
- lastUpdated: "2026-06-01",
1261
- originalModel: "GPT-5.2 Codex (xhigh)",
1146
+ lastUpdated: "2026-07-01",
1147
+ originalModel: "Ling-2.6-1T",
1262
1148
  },
1263
- "gpt-5.1-codex-high": {
1149
+ "ring-flash-2.0": {
1264
1150
  // AA specific benchmarks
1265
- codingIndex: 36.6,
1266
- mathIndex: 95.7,
1151
+ codingIndex: undefined,
1152
+ mathIndex: 83.7,
1267
1153
 
1268
1154
  // Academic benchmarks
1269
- mmluPro: 0.86,
1270
- gpqa: 0.86,
1271
- hle: 0.234,
1155
+ mmluPro: 0.793,
1156
+ gpqa: 0.725,
1157
+ hle: 0.089,
1272
1158
 
1273
1159
  // Capabilities
1274
1160
  contextWindow: 8192,
@@ -1276,18 +1162,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1276
1162
  supportsVision: false,
1277
1163
 
1278
1164
  // Metadata
1279
- lastUpdated: "2026-06-01",
1280
- originalModel: "GPT-5.1 Codex (high)",
1165
+ lastUpdated: "2026-07-01",
1166
+ originalModel: "Ring-flash-2.0",
1281
1167
  },
1282
- "gpt-5-nano-minimal": {
1168
+ "ring-2.6-1t": {
1283
1169
  // AA specific benchmarks
1284
- codingIndex: 14.2,
1285
- mathIndex: 27.3,
1170
+ codingIndex: 42.8,
1171
+ mathIndex: undefined,
1286
1172
 
1287
1173
  // Academic benchmarks
1288
- mmluPro: 0.556,
1289
- gpqa: 0.428,
1290
- hle: 0.041,
1174
+ mmluPro: undefined,
1175
+ gpqa: 0.857,
1176
+ hle: 0.183,
1291
1177
 
1292
1178
  // Capabilities
1293
1179
  contextWindow: 8192,
@@ -1295,18 +1181,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1295
1181
  supportsVision: false,
1296
1182
 
1297
1183
  // Metadata
1298
- lastUpdated: "2026-06-01",
1299
- originalModel: "GPT-5 nano (minimal)",
1184
+ lastUpdated: "2026-07-01",
1185
+ originalModel: "Ring-2.6-1T",
1300
1186
  },
1301
- "o3-mini-high": {
1187
+ "doubao-seed-code": {
1302
1188
  // AA specific benchmarks
1303
- codingIndex: 17.3,
1304
- mathIndex: undefined,
1189
+ codingIndex: undefined,
1190
+ mathIndex: 79.3,
1305
1191
 
1306
1192
  // Academic benchmarks
1307
- mmluPro: 0.802,
1308
- gpqa: 0.773,
1309
- hle: 0.123,
1193
+ mmluPro: 0.854,
1194
+ gpqa: 0.764,
1195
+ hle: 0.133,
1310
1196
 
1311
1197
  // Capabilities
1312
1198
  contextWindow: 8192,
@@ -1314,18 +1200,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1314
1200
  supportsVision: false,
1315
1201
 
1316
1202
  // Metadata
1317
- lastUpdated: "2026-06-01",
1318
- originalModel: "o3-mini (high)",
1203
+ lastUpdated: "2026-07-01",
1204
+ originalModel: "Doubao Seed Code",
1319
1205
  },
1320
- "gpt-5.4-non-reasoning": {
1206
+ "o1": {
1321
1207
  // AA specific benchmarks
1322
- codingIndex: 41,
1208
+ codingIndex: 39.7,
1323
1209
  mathIndex: undefined,
1324
1210
 
1325
1211
  // Academic benchmarks
1326
- mmluPro: undefined,
1327
- gpqa: 0.748,
1328
- hle: 0.106,
1212
+ mmluPro: 0.841,
1213
+ gpqa: 0.747,
1214
+ hle: 0.077,
1329
1215
 
1330
1216
  // Capabilities
1331
1217
  contextWindow: 8192,
@@ -1333,18 +1219,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1333
1219
  supportsVision: false,
1334
1220
 
1335
1221
  // Metadata
1336
- lastUpdated: "2026-06-01",
1337
- originalModel: "GPT-5.4 (Non-reasoning)",
1222
+ lastUpdated: "2026-07-01",
1223
+ originalModel: "o1",
1338
1224
  },
1339
- "gpt-5.4-low": {
1225
+ "o1-preview": {
1340
1226
  // AA specific benchmarks
1341
- codingIndex: 45.6,
1227
+ codingIndex: 34,
1342
1228
  mathIndex: undefined,
1343
1229
 
1344
1230
  // Academic benchmarks
1345
1231
  mmluPro: undefined,
1346
- gpqa: 0.871,
1347
- hle: 0.289,
1232
+ gpqa: undefined,
1233
+ hle: undefined,
1348
1234
 
1349
1235
  // Capabilities
1350
1236
  contextWindow: 8192,
@@ -1352,18 +1238,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1352
1238
  supportsVision: false,
1353
1239
 
1354
1240
  // Metadata
1355
- lastUpdated: "2026-06-01",
1356
- originalModel: "GPT-5.4 (low)",
1241
+ lastUpdated: "2026-07-01",
1242
+ originalModel: "o1-preview",
1357
1243
  },
1358
- "gpt-4o-march-2025-chatgpt-4o-latest": {
1244
+ "o1-mini": {
1359
1245
  // AA specific benchmarks
1360
1246
  codingIndex: undefined,
1361
- mathIndex: 25.7,
1247
+ mathIndex: undefined,
1362
1248
 
1363
1249
  // Academic benchmarks
1364
- mmluPro: 0.803,
1365
- gpqa: 0.655,
1366
- hle: 0.05,
1250
+ mmluPro: 0.742,
1251
+ gpqa: 0.603,
1252
+ hle: 0.049,
1367
1253
 
1368
1254
  // Capabilities
1369
1255
  contextWindow: 8192,
@@ -1371,18 +1257,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1371
1257
  supportsVision: false,
1372
1258
 
1373
1259
  // Metadata
1374
- lastUpdated: "2026-06-01",
1375
- originalModel: "GPT-4o (March 2025, chatgpt-4o-latest)",
1260
+ lastUpdated: "2026-07-01",
1261
+ originalModel: "o1-mini",
1376
1262
  },
1377
- "gpt-4.5-preview": {
1263
+ "gpt-4o-aug-24": {
1378
1264
  // AA specific benchmarks
1379
1265
  codingIndex: undefined,
1380
1266
  mathIndex: undefined,
1381
1267
 
1382
1268
  // Academic benchmarks
1383
1269
  mmluPro: undefined,
1384
- gpqa: undefined,
1385
- hle: undefined,
1270
+ gpqa: 0.521,
1271
+ hle: 0.029,
1386
1272
 
1387
1273
  // Capabilities
1388
1274
  contextWindow: 8192,
@@ -1390,18 +1276,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1390
1276
  supportsVision: false,
1391
1277
 
1392
1278
  // Metadata
1393
- lastUpdated: "2026-06-01",
1394
- originalModel: "GPT-4.5 (Preview)",
1279
+ lastUpdated: "2026-07-01",
1280
+ originalModel: "GPT-4o (Aug '24)",
1395
1281
  },
1396
- "gpt-5-codex-high": {
1282
+ "gpt-4o-may-24": {
1397
1283
  // AA specific benchmarks
1398
- codingIndex: 38.9,
1399
- mathIndex: 98.7,
1284
+ codingIndex: 24.2,
1285
+ mathIndex: undefined,
1400
1286
 
1401
1287
  // Academic benchmarks
1402
- mmluPro: 0.865,
1403
- gpqa: 0.837,
1404
- hle: 0.256,
1288
+ mmluPro: 0.74,
1289
+ gpqa: 0.526,
1290
+ hle: 0.028,
1405
1291
 
1406
1292
  // Capabilities
1407
1293
  contextWindow: 8192,
@@ -1409,18 +1295,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1409
1295
  supportsVision: false,
1410
1296
 
1411
1297
  // Metadata
1412
- lastUpdated: "2026-06-01",
1413
- originalModel: "GPT-5 Codex (high)",
1298
+ lastUpdated: "2026-07-01",
1299
+ originalModel: "GPT-4o (May '24)",
1414
1300
  },
1415
- "gpt-5-medium": {
1301
+ "gpt-4-turbo": {
1416
1302
  // AA specific benchmarks
1417
- codingIndex: 38.9,
1418
- mathIndex: 91.7,
1303
+ codingIndex: 21.5,
1304
+ mathIndex: undefined,
1419
1305
 
1420
1306
  // Academic benchmarks
1421
- mmluPro: 0.867,
1422
- gpqa: 0.842,
1423
- hle: 0.235,
1307
+ mmluPro: 0.694,
1308
+ gpqa: undefined,
1309
+ hle: 0.033,
1424
1310
 
1425
1311
  // Capabilities
1426
1312
  contextWindow: 8192,
@@ -1428,18 +1314,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1428
1314
  supportsVision: false,
1429
1315
 
1430
1316
  // Metadata
1431
- lastUpdated: "2026-06-01",
1432
- originalModel: "GPT-5 (medium)",
1317
+ lastUpdated: "2026-07-01",
1318
+ originalModel: "GPT-4 Turbo",
1433
1319
  },
1434
- "gpt-4": {
1320
+ "gpt-4o-nov-24": {
1435
1321
  // AA specific benchmarks
1436
- codingIndex: 13.1,
1437
- mathIndex: undefined,
1322
+ codingIndex: undefined,
1323
+ mathIndex: 6,
1438
1324
 
1439
1325
  // Academic benchmarks
1440
- mmluPro: undefined,
1441
- gpqa: undefined,
1442
- hle: undefined,
1326
+ mmluPro: 0.748,
1327
+ gpqa: 0.543,
1328
+ hle: 0.033,
1443
1329
 
1444
1330
  // Capabilities
1445
1331
  contextWindow: 8192,
@@ -1447,18 +1333,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1447
1333
  supportsVision: false,
1448
1334
 
1449
1335
  // Metadata
1450
- lastUpdated: "2026-06-01",
1451
- originalModel: "GPT-4",
1336
+ lastUpdated: "2026-07-01",
1337
+ originalModel: "GPT-4o (Nov '24)",
1452
1338
  },
1453
- "gpt-5.2-non-reasoning": {
1339
+ "gpt-4o-mini": {
1454
1340
  // AA specific benchmarks
1455
- codingIndex: 34.7,
1456
- mathIndex: 51,
1341
+ codingIndex: 11.4,
1342
+ mathIndex: 14.7,
1457
1343
 
1458
1344
  // Academic benchmarks
1459
- mmluPro: 0.814,
1460
- gpqa: 0.712,
1461
- hle: 0.073,
1345
+ mmluPro: 0.648,
1346
+ gpqa: 0.426,
1347
+ hle: 0.04,
1462
1348
 
1463
1349
  // Capabilities
1464
1350
  contextWindow: 8192,
@@ -1466,18 +1352,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1466
1352
  supportsVision: false,
1467
1353
 
1468
1354
  // Metadata
1469
- lastUpdated: "2026-06-01",
1470
- originalModel: "GPT-5.2 (Non-reasoning)",
1355
+ lastUpdated: "2026-07-01",
1356
+ originalModel: "GPT-4o mini",
1471
1357
  },
1472
- "gpt-4.1-nano": {
1358
+ "gpt-3.5-turbo": {
1473
1359
  // AA specific benchmarks
1474
- codingIndex: 11.2,
1475
- mathIndex: 24,
1360
+ codingIndex: 10.7,
1361
+ mathIndex: undefined,
1476
1362
 
1477
1363
  // Academic benchmarks
1478
- mmluPro: 0.657,
1479
- gpqa: 0.512,
1480
- hle: 0.039,
1364
+ mmluPro: 0.462,
1365
+ gpqa: 0.297,
1366
+ hle: undefined,
1481
1367
 
1482
1368
  // Capabilities
1483
1369
  contextWindow: 8192,
@@ -1485,12 +1371,12 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1485
1371
  supportsVision: false,
1486
1372
 
1487
1373
  // Metadata
1488
- lastUpdated: "2026-06-01",
1489
- originalModel: "GPT-4.1 nano",
1374
+ lastUpdated: "2026-07-01",
1375
+ originalModel: "GPT-3.5 Turbo",
1490
1376
  },
1491
1377
  "gpt-5-low": {
1492
1378
  // AA specific benchmarks
1493
- codingIndex: 30.7,
1379
+ codingIndex: undefined,
1494
1380
  mathIndex: 83,
1495
1381
 
1496
1382
  // Academic benchmarks
@@ -1504,12 +1390,12 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1504
1390
  supportsVision: false,
1505
1391
 
1506
1392
  // Metadata
1507
- lastUpdated: "2026-06-01",
1393
+ lastUpdated: "2026-07-01",
1508
1394
  originalModel: "GPT-5 (low)",
1509
1395
  },
1510
1396
  "o4-mini-high": {
1511
1397
  // AA specific benchmarks
1512
- codingIndex: 25.6,
1398
+ codingIndex: undefined,
1513
1399
  mathIndex: 90.7,
1514
1400
 
1515
1401
  // Academic benchmarks
@@ -1523,12 +1409,12 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1523
1409
  supportsVision: false,
1524
1410
 
1525
1411
  // Metadata
1526
- lastUpdated: "2026-06-01",
1412
+ lastUpdated: "2026-07-01",
1527
1413
  originalModel: "o4-mini (high)",
1528
1414
  },
1529
1415
  "gpt-5.2-medium": {
1530
1416
  // AA specific benchmarks
1531
- codingIndex: 44.2,
1417
+ codingIndex: undefined,
1532
1418
  mathIndex: 96.7,
1533
1419
 
1534
1420
  // Academic benchmarks
@@ -1542,12 +1428,31 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1542
1428
  supportsVision: false,
1543
1429
 
1544
1430
  // Metadata
1545
- lastUpdated: "2026-06-01",
1431
+ lastUpdated: "2026-07-01",
1546
1432
  originalModel: "GPT-5.2 (medium)",
1547
1433
  },
1434
+ "gpt-5-mini-high": {
1435
+ // AA specific benchmarks
1436
+ codingIndex: undefined,
1437
+ mathIndex: 90.7,
1438
+
1439
+ // Academic benchmarks
1440
+ mmluPro: 0.837,
1441
+ gpqa: 0.828,
1442
+ hle: 0.197,
1443
+
1444
+ // Capabilities
1445
+ contextWindow: 8192,
1446
+ supportsReasoning: false,
1447
+ supportsVision: false,
1448
+
1449
+ // Metadata
1450
+ lastUpdated: "2026-07-01",
1451
+ originalModel: "GPT-5 mini (high)",
1452
+ },
1548
1453
  "gpt-5-nano-medium": {
1549
1454
  // AA specific benchmarks
1550
- codingIndex: 22.9,
1455
+ codingIndex: undefined,
1551
1456
  mathIndex: 78.3,
1552
1457
 
1553
1458
  // Academic benchmarks
@@ -1561,18 +1466,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1561
1466
  supportsVision: false,
1562
1467
 
1563
1468
  // Metadata
1564
- lastUpdated: "2026-06-01",
1469
+ lastUpdated: "2026-07-01",
1565
1470
  originalModel: "GPT-5 nano (medium)",
1566
1471
  },
1567
- "gpt-4.1-mini": {
1472
+ "gpt-5.1-codex-mini-high": {
1568
1473
  // AA specific benchmarks
1569
- codingIndex: 18.5,
1570
- mathIndex: 46.3,
1474
+ codingIndex: undefined,
1475
+ mathIndex: 91.7,
1571
1476
 
1572
1477
  // Academic benchmarks
1573
- mmluPro: 0.781,
1574
- gpqa: 0.664,
1575
- hle: 0.046,
1478
+ mmluPro: 0.82,
1479
+ gpqa: 0.813,
1480
+ hle: 0.169,
1576
1481
 
1577
1482
  // Capabilities
1578
1483
  contextWindow: 8192,
@@ -1580,18 +1485,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1580
1485
  supportsVision: false,
1581
1486
 
1582
1487
  // Metadata
1583
- lastUpdated: "2026-06-01",
1584
- originalModel: "GPT-4.1 mini",
1488
+ lastUpdated: "2026-07-01",
1489
+ originalModel: "GPT-5.1 Codex mini (high)",
1585
1490
  },
1586
- "gpt-5.4-xhigh": {
1491
+ "gpt-5-mini-minimal": {
1587
1492
  // AA specific benchmarks
1588
- codingIndex: 57.2,
1589
- mathIndex: undefined,
1493
+ codingIndex: undefined,
1494
+ mathIndex: 46.7,
1590
1495
 
1591
1496
  // Academic benchmarks
1592
- mmluPro: undefined,
1593
- gpqa: 0.92,
1594
- hle: 0.416,
1497
+ mmluPro: 0.775,
1498
+ gpqa: 0.687,
1499
+ hle: 0.05,
1595
1500
 
1596
1501
  // Capabilities
1597
1502
  contextWindow: 8192,
@@ -1599,8 +1504,8 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1599
1504
  supportsVision: false,
1600
1505
 
1601
1506
  // Metadata
1602
- lastUpdated: "2026-06-01",
1603
- originalModel: "GPT-5.4 (xhigh)",
1507
+ lastUpdated: "2026-07-01",
1508
+ originalModel: "GPT-5 mini (minimal)",
1604
1509
  },
1605
1510
  "gpt-4o-chatgpt": {
1606
1511
  // AA specific benchmarks
@@ -1618,18 +1523,18 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1618
1523
  supportsVision: false,
1619
1524
 
1620
1525
  // Metadata
1621
- lastUpdated: "2026-06-01",
1526
+ lastUpdated: "2026-07-01",
1622
1527
  originalModel: "GPT-4o (ChatGPT)",
1623
1528
  },
1624
- "gpt-5-mini-minimal": {
1529
+ "gpt-5.4-xhigh": {
1625
1530
  // AA specific benchmarks
1626
- codingIndex: 21.9,
1627
- mathIndex: 46.7,
1531
+ codingIndex: 71.1,
1532
+ mathIndex: undefined,
1628
1533
 
1629
1534
  // Academic benchmarks
1630
- mmluPro: 0.775,
1631
- gpqa: 0.687,
1632
- hle: 0.05,
1535
+ mmluPro: undefined,
1536
+ gpqa: 0.92,
1537
+ hle: 0.416,
1633
1538
 
1634
1539
  // Capabilities
1635
1540
  contextWindow: 8192,
@@ -1637,12 +1542,31 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1637
1542
  supportsVision: false,
1638
1543
 
1639
1544
  // Metadata
1640
- lastUpdated: "2026-06-01",
1641
- originalModel: "GPT-5 mini (minimal)",
1545
+ lastUpdated: "2026-07-01",
1546
+ originalModel: "GPT-5.4 (xhigh)",
1547
+ },
1548
+ "gpt-4.1-mini": {
1549
+ // AA specific benchmarks
1550
+ codingIndex: undefined,
1551
+ mathIndex: 46.3,
1552
+
1553
+ // Academic benchmarks
1554
+ mmluPro: 0.781,
1555
+ gpqa: 0.664,
1556
+ hle: 0.046,
1557
+
1558
+ // Capabilities
1559
+ contextWindow: 8192,
1560
+ supportsReasoning: false,
1561
+ supportsVision: false,
1562
+
1563
+ // Metadata
1564
+ lastUpdated: "2026-07-01",
1565
+ originalModel: "GPT-4.1 mini",
1642
1566
  },
1643
1567
  "gpt-5-mini-medium": {
1644
1568
  // AA specific benchmarks
1645
- codingIndex: 32.8,
1569
+ codingIndex: undefined,
1646
1570
  mathIndex: 85,
1647
1571
 
1648
1572
  // Academic benchmarks
@@ -1656,17 +1580,36 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1656
1580
  supportsVision: false,
1657
1581
 
1658
1582
  // Metadata
1659
- lastUpdated: "2026-06-01",
1583
+ lastUpdated: "2026-07-01",
1660
1584
  originalModel: "GPT-5 mini (medium)",
1661
1585
  },
1662
- "o1-pro": {
1586
+ "gpt-5-minimal": {
1587
+ // AA specific benchmarks
1588
+ codingIndex: undefined,
1589
+ mathIndex: 31.7,
1590
+
1591
+ // Academic benchmarks
1592
+ mmluPro: 0.806,
1593
+ gpqa: 0.673,
1594
+ hle: 0.054,
1595
+
1596
+ // Capabilities
1597
+ contextWindow: 8192,
1598
+ supportsReasoning: false,
1599
+ supportsVision: false,
1600
+
1601
+ // Metadata
1602
+ lastUpdated: "2026-07-01",
1603
+ originalModel: "GPT-5 (minimal)",
1604
+ },
1605
+ "o3-pro": {
1663
1606
  // AA specific benchmarks
1664
1607
  codingIndex: undefined,
1665
1608
  mathIndex: undefined,
1666
1609
 
1667
1610
  // Academic benchmarks
1668
1611
  mmluPro: undefined,
1669
- gpqa: undefined,
1612
+ gpqa: 0.845,
1670
1613
  hle: undefined,
1671
1614
 
1672
1615
  // Capabilities
@@ -1675,18 +1618,56 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1675
1618
  supportsVision: false,
1676
1619
 
1677
1620
  // Metadata
1678
- lastUpdated: "2026-06-01",
1679
- originalModel: "o1-pro",
1621
+ lastUpdated: "2026-07-01",
1622
+ originalModel: "o3-pro",
1680
1623
  },
1681
- "gpt-5-chatgpt": {
1624
+ "gpt-4.1-nano": {
1682
1625
  // AA specific benchmarks
1683
- codingIndex: 21.2,
1684
- mathIndex: 48.3,
1626
+ codingIndex: 11.1,
1627
+ mathIndex: 24,
1685
1628
 
1686
1629
  // Academic benchmarks
1687
- mmluPro: 0.82,
1688
- gpqa: 0.686,
1689
- hle: 0.058,
1630
+ mmluPro: 0.657,
1631
+ gpqa: 0.512,
1632
+ hle: 0.039,
1633
+
1634
+ // Capabilities
1635
+ contextWindow: 8192,
1636
+ supportsReasoning: false,
1637
+ supportsVision: false,
1638
+
1639
+ // Metadata
1640
+ lastUpdated: "2026-07-01",
1641
+ originalModel: "GPT-4.1 nano",
1642
+ },
1643
+ "gpt-5.1-non-reasoning": {
1644
+ // AA specific benchmarks
1645
+ codingIndex: undefined,
1646
+ mathIndex: 38,
1647
+
1648
+ // Academic benchmarks
1649
+ mmluPro: 0.801,
1650
+ gpqa: 0.643,
1651
+ hle: 0.052,
1652
+
1653
+ // Capabilities
1654
+ contextWindow: 8192,
1655
+ supportsReasoning: false,
1656
+ supportsVision: false,
1657
+
1658
+ // Metadata
1659
+ lastUpdated: "2026-07-01",
1660
+ originalModel: "GPT-5.1 (Non-reasoning)",
1661
+ },
1662
+ "gpt-5.2-non-reasoning": {
1663
+ // AA specific benchmarks
1664
+ codingIndex: undefined,
1665
+ mathIndex: 51,
1666
+
1667
+ // Academic benchmarks
1668
+ mmluPro: 0.814,
1669
+ gpqa: 0.712,
1670
+ hle: 0.073,
1690
1671
 
1691
1672
  // Capabilities
1692
1673
  contextWindow: 8192,
@@ -1694,18 +1675,37 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1694
1675
  supportsVision: false,
1695
1676
 
1696
1677
  // Metadata
1697
- lastUpdated: "2026-06-01",
1698
- originalModel: "GPT-5 (ChatGPT)",
1678
+ lastUpdated: "2026-07-01",
1679
+ originalModel: "GPT-5.2 (Non-reasoning)",
1699
1680
  },
1700
- "o3-mini": {
1681
+ "gpt-4": {
1701
1682
  // AA specific benchmarks
1702
- codingIndex: 17.9,
1683
+ codingIndex: 13.1,
1703
1684
  mathIndex: undefined,
1704
1685
 
1705
1686
  // Academic benchmarks
1706
- mmluPro: 0.791,
1707
- gpqa: 0.748,
1708
- hle: 0.087,
1687
+ mmluPro: undefined,
1688
+ gpqa: undefined,
1689
+ hle: undefined,
1690
+
1691
+ // Capabilities
1692
+ contextWindow: 8192,
1693
+ supportsReasoning: false,
1694
+ supportsVision: false,
1695
+
1696
+ // Metadata
1697
+ lastUpdated: "2026-07-01",
1698
+ originalModel: "GPT-4",
1699
+ },
1700
+ "gpt-5-nano-high": {
1701
+ // AA specific benchmarks
1702
+ codingIndex: undefined,
1703
+ mathIndex: 83.7,
1704
+
1705
+ // Academic benchmarks
1706
+ mmluPro: 0.78,
1707
+ gpqa: 0.676,
1708
+ hle: 0.082,
1709
1709
 
1710
1710
  // Capabilities
1711
1711
  contextWindow: 8192,
@@ -1713,7 +1713,7 @@ export const BENCHMARKS_CHUNK_2: Record<string, HardcodedBenchmark> = {
1713
1713
  supportsVision: false,
1714
1714
 
1715
1715
  // Metadata
1716
- lastUpdated: "2026-06-01",
1717
- originalModel: "o3-mini",
1716
+ lastUpdated: "2026-07-01",
1717
+ originalModel: "GPT-5 nano (high)",
1718
1718
  },
1719
1719
  };