@mariozechner/pi-ai 0.5.42 → 0.5.44

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 (19) hide show
  1. package/dist/models.generated.d.ts +453 -232
  2. package/dist/models.generated.d.ts.map +1 -1
  3. package/dist/models.generated.js +960 -739
  4. package/dist/models.generated.js.map +1 -1
  5. package/dist/providers/anthropic.js +24 -1
  6. package/dist/providers/anthropic.js.map +1 -1
  7. package/dist/providers/google.js +1 -1
  8. package/dist/providers/google.js.map +1 -1
  9. package/dist/providers/openai-responses.js +0 -4
  10. package/dist/providers/openai-responses.js.map +1 -1
  11. package/dist/providers/transorm-messages.d.ts.map +1 -1
  12. package/dist/providers/transorm-messages.js +46 -1
  13. package/dist/providers/transorm-messages.js.map +1 -1
  14. package/dist/utils/event-stream.js +3 -7
  15. package/dist/utils/event-stream.js.map +1 -1
  16. package/dist/utils/validation.d.ts.map +1 -1
  17. package/dist/utils/validation.js +25 -3
  18. package/dist/utils/validation.js.map +1 -1
  19. package/package.json +3 -3
package/dist/models.generated.js CHANGED
@@ -2,13 +2,13 @@
2
2
  // Do not edit manually - run 'npm run generate-models' to update
3
3
  export const MODELS = {
4
4
  anthropic: {
5
- "claude-3-7-sonnet-20250219": {
6
- id: "claude-3-7-sonnet-20250219",
7
- name: "Claude Sonnet 3.7",
5
+ "claude-3-5-sonnet-20241022": {
6
+ id: "claude-3-5-sonnet-20241022",
7
+ name: "Claude Sonnet 3.5 v2",
8
8
  api: "anthropic-messages",
9
9
  provider: "anthropic",
10
10
  baseUrl: "https://api.anthropic.com",
11
- reasoning: true,
11
+ reasoning: false,
12
12
  input: ["text", "image"],
13
13
  cost: {
14
14
  input: 3,
@@ -17,15 +17,32 @@ export const MODELS = {
17
17
  cacheWrite: 3.75,
18
18
  },
19
19
  contextWindow: 200000,
20
- maxTokens: 64000,
20
+ maxTokens: 8192,
21
21
  },
22
- "claude-opus-4-1-20250805": {
23
- id: "claude-opus-4-1-20250805",
24
- name: "Claude Opus 4.1",
22
+ "claude-3-5-sonnet-20240620": {
23
+ id: "claude-3-5-sonnet-20240620",
24
+ name: "Claude Sonnet 3.5",
25
25
  api: "anthropic-messages",
26
26
  provider: "anthropic",
27
27
  baseUrl: "https://api.anthropic.com",
28
- reasoning: true,
28
+ reasoning: false,
29
+ input: ["text", "image"],
30
+ cost: {
31
+ input: 3,
32
+ output: 15,
33
+ cacheRead: 0.3,
34
+ cacheWrite: 3.75,
35
+ },
36
+ contextWindow: 200000,
37
+ maxTokens: 8192,
38
+ },
39
+ "claude-3-opus-20240229": {
40
+ id: "claude-3-opus-20240229",
41
+ name: "Claude Opus 3",
42
+ api: "anthropic-messages",
43
+ provider: "anthropic",
44
+ baseUrl: "https://api.anthropic.com",
45
+ reasoning: false,
29
46
  input: ["text", "image"],
30
47
  cost: {
31
48
  input: 15,
@@ -34,41 +51,41 @@ export const MODELS = {
34
51
  cacheWrite: 18.75,
35
52
  },
36
53
  contextWindow: 200000,
37
- maxTokens: 32000,
54
+ maxTokens: 4096,
38
55
  },
39
- "claude-3-haiku-20240307": {
40
- id: "claude-3-haiku-20240307",
41
- name: "Claude Haiku 3",
56
+ "claude-sonnet-4-5-20250929": {
57
+ id: "claude-sonnet-4-5-20250929",
58
+ name: "Claude Sonnet 4.5",
42
59
  api: "anthropic-messages",
43
60
  provider: "anthropic",
44
61
  baseUrl: "https://api.anthropic.com",
45
- reasoning: false,
62
+ reasoning: true,
46
63
  input: ["text", "image"],
47
64
  cost: {
48
- input: 0.25,
49
- output: 1.25,
50
- cacheRead: 0.03,
51
- cacheWrite: 0.3,
65
+ input: 3,
66
+ output: 15,
67
+ cacheRead: 0.3,
68
+ cacheWrite: 3.75,
52
69
  },
53
70
  contextWindow: 200000,
54
- maxTokens: 4096,
71
+ maxTokens: 64000,
55
72
  },
56
- "claude-3-5-haiku-20241022": {
57
- id: "claude-3-5-haiku-20241022",
58
- name: "Claude Haiku 3.5",
73
+ "claude-sonnet-4-20250514": {
74
+ id: "claude-sonnet-4-20250514",
75
+ name: "Claude Sonnet 4",
59
76
  api: "anthropic-messages",
60
77
  provider: "anthropic",
61
78
  baseUrl: "https://api.anthropic.com",
62
- reasoning: false,
79
+ reasoning: true,
63
80
  input: ["text", "image"],
64
81
  cost: {
65
- input: 0.8,
66
- output: 4,
67
- cacheRead: 0.08,
68
- cacheWrite: 1,
82
+ input: 3,
83
+ output: 15,
84
+ cacheRead: 0.3,
85
+ cacheWrite: 3.75,
69
86
  },
70
87
  contextWindow: 200000,
71
- maxTokens: 8192,
88
+ maxTokens: 64000,
72
89
  },
73
90
  "claude-opus-4-20250514": {
74
91
  id: "claude-opus-4-20250514",
@@ -87,87 +104,87 @@ export const MODELS = {
87
104
  contextWindow: 200000,
88
105
  maxTokens: 32000,
89
106
  },
90
- "claude-3-5-sonnet-20241022": {
91
- id: "claude-3-5-sonnet-20241022",
92
- name: "Claude Sonnet 3.5 v2",
107
+ "claude-3-5-haiku-20241022": {
108
+ id: "claude-3-5-haiku-20241022",
109
+ name: "Claude Haiku 3.5",
93
110
  api: "anthropic-messages",
94
111
  provider: "anthropic",
95
112
  baseUrl: "https://api.anthropic.com",
96
113
  reasoning: false,
97
114
  input: ["text", "image"],
98
115
  cost: {
99
- input: 3,
100
- output: 15,
101
- cacheRead: 0.3,
102
- cacheWrite: 3.75,
116
+ input: 0.8,
117
+ output: 4,
118
+ cacheRead: 0.08,
119
+ cacheWrite: 1,
103
120
  },
104
121
  contextWindow: 200000,
105
122
  maxTokens: 8192,
106
123
  },
107
- "claude-3-5-sonnet-20240620": {
108
- id: "claude-3-5-sonnet-20240620",
109
- name: "Claude Sonnet 3.5",
124
+ "claude-3-haiku-20240307": {
125
+ id: "claude-3-haiku-20240307",
126
+ name: "Claude Haiku 3",
110
127
  api: "anthropic-messages",
111
128
  provider: "anthropic",
112
129
  baseUrl: "https://api.anthropic.com",
113
130
  reasoning: false,
114
131
  input: ["text", "image"],
115
132
  cost: {
116
- input: 3,
117
- output: 15,
118
- cacheRead: 0.3,
119
- cacheWrite: 3.75,
133
+ input: 0.25,
134
+ output: 1.25,
135
+ cacheRead: 0.03,
136
+ cacheWrite: 0.3,
120
137
  },
121
138
  contextWindow: 200000,
122
- maxTokens: 8192,
139
+ maxTokens: 4096,
123
140
  },
124
- "claude-3-sonnet-20240229": {
125
- id: "claude-3-sonnet-20240229",
126
- name: "Claude Sonnet 3",
141
+ "claude-3-7-sonnet-20250219": {
142
+ id: "claude-3-7-sonnet-20250219",
143
+ name: "Claude Sonnet 3.7",
127
144
  api: "anthropic-messages",
128
145
  provider: "anthropic",
129
146
  baseUrl: "https://api.anthropic.com",
130
- reasoning: false,
147
+ reasoning: true,
131
148
  input: ["text", "image"],
132
149
  cost: {
133
150
  input: 3,
134
151
  output: 15,
135
152
  cacheRead: 0.3,
136
- cacheWrite: 0.3,
153
+ cacheWrite: 3.75,
137
154
  },
138
155
  contextWindow: 200000,
139
- maxTokens: 4096,
156
+ maxTokens: 64000,
140
157
  },
141
- "claude-sonnet-4-20250514": {
142
- id: "claude-sonnet-4-20250514",
143
- name: "Claude Sonnet 4",
158
+ "claude-opus-4-1-20250805": {
159
+ id: "claude-opus-4-1-20250805",
160
+ name: "Claude Opus 4.1",
144
161
  api: "anthropic-messages",
145
162
  provider: "anthropic",
146
163
  baseUrl: "https://api.anthropic.com",
147
164
  reasoning: true,
148
165
  input: ["text", "image"],
149
166
  cost: {
150
- input: 3,
151
- output: 15,
152
- cacheRead: 0.3,
153
- cacheWrite: 3.75,
167
+ input: 15,
168
+ output: 75,
169
+ cacheRead: 1.5,
170
+ cacheWrite: 18.75,
154
171
  },
155
172
  contextWindow: 200000,
156
- maxTokens: 64000,
173
+ maxTokens: 32000,
157
174
  },
158
- "claude-3-opus-20240229": {
159
- id: "claude-3-opus-20240229",
160
- name: "Claude Opus 3",
175
+ "claude-3-sonnet-20240229": {
176
+ id: "claude-3-sonnet-20240229",
177
+ name: "Claude Sonnet 3",
161
178
  api: "anthropic-messages",
162
179
  provider: "anthropic",
163
180
  baseUrl: "https://api.anthropic.com",
164
181
  reasoning: false,
165
182
  input: ["text", "image"],
166
183
  cost: {
167
- input: 15,
168
- output: 75,
169
- cacheRead: 1.5,
170
- cacheWrite: 18.75,
184
+ input: 3,
185
+ output: 15,
186
+ cacheRead: 0.3,
187
+ cacheWrite: 0.3,
171
188
  },
172
189
  contextWindow: 200000,
173
190
  maxTokens: 4096,
@@ -191,39 +208,73 @@ export const MODELS = {
191
208
  contextWindow: 1048576,
192
209
  maxTokens: 65536,
193
210
  },
194
- "gemini-2.5-pro": {
195
- id: "gemini-2.5-pro",
196
- name: "Gemini 2.5 Pro",
211
+ "gemini-flash-lite-latest": {
212
+ id: "gemini-flash-lite-latest",
213
+ name: "Gemini Flash-Lite Latest",
197
214
  api: "google-generative-ai",
198
215
  provider: "google",
199
216
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
200
217
  reasoning: true,
201
218
  input: ["text", "image"],
202
219
  cost: {
203
- input: 1.25,
204
- output: 10,
205
- cacheRead: 0.31,
220
+ input: 0.1,
221
+ output: 0.4,
222
+ cacheRead: 0.025,
206
223
  cacheWrite: 0,
207
224
  },
208
225
  contextWindow: 1048576,
209
226
  maxTokens: 65536,
210
227
  },
211
- "gemini-1.5-flash": {
212
- id: "gemini-1.5-flash",
213
- name: "Gemini 1.5 Flash",
228
+ "gemini-2.5-flash": {
229
+ id: "gemini-2.5-flash",
230
+ name: "Gemini 2.5 Flash",
214
231
  api: "google-generative-ai",
215
232
  provider: "google",
216
233
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
217
- reasoning: false,
234
+ reasoning: true,
218
235
  input: ["text", "image"],
219
236
  cost: {
220
- input: 0.075,
221
- output: 0.3,
222
- cacheRead: 0.01875,
237
+ input: 0.3,
238
+ output: 2.5,
239
+ cacheRead: 0.075,
223
240
  cacheWrite: 0,
224
241
  },
225
- contextWindow: 1000000,
226
- maxTokens: 8192,
242
+ contextWindow: 1048576,
243
+ maxTokens: 65536,
244
+ },
245
+ "gemini-flash-latest": {
246
+ id: "gemini-flash-latest",
247
+ name: "Gemini Flash Latest",
248
+ api: "google-generative-ai",
249
+ provider: "google",
250
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
251
+ reasoning: true,
252
+ input: ["text", "image"],
253
+ cost: {
254
+ input: 0.15,
255
+ output: 0.6,
256
+ cacheRead: 0.0375,
257
+ cacheWrite: 0,
258
+ },
259
+ contextWindow: 1048576,
260
+ maxTokens: 65536,
261
+ },
262
+ "gemini-2.5-pro-preview-05-06": {
263
+ id: "gemini-2.5-pro-preview-05-06",
264
+ name: "Gemini 2.5 Pro Preview 05-06",
265
+ api: "google-generative-ai",
266
+ provider: "google",
267
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
268
+ reasoning: true,
269
+ input: ["text", "image"],
270
+ cost: {
271
+ input: 1.25,
272
+ output: 10,
273
+ cacheRead: 0.31,
274
+ cacheWrite: 0,
275
+ },
276
+ contextWindow: 1048576,
277
+ maxTokens: 65536,
227
278
  },
228
279
  "gemini-2.0-flash-lite": {
229
280
  id: "gemini-2.0-flash-lite",
@@ -242,52 +293,52 @@ export const MODELS = {
242
293
  contextWindow: 1048576,
243
294
  maxTokens: 8192,
244
295
  },
245
- "gemini-1.5-pro": {
246
- id: "gemini-1.5-pro",
247
- name: "Gemini 1.5 Pro",
296
+ "gemini-live-2.5-flash-preview-native-audio": {
297
+ id: "gemini-live-2.5-flash-preview-native-audio",
298
+ name: "Gemini Live 2.5 Flash Preview Native Audio",
248
299
  api: "google-generative-ai",
249
300
  provider: "google",
250
301
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
251
- reasoning: false,
252
- input: ["text", "image"],
302
+ reasoning: true,
303
+ input: ["text"],
253
304
  cost: {
254
- input: 1.25,
255
- output: 5,
256
- cacheRead: 0.3125,
305
+ input: 0.5,
306
+ output: 2,
307
+ cacheRead: 0,
257
308
  cacheWrite: 0,
258
309
  },
259
- contextWindow: 1000000,
260
- maxTokens: 8192,
310
+ contextWindow: 131072,
311
+ maxTokens: 65536,
261
312
  },
262
- "gemini-1.5-flash-8b": {
263
- id: "gemini-1.5-flash-8b",
264
- name: "Gemini 1.5 Flash-8B",
313
+ "gemini-2.0-flash": {
314
+ id: "gemini-2.0-flash",
315
+ name: "Gemini 2.0 Flash",
265
316
  api: "google-generative-ai",
266
317
  provider: "google",
267
318
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
268
319
  reasoning: false,
269
320
  input: ["text", "image"],
270
321
  cost: {
271
- input: 0.0375,
272
- output: 0.15,
273
- cacheRead: 0.01,
322
+ input: 0.1,
323
+ output: 0.4,
324
+ cacheRead: 0.025,
274
325
  cacheWrite: 0,
275
326
  },
276
- contextWindow: 1000000,
327
+ contextWindow: 1048576,
277
328
  maxTokens: 8192,
278
329
  },
279
- "gemini-2.5-flash": {
280
- id: "gemini-2.5-flash",
281
- name: "Gemini 2.5 Flash",
330
+ "gemini-2.5-flash-lite": {
331
+ id: "gemini-2.5-flash-lite",
332
+ name: "Gemini 2.5 Flash Lite",
282
333
  api: "google-generative-ai",
283
334
  provider: "google",
284
335
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
285
336
  reasoning: true,
286
337
  input: ["text", "image"],
287
338
  cost: {
288
- input: 0.3,
289
- output: 2.5,
290
- cacheRead: 0.075,
339
+ input: 0.1,
340
+ output: 0.4,
341
+ cacheRead: 0.025,
291
342
  cacheWrite: 0,
292
343
  },
293
344
  contextWindow: 1048576,
@@ -310,43 +361,111 @@ export const MODELS = {
310
361
  contextWindow: 1048576,
311
362
  maxTokens: 65536,
312
363
  },
313
- "gemini-2.5-pro-preview-05-06": {
314
- id: "gemini-2.5-pro-preview-05-06",
315
- name: "Gemini 2.5 Pro Preview 05-06",
364
+ "gemini-2.5-flash-lite-preview-06-17": {
365
+ id: "gemini-2.5-flash-lite-preview-06-17",
366
+ name: "Gemini 2.5 Flash Lite Preview 06-17",
316
367
  api: "google-generative-ai",
317
368
  provider: "google",
318
369
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
319
370
  reasoning: true,
320
371
  input: ["text", "image"],
321
372
  cost: {
322
- input: 1.25,
323
- output: 10,
324
- cacheRead: 0.31,
373
+ input: 0.1,
374
+ output: 0.4,
375
+ cacheRead: 0.025,
325
376
  cacheWrite: 0,
326
377
  },
327
378
  contextWindow: 1048576,
328
379
  maxTokens: 65536,
329
380
  },
330
- "gemini-2.0-flash": {
331
- id: "gemini-2.0-flash",
332
- name: "Gemini 2.0 Flash",
381
+ "gemini-2.5-flash-preview-09-2025": {
382
+ id: "gemini-2.5-flash-preview-09-2025",
383
+ name: "Gemini 2.5 Flash Preview 09-25",
333
384
  api: "google-generative-ai",
334
385
  provider: "google",
335
386
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
336
- reasoning: false,
387
+ reasoning: true,
337
388
  input: ["text", "image"],
338
389
  cost: {
339
- input: 0.1,
340
- output: 0.4,
341
- cacheRead: 0.025,
390
+ input: 0.15,
391
+ output: 0.6,
392
+ cacheRead: 0.0375,
393
+ cacheWrite: 0,
394
+ },
395
+ contextWindow: 1048576,
396
+ maxTokens: 65536,
397
+ },
398
+ "gemini-2.5-flash-preview-04-17": {
399
+ id: "gemini-2.5-flash-preview-04-17",
400
+ name: "Gemini 2.5 Flash Preview 04-17",
401
+ api: "google-generative-ai",
402
+ provider: "google",
403
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
404
+ reasoning: true,
405
+ input: ["text", "image"],
406
+ cost: {
407
+ input: 0.15,
408
+ output: 0.6,
409
+ cacheRead: 0.0375,
410
+ cacheWrite: 0,
411
+ },
412
+ contextWindow: 1048576,
413
+ maxTokens: 65536,
414
+ },
415
+ "gemini-2.5-pro": {
416
+ id: "gemini-2.5-pro",
417
+ name: "Gemini 2.5 Pro",
418
+ api: "google-generative-ai",
419
+ provider: "google",
420
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
421
+ reasoning: true,
422
+ input: ["text", "image"],
423
+ cost: {
424
+ input: 1.25,
425
+ output: 10,
426
+ cacheRead: 0.31,
342
427
  cacheWrite: 0,
343
428
  },
344
429
  contextWindow: 1048576,
430
+ maxTokens: 65536,
431
+ },
432
+ "gemini-1.5-flash": {
433
+ id: "gemini-1.5-flash",
434
+ name: "Gemini 1.5 Flash",
435
+ api: "google-generative-ai",
436
+ provider: "google",
437
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
438
+ reasoning: false,
439
+ input: ["text", "image"],
440
+ cost: {
441
+ input: 0.075,
442
+ output: 0.3,
443
+ cacheRead: 0.01875,
444
+ cacheWrite: 0,
445
+ },
446
+ contextWindow: 1000000,
345
447
  maxTokens: 8192,
346
448
  },
347
- "gemini-2.5-flash-lite-preview-06-17": {
348
- id: "gemini-2.5-flash-lite-preview-06-17",
349
- name: "Gemini 2.5 Flash Lite Preview 06-17",
449
+ "gemini-1.5-flash-8b": {
450
+ id: "gemini-1.5-flash-8b",
451
+ name: "Gemini 1.5 Flash-8B",
452
+ api: "google-generative-ai",
453
+ provider: "google",
454
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
455
+ reasoning: false,
456
+ input: ["text", "image"],
457
+ cost: {
458
+ input: 0.0375,
459
+ output: 0.15,
460
+ cacheRead: 0.01,
461
+ cacheWrite: 0,
462
+ },
463
+ contextWindow: 1000000,
464
+ maxTokens: 8192,
465
+ },
466
+ "gemini-2.5-flash-lite-preview-09-2025": {
467
+ id: "gemini-2.5-flash-lite-preview-09-2025",
468
+ name: "Gemini 2.5 Flash Lite Preview 09-25",
350
469
  api: "google-generative-ai",
351
470
  provider: "google",
352
471
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
@@ -358,96 +477,147 @@ export const MODELS = {
358
477
  cacheRead: 0.025,
359
478
  cacheWrite: 0,
360
479
  },
361
- contextWindow: 65536,
480
+ contextWindow: 1048576,
362
481
  maxTokens: 65536,
363
482
  },
364
- "gemini-2.5-flash-preview-04-17": {
365
- id: "gemini-2.5-flash-preview-04-17",
366
- name: "Gemini 2.5 Flash Preview 04-17",
483
+ "gemini-1.5-pro": {
484
+ id: "gemini-1.5-pro",
485
+ name: "Gemini 1.5 Pro",
367
486
  api: "google-generative-ai",
368
487
  provider: "google",
369
488
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
370
- reasoning: true,
489
+ reasoning: false,
371
490
  input: ["text", "image"],
372
491
  cost: {
373
- input: 0.15,
374
- output: 0.6,
375
- cacheRead: 0.0375,
492
+ input: 1.25,
493
+ output: 5,
494
+ cacheRead: 0.3125,
376
495
  cacheWrite: 0,
377
496
  },
378
- contextWindow: 1048576,
379
- maxTokens: 65536,
497
+ contextWindow: 1000000,
498
+ maxTokens: 8192,
380
499
  },
381
500
  },
382
501
  openai: {
383
- "gpt-5-nano": {
384
- id: "gpt-5-nano",
385
- name: "GPT-5 Nano",
502
+ "gpt-4.1-nano": {
503
+ id: "gpt-4.1-nano",
504
+ name: "GPT-4.1 nano",
386
505
  api: "openai-responses",
387
506
  provider: "openai",
388
507
  baseUrl: "https://api.openai.com/v1",
389
- reasoning: true,
508
+ reasoning: false,
390
509
  input: ["text", "image"],
391
510
  cost: {
392
- input: 0.05,
511
+ input: 0.1,
393
512
  output: 0.4,
394
- cacheRead: 0.01,
513
+ cacheRead: 0.03,
395
514
  cacheWrite: 0,
396
515
  },
397
- contextWindow: 400000,
398
- maxTokens: 128000,
516
+ contextWindow: 1047576,
517
+ maxTokens: 32768,
399
518
  },
400
- "o3-pro": {
401
- id: "o3-pro",
402
- name: "o3-pro",
519
+ "gpt-4": {
520
+ id: "gpt-4",
521
+ name: "GPT-4",
522
+ api: "openai-responses",
523
+ provider: "openai",
524
+ baseUrl: "https://api.openai.com/v1",
525
+ reasoning: false,
526
+ input: ["text"],
527
+ cost: {
528
+ input: 30,
529
+ output: 60,
530
+ cacheRead: 0,
531
+ cacheWrite: 0,
532
+ },
533
+ contextWindow: 8192,
534
+ maxTokens: 8192,
535
+ },
536
+ "o1-pro": {
537
+ id: "o1-pro",
538
+ name: "o1-pro",
403
539
  api: "openai-responses",
404
540
  provider: "openai",
405
541
  baseUrl: "https://api.openai.com/v1",
406
542
  reasoning: true,
407
543
  input: ["text", "image"],
408
544
  cost: {
409
- input: 20,
410
- output: 80,
545
+ input: 150,
546
+ output: 600,
411
547
  cacheRead: 0,
412
548
  cacheWrite: 0,
413
549
  },
414
550
  contextWindow: 200000,
415
551
  maxTokens: 100000,
416
552
  },
417
- "codex-mini-latest": {
418
- id: "codex-mini-latest",
419
- name: "Codex Mini",
553
+ "gpt-4o-2024-05-13": {
554
+ id: "gpt-4o-2024-05-13",
555
+ name: "GPT-4o (2024-05-13)",
420
556
  api: "openai-responses",
421
557
  provider: "openai",
422
558
  baseUrl: "https://api.openai.com/v1",
423
- reasoning: true,
424
- input: ["text"],
559
+ reasoning: false,
560
+ input: ["text", "image"],
425
561
  cost: {
426
- input: 1.5,
427
- output: 6,
428
- cacheRead: 0.375,
562
+ input: 5,
563
+ output: 15,
564
+ cacheRead: 0,
429
565
  cacheWrite: 0,
430
566
  },
431
- contextWindow: 200000,
432
- maxTokens: 100000,
567
+ contextWindow: 128000,
568
+ maxTokens: 4096,
433
569
  },
434
- "gpt-4.1": {
435
- id: "gpt-4.1",
436
- name: "GPT-4.1",
570
+ "gpt-4o-2024-08-06": {
571
+ id: "gpt-4o-2024-08-06",
572
+ name: "GPT-4o (2024-08-06)",
437
573
  api: "openai-responses",
438
574
  provider: "openai",
439
575
  baseUrl: "https://api.openai.com/v1",
440
576
  reasoning: false,
441
577
  input: ["text", "image"],
442
578
  cost: {
443
- input: 2,
444
- output: 8,
445
- cacheRead: 0.5,
579
+ input: 2.5,
580
+ output: 10,
581
+ cacheRead: 1.25,
582
+ cacheWrite: 0,
583
+ },
584
+ contextWindow: 128000,
585
+ maxTokens: 16384,
586
+ },
587
+ "gpt-4.1-mini": {
588
+ id: "gpt-4.1-mini",
589
+ name: "GPT-4.1 mini",
590
+ api: "openai-responses",
591
+ provider: "openai",
592
+ baseUrl: "https://api.openai.com/v1",
593
+ reasoning: false,
594
+ input: ["text", "image"],
595
+ cost: {
596
+ input: 0.4,
597
+ output: 1.6,
598
+ cacheRead: 0.1,
446
599
  cacheWrite: 0,
447
600
  },
448
601
  contextWindow: 1047576,
449
602
  maxTokens: 32768,
450
603
  },
604
+ "o3-deep-research": {
605
+ id: "o3-deep-research",
606
+ name: "o3-deep-research",
607
+ api: "openai-responses",
608
+ provider: "openai",
609
+ baseUrl: "https://api.openai.com/v1",
610
+ reasoning: true,
611
+ input: ["text", "image"],
612
+ cost: {
613
+ input: 10,
614
+ output: 40,
615
+ cacheRead: 2.5,
616
+ cacheWrite: 0,
617
+ },
618
+ contextWindow: 200000,
619
+ maxTokens: 100000,
620
+ },
451
621
  "gpt-4-turbo": {
452
622
  id: "gpt-4-turbo",
453
623
  name: "GPT-4 Turbo",
@@ -465,86 +635,137 @@ export const MODELS = {
465
635
  contextWindow: 128000,
466
636
  maxTokens: 4096,
467
637
  },
468
- "o1": {
469
- id: "o1",
470
- name: "o1",
638
+ "o3-mini": {
639
+ id: "o3-mini",
640
+ name: "o3-mini",
471
641
  api: "openai-responses",
472
642
  provider: "openai",
473
643
  baseUrl: "https://api.openai.com/v1",
474
644
  reasoning: true,
475
- input: ["text", "image"],
645
+ input: ["text"],
476
646
  cost: {
477
- input: 15,
478
- output: 60,
479
- cacheRead: 7.5,
647
+ input: 1.1,
648
+ output: 4.4,
649
+ cacheRead: 0.55,
480
650
  cacheWrite: 0,
481
651
  },
482
652
  contextWindow: 200000,
483
653
  maxTokens: 100000,
484
654
  },
485
- "o3-deep-research": {
486
- id: "o3-deep-research",
487
- name: "o3-deep-research",
655
+ "codex-mini-latest": {
656
+ id: "codex-mini-latest",
657
+ name: "Codex Mini",
488
658
  api: "openai-responses",
489
659
  provider: "openai",
490
660
  baseUrl: "https://api.openai.com/v1",
491
661
  reasoning: true,
492
- input: ["text", "image"],
662
+ input: ["text"],
493
663
  cost: {
494
- input: 10,
495
- output: 40,
496
- cacheRead: 2.5,
664
+ input: 1.5,
665
+ output: 6,
666
+ cacheRead: 0.375,
497
667
  cacheWrite: 0,
498
668
  },
499
669
  contextWindow: 200000,
500
670
  maxTokens: 100000,
501
671
  },
502
- "gpt-5": {
503
- id: "gpt-5",
504
- name: "GPT-5",
672
+ "gpt-5-nano": {
673
+ id: "gpt-5-nano",
674
+ name: "GPT-5 Nano",
505
675
  api: "openai-responses",
506
676
  provider: "openai",
507
677
  baseUrl: "https://api.openai.com/v1",
508
678
  reasoning: true,
509
679
  input: ["text", "image"],
510
680
  cost: {
511
- input: 1.25,
512
- output: 10,
513
- cacheRead: 0.13,
681
+ input: 0.05,
682
+ output: 0.4,
683
+ cacheRead: 0.01,
684
+ cacheWrite: 0,
685
+ },
686
+ contextWindow: 400000,
687
+ maxTokens: 128000,
688
+ },
689
+ "gpt-5-codex": {
690
+ id: "gpt-5-codex",
691
+ name: "GPT-5-Codex",
692
+ api: "openai-responses",
693
+ provider: "openai",
694
+ baseUrl: "https://api.openai.com/v1",
695
+ reasoning: true,
696
+ input: ["text", "image"],
697
+ cost: {
698
+ input: 0,
699
+ output: 0,
700
+ cacheRead: 0,
514
701
  cacheWrite: 0,
515
702
  },
516
703
  contextWindow: 400000,
517
704
  maxTokens: 128000,
518
705
  },
519
- "o1-pro": {
520
- id: "o1-pro",
521
- name: "o1-pro",
706
+ "gpt-4o": {
707
+ id: "gpt-4o",
708
+ name: "GPT-4o",
709
+ api: "openai-responses",
710
+ provider: "openai",
711
+ baseUrl: "https://api.openai.com/v1",
712
+ reasoning: false,
713
+ input: ["text", "image"],
714
+ cost: {
715
+ input: 2.5,
716
+ output: 10,
717
+ cacheRead: 1.25,
718
+ cacheWrite: 0,
719
+ },
720
+ contextWindow: 128000,
721
+ maxTokens: 16384,
722
+ },
723
+ "gpt-4.1": {
724
+ id: "gpt-4.1",
725
+ name: "GPT-4.1",
726
+ api: "openai-responses",
727
+ provider: "openai",
728
+ baseUrl: "https://api.openai.com/v1",
729
+ reasoning: false,
730
+ input: ["text", "image"],
731
+ cost: {
732
+ input: 2,
733
+ output: 8,
734
+ cacheRead: 0.5,
735
+ cacheWrite: 0,
736
+ },
737
+ contextWindow: 1047576,
738
+ maxTokens: 32768,
739
+ },
740
+ "o4-mini": {
741
+ id: "o4-mini",
742
+ name: "o4-mini",
522
743
  api: "openai-responses",
523
744
  provider: "openai",
524
745
  baseUrl: "https://api.openai.com/v1",
525
746
  reasoning: true,
526
747
  input: ["text", "image"],
527
748
  cost: {
528
- input: 150,
529
- output: 600,
530
- cacheRead: 0,
749
+ input: 1.1,
750
+ output: 4.4,
751
+ cacheRead: 0.28,
531
752
  cacheWrite: 0,
532
753
  },
533
754
  contextWindow: 200000,
534
755
  maxTokens: 100000,
535
756
  },
536
- "o3": {
537
- id: "o3",
538
- name: "o3",
757
+ "o1": {
758
+ id: "o1",
759
+ name: "o1",
539
760
  api: "openai-responses",
540
761
  provider: "openai",
541
762
  baseUrl: "https://api.openai.com/v1",
542
763
  reasoning: true,
543
764
  input: ["text", "image"],
544
765
  cost: {
545
- input: 2,
546
- output: 8,
547
- cacheRead: 0.5,
766
+ input: 15,
767
+ output: 60,
768
+ cacheRead: 7.5,
548
769
  cacheWrite: 0,
549
770
  },
550
771
  contextWindow: 200000,
@@ -567,141 +788,107 @@ export const MODELS = {
567
788
  contextWindow: 400000,
568
789
  maxTokens: 128000,
569
790
  },
570
- "o4-mini-deep-research": {
571
- id: "o4-mini-deep-research",
572
- name: "o4-mini-deep-research",
791
+ "o3-pro": {
792
+ id: "o3-pro",
793
+ name: "o3-pro",
573
794
  api: "openai-responses",
574
795
  provider: "openai",
575
796
  baseUrl: "https://api.openai.com/v1",
576
797
  reasoning: true,
577
798
  input: ["text", "image"],
578
799
  cost: {
579
- input: 2,
580
- output: 8,
581
- cacheRead: 0.5,
800
+ input: 20,
801
+ output: 80,
802
+ cacheRead: 0,
582
803
  cacheWrite: 0,
583
804
  },
584
805
  contextWindow: 200000,
585
806
  maxTokens: 100000,
586
807
  },
587
- "gpt-4o-mini": {
588
- id: "gpt-4o-mini",
589
- name: "GPT-4o mini",
808
+ "gpt-4o-2024-11-20": {
809
+ id: "gpt-4o-2024-11-20",
810
+ name: "GPT-4o (2024-11-20)",
590
811
  api: "openai-responses",
591
812
  provider: "openai",
592
813
  baseUrl: "https://api.openai.com/v1",
593
814
  reasoning: false,
594
815
  input: ["text", "image"],
595
816
  cost: {
596
- input: 0.15,
597
- output: 0.6,
598
- cacheRead: 0.08,
817
+ input: 2.5,
818
+ output: 10,
819
+ cacheRead: 1.25,
599
820
  cacheWrite: 0,
600
821
  },
601
822
  contextWindow: 128000,
602
823
  maxTokens: 16384,
603
824
  },
604
- "gpt-4.1-nano": {
605
- id: "gpt-4.1-nano",
606
- name: "GPT-4.1 nano",
825
+ "o3": {
826
+ id: "o3",
827
+ name: "o3",
607
828
  api: "openai-responses",
608
829
  provider: "openai",
609
830
  baseUrl: "https://api.openai.com/v1",
610
- reasoning: false,
831
+ reasoning: true,
611
832
  input: ["text", "image"],
612
833
  cost: {
613
- input: 0.1,
614
- output: 0.4,
615
- cacheRead: 0.03,
834
+ input: 2,
835
+ output: 8,
836
+ cacheRead: 0.5,
616
837
  cacheWrite: 0,
617
838
  },
618
- contextWindow: 1047576,
619
- maxTokens: 32768,
839
+ contextWindow: 200000,
840
+ maxTokens: 100000,
620
841
  },
621
- "gpt-4.1-mini": {
622
- id: "gpt-4.1-mini",
623
- name: "GPT-4.1 mini",
842
+ "o4-mini-deep-research": {
843
+ id: "o4-mini-deep-research",
844
+ name: "o4-mini-deep-research",
624
845
  api: "openai-responses",
625
846
  provider: "openai",
626
847
  baseUrl: "https://api.openai.com/v1",
627
- reasoning: false,
848
+ reasoning: true,
628
849
  input: ["text", "image"],
629
850
  cost: {
630
- input: 0.4,
631
- output: 1.6,
632
- cacheRead: 0.1,
851
+ input: 2,
852
+ output: 8,
853
+ cacheRead: 0.5,
633
854
  cacheWrite: 0,
634
855
  },
635
- contextWindow: 1047576,
636
- maxTokens: 32768,
856
+ contextWindow: 200000,
857
+ maxTokens: 100000,
637
858
  },
638
- "gpt-4o": {
639
- id: "gpt-4o",
640
- name: "GPT-4o",
859
+ "gpt-4o-mini": {
860
+ id: "gpt-4o-mini",
861
+ name: "GPT-4o mini",
641
862
  api: "openai-responses",
642
863
  provider: "openai",
643
864
  baseUrl: "https://api.openai.com/v1",
644
865
  reasoning: false,
645
866
  input: ["text", "image"],
646
867
  cost: {
647
- input: 2.5,
648
- output: 10,
649
- cacheRead: 1.25,
868
+ input: 0.15,
869
+ output: 0.6,
870
+ cacheRead: 0.08,
650
871
  cacheWrite: 0,
651
872
  },
652
873
  contextWindow: 128000,
653
874
  maxTokens: 16384,
654
875
  },
655
- "gpt-4": {
656
- id: "gpt-4",
657
- name: "GPT-4",
658
- api: "openai-responses",
659
- provider: "openai",
660
- baseUrl: "https://api.openai.com/v1",
661
- reasoning: false,
662
- input: ["text"],
663
- cost: {
664
- input: 30,
665
- output: 60,
666
- cacheRead: 0,
667
- cacheWrite: 0,
668
- },
669
- contextWindow: 8192,
670
- maxTokens: 8192,
671
- },
672
- "o4-mini": {
673
- id: "o4-mini",
674
- name: "o4-mini",
876
+ "gpt-5": {
877
+ id: "gpt-5",
878
+ name: "GPT-5",
675
879
  api: "openai-responses",
676
880
  provider: "openai",
677
881
  baseUrl: "https://api.openai.com/v1",
678
882
  reasoning: true,
679
883
  input: ["text", "image"],
680
884
  cost: {
681
- input: 1.1,
682
- output: 4.4,
683
- cacheRead: 0.28,
684
- cacheWrite: 0,
685
- },
686
- contextWindow: 200000,
687
- maxTokens: 100000,
688
- },
689
- "o3-mini": {
690
- id: "o3-mini",
691
- name: "o3-mini",
692
- api: "openai-responses",
693
- provider: "openai",
694
- baseUrl: "https://api.openai.com/v1",
695
- reasoning: true,
696
- input: ["text"],
697
- cost: {
698
- input: 1.1,
699
- output: 4.4,
700
- cacheRead: 0.55,
885
+ input: 1.25,
886
+ output: 10,
887
+ cacheRead: 0.13,
701
888
  cacheWrite: 0,
702
889
  },
703
- contextWindow: 200000,
704
- maxTokens: 100000,
890
+ contextWindow: 400000,
891
+ maxTokens: 128000,
705
892
  },
706
893
  "gpt-5-chat-latest": {
707
894
  id: "gpt-5-chat-latest",
@@ -739,6 +926,40 @@ export const MODELS = {
739
926
  contextWindow: 131072,
740
927
  maxTokens: 8192,
741
928
  },
929
+ "mistral-saba-24b": {
930
+ id: "mistral-saba-24b",
931
+ name: "Mistral Saba 24B",
932
+ api: "openai-completions",
933
+ provider: "groq",
934
+ baseUrl: "https://api.groq.com/openai/v1",
935
+ reasoning: false,
936
+ input: ["text"],
937
+ cost: {
938
+ input: 0.79,
939
+ output: 0.79,
940
+ cacheRead: 0,
941
+ cacheWrite: 0,
942
+ },
943
+ contextWindow: 32768,
944
+ maxTokens: 32768,
945
+ },
946
+ "llama3-8b-8192": {
947
+ id: "llama3-8b-8192",
948
+ name: "Llama 3 8B",
949
+ api: "openai-completions",
950
+ provider: "groq",
951
+ baseUrl: "https://api.groq.com/openai/v1",
952
+ reasoning: false,
953
+ input: ["text"],
954
+ cost: {
955
+ input: 0.05,
956
+ output: 0.08,
957
+ cacheRead: 0,
958
+ cacheWrite: 0,
959
+ },
960
+ contextWindow: 8192,
961
+ maxTokens: 8192,
962
+ },
742
963
  "qwen-qwq-32b": {
743
964
  id: "qwen-qwq-32b",
744
965
  name: "Qwen QwQ 32B",
@@ -790,23 +1011,6 @@ export const MODELS = {
790
1011
  contextWindow: 131072,
791
1012
  maxTokens: 8192,
792
1013
  },
793
- "llama3-8b-8192": {
794
- id: "llama3-8b-8192",
795
- name: "Llama 3 8B",
796
- api: "openai-completions",
797
- provider: "groq",
798
- baseUrl: "https://api.groq.com/openai/v1",
799
- reasoning: false,
800
- input: ["text"],
801
- cost: {
802
- input: 0.05,
803
- output: 0.08,
804
- cacheRead: 0,
805
- cacheWrite: 0,
806
- },
807
- contextWindow: 8192,
808
- maxTokens: 8192,
809
- },
810
1014
  "gemma2-9b-it": {
811
1015
  id: "gemma2-9b-it",
812
1016
  name: "Gemma 2 9B",
@@ -841,90 +1045,73 @@ export const MODELS = {
841
1045
  contextWindow: 131072,
842
1046
  maxTokens: 32768,
843
1047
  },
844
- "mistral-saba-24b": {
845
- id: "mistral-saba-24b",
846
- name: "Mistral Saba 24B",
1048
+ "moonshotai/kimi-k2-instruct-0905": {
1049
+ id: "moonshotai/kimi-k2-instruct-0905",
1050
+ name: "Kimi K2 Instruct 0905",
847
1051
  api: "openai-completions",
848
1052
  provider: "groq",
849
1053
  baseUrl: "https://api.groq.com/openai/v1",
850
1054
  reasoning: false,
851
1055
  input: ["text"],
852
1056
  cost: {
853
- input: 0.79,
854
- output: 0.79,
1057
+ input: 1,
1058
+ output: 3,
855
1059
  cacheRead: 0,
856
1060
  cacheWrite: 0,
857
1061
  },
858
- contextWindow: 32768,
859
- maxTokens: 32768,
1062
+ contextWindow: 262144,
1063
+ maxTokens: 16384,
860
1064
  },
861
- "openai/gpt-oss-20b": {
862
- id: "openai/gpt-oss-20b",
863
- name: "GPT OSS 20B",
1065
+ "moonshotai/kimi-k2-instruct": {
1066
+ id: "moonshotai/kimi-k2-instruct",
1067
+ name: "Kimi K2 Instruct",
864
1068
  api: "openai-completions",
865
1069
  provider: "groq",
866
1070
  baseUrl: "https://api.groq.com/openai/v1",
867
- reasoning: true,
1071
+ reasoning: false,
868
1072
  input: ["text"],
869
1073
  cost: {
870
- input: 0.1,
871
- output: 0.5,
1074
+ input: 1,
1075
+ output: 3,
872
1076
  cacheRead: 0,
873
1077
  cacheWrite: 0,
874
1078
  },
875
1079
  contextWindow: 131072,
876
- maxTokens: 32768,
1080
+ maxTokens: 16384,
877
1081
  },
878
- "openai/gpt-oss-120b": {
879
- id: "openai/gpt-oss-120b",
880
- name: "GPT OSS 120B",
1082
+ "openai/gpt-oss-20b": {
1083
+ id: "openai/gpt-oss-20b",
1084
+ name: "GPT OSS 20B",
881
1085
  api: "openai-completions",
882
1086
  provider: "groq",
883
1087
  baseUrl: "https://api.groq.com/openai/v1",
884
1088
  reasoning: true,
885
1089
  input: ["text"],
886
1090
  cost: {
887
- input: 0.15,
888
- output: 0.75,
1091
+ input: 0.1,
1092
+ output: 0.5,
889
1093
  cacheRead: 0,
890
1094
  cacheWrite: 0,
891
1095
  },
892
1096
  contextWindow: 131072,
893
1097
  maxTokens: 32768,
894
1098
  },
895
- "meta-llama/llama-4-maverick-17b-128e-instruct": {
896
- id: "meta-llama/llama-4-maverick-17b-128e-instruct",
897
- name: "Llama 4 Maverick 17B",
898
- api: "openai-completions",
899
- provider: "groq",
900
- baseUrl: "https://api.groq.com/openai/v1",
901
- reasoning: false,
902
- input: ["text", "image"],
903
- cost: {
904
- input: 0.2,
905
- output: 0.6,
906
- cacheRead: 0,
907
- cacheWrite: 0,
908
- },
909
- contextWindow: 131072,
910
- maxTokens: 8192,
911
- },
912
- "meta-llama/llama-4-scout-17b-16e-instruct": {
913
- id: "meta-llama/llama-4-scout-17b-16e-instruct",
914
- name: "Llama 4 Scout 17B",
1099
+ "openai/gpt-oss-120b": {
1100
+ id: "openai/gpt-oss-120b",
1101
+ name: "GPT OSS 120B",
915
1102
  api: "openai-completions",
916
1103
  provider: "groq",
917
1104
  baseUrl: "https://api.groq.com/openai/v1",
918
- reasoning: false,
919
- input: ["text", "image"],
1105
+ reasoning: true,
1106
+ input: ["text"],
920
1107
  cost: {
921
- input: 0.11,
922
- output: 0.34,
1108
+ input: 0.15,
1109
+ output: 0.75,
923
1110
  cacheRead: 0,
924
1111
  cacheWrite: 0,
925
1112
  },
926
1113
  contextWindow: 131072,
927
- maxTokens: 8192,
1114
+ maxTokens: 32768,
928
1115
  },
929
1116
  "qwen/qwen3-32b": {
930
1117
  id: "qwen/qwen3-32b",
@@ -943,39 +1130,39 @@ export const MODELS = {
943
1130
  contextWindow: 131072,
944
1131
  maxTokens: 16384,
945
1132
  },
946
- "moonshotai/kimi-k2-instruct-0905": {
947
- id: "moonshotai/kimi-k2-instruct-0905",
948
- name: "Kimi K2 Instruct 0905",
1133
+ "meta-llama/llama-4-scout-17b-16e-instruct": {
1134
+ id: "meta-llama/llama-4-scout-17b-16e-instruct",
1135
+ name: "Llama 4 Scout 17B",
949
1136
  api: "openai-completions",
950
1137
  provider: "groq",
951
1138
  baseUrl: "https://api.groq.com/openai/v1",
952
1139
  reasoning: false,
953
- input: ["text"],
1140
+ input: ["text", "image"],
954
1141
  cost: {
955
- input: 1,
956
- output: 3,
1142
+ input: 0.11,
1143
+ output: 0.34,
957
1144
  cacheRead: 0,
958
1145
  cacheWrite: 0,
959
1146
  },
960
- contextWindow: 262144,
961
- maxTokens: 16384,
1147
+ contextWindow: 131072,
1148
+ maxTokens: 8192,
962
1149
  },
963
- "moonshotai/kimi-k2-instruct": {
964
- id: "moonshotai/kimi-k2-instruct",
965
- name: "Kimi K2 Instruct",
1150
+ "meta-llama/llama-4-maverick-17b-128e-instruct": {
1151
+ id: "meta-llama/llama-4-maverick-17b-128e-instruct",
1152
+ name: "Llama 4 Maverick 17B",
966
1153
  api: "openai-completions",
967
1154
  provider: "groq",
968
1155
  baseUrl: "https://api.groq.com/openai/v1",
969
1156
  reasoning: false,
970
- input: ["text"],
1157
+ input: ["text", "image"],
971
1158
  cost: {
972
- input: 1,
973
- output: 3,
1159
+ input: 0.2,
1160
+ output: 0.6,
974
1161
  cacheRead: 0,
975
1162
  cacheWrite: 0,
976
1163
  },
977
1164
  contextWindow: 131072,
978
- maxTokens: 16384,
1165
+ maxTokens: 8192,
979
1166
  },
980
1167
  },
981
1168
  cerebras: {
@@ -996,6 +1183,23 @@ export const MODELS = {
996
1183
  contextWindow: 131000,
997
1184
  maxTokens: 32000,
998
1185
  },
1186
+ "qwen-3-coder-480b": {
1187
+ id: "qwen-3-coder-480b",
1188
+ name: "Qwen 3 Coder 480B",
1189
+ api: "openai-completions",
1190
+ provider: "cerebras",
1191
+ baseUrl: "https://api.cerebras.ai/v1",
1192
+ reasoning: false,
1193
+ input: ["text"],
1194
+ cost: {
1195
+ input: 2,
1196
+ output: 2,
1197
+ cacheRead: 0,
1198
+ cacheWrite: 0,
1199
+ },
1200
+ contextWindow: 131000,
1201
+ maxTokens: 32000,
1202
+ },
999
1203
  "gpt-oss-120b": {
1000
1204
  id: "gpt-oss-120b",
1001
1205
  name: "GPT OSS 120B",
@@ -1013,122 +1217,122 @@ export const MODELS = {
1013
1217
  contextWindow: 131072,
1014
1218
  maxTokens: 32768,
1015
1219
  },
1016
- "qwen-3-coder-480b": {
1017
- id: "qwen-3-coder-480b",
1018
- name: "Qwen 3 Coder 480B",
1220
+ },
1221
+ xai: {
1222
+ "grok-4-fast-non-reasoning": {
1223
+ id: "grok-4-fast-non-reasoning",
1224
+ name: "Grok 4 Fast (Non-Reasoning)",
1019
1225
  api: "openai-completions",
1020
- provider: "cerebras",
1021
- baseUrl: "https://api.cerebras.ai/v1",
1226
+ provider: "xai",
1227
+ baseUrl: "https://api.x.ai/v1",
1022
1228
  reasoning: false,
1023
- input: ["text"],
1229
+ input: ["text", "image"],
1024
1230
  cost: {
1025
- input: 2,
1026
- output: 2,
1027
- cacheRead: 0,
1231
+ input: 0.2,
1232
+ output: 0.5,
1233
+ cacheRead: 0.05,
1028
1234
  cacheWrite: 0,
1029
1235
  },
1030
- contextWindow: 131000,
1031
- maxTokens: 32000,
1236
+ contextWindow: 2000000,
1237
+ maxTokens: 30000,
1032
1238
  },
1033
- },
1034
- xai: {
1035
- "grok-3-mini-fast-latest": {
1036
- id: "grok-3-mini-fast-latest",
1037
- name: "Grok 3 Mini Fast Latest",
1239
+ "grok-3-fast": {
1240
+ id: "grok-3-fast",
1241
+ name: "Grok 3 Fast",
1038
1242
  api: "openai-completions",
1039
1243
  provider: "xai",
1040
1244
  baseUrl: "https://api.x.ai/v1",
1041
- reasoning: true,
1245
+ reasoning: false,
1042
1246
  input: ["text"],
1043
1247
  cost: {
1044
- input: 0.6,
1045
- output: 4,
1046
- cacheRead: 0.15,
1248
+ input: 5,
1249
+ output: 25,
1250
+ cacheRead: 1.25,
1047
1251
  cacheWrite: 0,
1048
1252
  },
1049
1253
  contextWindow: 131072,
1050
1254
  maxTokens: 8192,
1051
1255
  },
1052
- "grok-3-mini-latest": {
1053
- id: "grok-3-mini-latest",
1054
- name: "Grok 3 Mini Latest",
1256
+ "grok-4": {
1257
+ id: "grok-4",
1258
+ name: "Grok 4",
1055
1259
  api: "openai-completions",
1056
1260
  provider: "xai",
1057
1261
  baseUrl: "https://api.x.ai/v1",
1058
1262
  reasoning: true,
1059
1263
  input: ["text"],
1060
1264
  cost: {
1061
- input: 0.3,
1062
- output: 0.5,
1063
- cacheRead: 0.075,
1265
+ input: 3,
1266
+ output: 15,
1267
+ cacheRead: 0.75,
1064
1268
  cacheWrite: 0,
1065
1269
  },
1066
- contextWindow: 131072,
1067
- maxTokens: 8192,
1270
+ contextWindow: 256000,
1271
+ maxTokens: 64000,
1068
1272
  },
1069
- "grok-beta": {
1070
- id: "grok-beta",
1071
- name: "Grok Beta",
1273
+ "grok-2-vision": {
1274
+ id: "grok-2-vision",
1275
+ name: "Grok 2 Vision",
1072
1276
  api: "openai-completions",
1073
1277
  provider: "xai",
1074
1278
  baseUrl: "https://api.x.ai/v1",
1075
1279
  reasoning: false,
1076
- input: ["text"],
1280
+ input: ["text", "image"],
1077
1281
  cost: {
1078
- input: 5,
1079
- output: 15,
1080
- cacheRead: 5,
1282
+ input: 2,
1283
+ output: 10,
1284
+ cacheRead: 2,
1081
1285
  cacheWrite: 0,
1082
1286
  },
1083
- contextWindow: 131072,
1287
+ contextWindow: 8192,
1084
1288
  maxTokens: 4096,
1085
1289
  },
1086
- "grok-3-fast-latest": {
1087
- id: "grok-3-fast-latest",
1088
- name: "Grok 3 Fast Latest",
1290
+ "grok-code-fast-1": {
1291
+ id: "grok-code-fast-1",
1292
+ name: "Grok Code Fast 1",
1089
1293
  api: "openai-completions",
1090
1294
  provider: "xai",
1091
1295
  baseUrl: "https://api.x.ai/v1",
1092
- reasoning: false,
1296
+ reasoning: true,
1093
1297
  input: ["text"],
1094
1298
  cost: {
1095
- input: 5,
1096
- output: 25,
1097
- cacheRead: 1.25,
1299
+ input: 0.2,
1300
+ output: 1.5,
1301
+ cacheRead: 0.02,
1098
1302
  cacheWrite: 0,
1099
1303
  },
1100
- contextWindow: 131072,
1101
- maxTokens: 8192,
1304
+ contextWindow: 256000,
1305
+ maxTokens: 10000,
1102
1306
  },
1103
- "grok-3": {
1104
- id: "grok-3",
1105
- name: "Grok 3",
1307
+ "grok-2": {
1308
+ id: "grok-2",
1309
+ name: "Grok 2",
1106
1310
  api: "openai-completions",
1107
1311
  provider: "xai",
1108
1312
  baseUrl: "https://api.x.ai/v1",
1109
1313
  reasoning: false,
1110
1314
  input: ["text"],
1111
1315
  cost: {
1112
- input: 3,
1113
- output: 15,
1114
- cacheRead: 0.75,
1316
+ input: 2,
1317
+ output: 10,
1318
+ cacheRead: 2,
1115
1319
  cacheWrite: 0,
1116
1320
  },
1117
1321
  contextWindow: 131072,
1118
1322
  maxTokens: 8192,
1119
1323
  },
1120
- "grok-3-mini": {
1121
- id: "grok-3-mini",
1122
- name: "Grok 3 Mini",
1324
+ "grok-3-mini-fast-latest": {
1325
+ id: "grok-3-mini-fast-latest",
1326
+ name: "Grok 3 Mini Fast Latest",
1123
1327
  api: "openai-completions",
1124
1328
  provider: "xai",
1125
1329
  baseUrl: "https://api.x.ai/v1",
1126
1330
  reasoning: true,
1127
1331
  input: ["text"],
1128
1332
  cost: {
1129
- input: 0.3,
1130
- output: 0.5,
1131
- cacheRead: 0.075,
1333
+ input: 0.6,
1334
+ output: 4,
1335
+ cacheRead: 0.15,
1132
1336
  cacheWrite: 0,
1133
1337
  },
1134
1338
  contextWindow: 131072,
@@ -1151,9 +1355,43 @@ export const MODELS = {
1151
1355
  contextWindow: 8192,
1152
1356
  maxTokens: 4096,
1153
1357
  },
1154
- "grok-2": {
1155
- id: "grok-2",
1156
- name: "Grok 2",
1358
+ "grok-3": {
1359
+ id: "grok-3",
1360
+ name: "Grok 3",
1361
+ api: "openai-completions",
1362
+ provider: "xai",
1363
+ baseUrl: "https://api.x.ai/v1",
1364
+ reasoning: false,
1365
+ input: ["text"],
1366
+ cost: {
1367
+ input: 3,
1368
+ output: 15,
1369
+ cacheRead: 0.75,
1370
+ cacheWrite: 0,
1371
+ },
1372
+ contextWindow: 131072,
1373
+ maxTokens: 8192,
1374
+ },
1375
+ "grok-4-fast": {
1376
+ id: "grok-4-fast",
1377
+ name: "Grok 4 Fast",
1378
+ api: "openai-completions",
1379
+ provider: "xai",
1380
+ baseUrl: "https://api.x.ai/v1",
1381
+ reasoning: true,
1382
+ input: ["text", "image"],
1383
+ cost: {
1384
+ input: 0.2,
1385
+ output: 0.5,
1386
+ cacheRead: 0.05,
1387
+ cacheWrite: 0,
1388
+ },
1389
+ contextWindow: 2000000,
1390
+ maxTokens: 30000,
1391
+ },
1392
+ "grok-2-latest": {
1393
+ id: "grok-2-latest",
1394
+ name: "Grok 2 Latest",
1157
1395
  api: "openai-completions",
1158
1396
  provider: "xai",
1159
1397
  baseUrl: "https://api.x.ai/v1",
@@ -1168,22 +1406,39 @@ export const MODELS = {
1168
1406
  contextWindow: 131072,
1169
1407
  maxTokens: 8192,
1170
1408
  },
1171
- "grok-2-vision-latest": {
1172
- id: "grok-2-vision-latest",
1173
- name: "Grok 2 Vision Latest",
1409
+ "grok-2-1212": {
1410
+ id: "grok-2-1212",
1411
+ name: "Grok 2 (1212)",
1174
1412
  api: "openai-completions",
1175
1413
  provider: "xai",
1176
1414
  baseUrl: "https://api.x.ai/v1",
1177
1415
  reasoning: false,
1178
- input: ["text", "image"],
1416
+ input: ["text"],
1179
1417
  cost: {
1180
1418
  input: 2,
1181
1419
  output: 10,
1182
1420
  cacheRead: 2,
1183
1421
  cacheWrite: 0,
1184
1422
  },
1185
- contextWindow: 8192,
1186
- maxTokens: 4096,
1423
+ contextWindow: 131072,
1424
+ maxTokens: 8192,
1425
+ },
1426
+ "grok-3-fast-latest": {
1427
+ id: "grok-3-fast-latest",
1428
+ name: "Grok 3 Fast Latest",
1429
+ api: "openai-completions",
1430
+ provider: "xai",
1431
+ baseUrl: "https://api.x.ai/v1",
1432
+ reasoning: false,
1433
+ input: ["text"],
1434
+ cost: {
1435
+ input: 5,
1436
+ output: 25,
1437
+ cacheRead: 1.25,
1438
+ cacheWrite: 0,
1439
+ },
1440
+ contextWindow: 131072,
1441
+ maxTokens: 8192,
1187
1442
  },
1188
1443
  "grok-3-latest": {
1189
1444
  id: "grok-3-latest",
@@ -1202,9 +1457,9 @@ export const MODELS = {
1202
1457
  contextWindow: 131072,
1203
1458
  maxTokens: 8192,
1204
1459
  },
1205
- "grok-2-vision": {
1206
- id: "grok-2-vision",
1207
- name: "Grok 2 Vision",
1460
+ "grok-2-vision-latest": {
1461
+ id: "grok-2-vision-latest",
1462
+ name: "Grok 2 Vision Latest",
1208
1463
  api: "openai-completions",
1209
1464
  provider: "xai",
1210
1465
  baseUrl: "https://api.x.ai/v1",
@@ -1218,91 +1473,74 @@ export const MODELS = {
1218
1473
  },
1219
1474
  contextWindow: 8192,
1220
1475
  maxTokens: 4096,
1221
- },
1222
- "grok-2-latest": {
1223
- id: "grok-2-latest",
1224
- name: "Grok 2 Latest",
1225
- api: "openai-completions",
1226
- provider: "xai",
1227
- baseUrl: "https://api.x.ai/v1",
1228
- reasoning: false,
1229
- input: ["text"],
1230
- cost: {
1231
- input: 2,
1232
- output: 10,
1233
- cacheRead: 2,
1234
- cacheWrite: 0,
1235
- },
1236
- contextWindow: 131072,
1237
- maxTokens: 8192,
1238
- },
1239
- "grok-3-fast": {
1240
- id: "grok-3-fast",
1241
- name: "Grok 3 Fast",
1476
+ },
1477
+ "grok-vision-beta": {
1478
+ id: "grok-vision-beta",
1479
+ name: "Grok Vision Beta",
1242
1480
  api: "openai-completions",
1243
1481
  provider: "xai",
1244
1482
  baseUrl: "https://api.x.ai/v1",
1245
1483
  reasoning: false,
1246
- input: ["text"],
1484
+ input: ["text", "image"],
1247
1485
  cost: {
1248
1486
  input: 5,
1249
- output: 25,
1250
- cacheRead: 1.25,
1487
+ output: 15,
1488
+ cacheRead: 5,
1251
1489
  cacheWrite: 0,
1252
1490
  },
1253
- contextWindow: 131072,
1254
- maxTokens: 8192,
1491
+ contextWindow: 8192,
1492
+ maxTokens: 4096,
1255
1493
  },
1256
- "grok-2-1212": {
1257
- id: "grok-2-1212",
1258
- name: "Grok 2 (1212)",
1494
+ "grok-3-mini": {
1495
+ id: "grok-3-mini",
1496
+ name: "Grok 3 Mini",
1259
1497
  api: "openai-completions",
1260
1498
  provider: "xai",
1261
1499
  baseUrl: "https://api.x.ai/v1",
1262
- reasoning: false,
1500
+ reasoning: true,
1263
1501
  input: ["text"],
1264
1502
  cost: {
1265
- input: 2,
1266
- output: 10,
1267
- cacheRead: 2,
1503
+ input: 0.3,
1504
+ output: 0.5,
1505
+ cacheRead: 0.075,
1268
1506
  cacheWrite: 0,
1269
1507
  },
1270
1508
  contextWindow: 131072,
1271
1509
  maxTokens: 8192,
1272
1510
  },
1273
- "grok-4": {
1274
- id: "grok-4",
1275
- name: "Grok 4",
1511
+ "grok-beta": {
1512
+ id: "grok-beta",
1513
+ name: "Grok Beta",
1276
1514
  api: "openai-completions",
1277
1515
  provider: "xai",
1278
1516
  baseUrl: "https://api.x.ai/v1",
1279
- reasoning: true,
1517
+ reasoning: false,
1280
1518
  input: ["text"],
1281
1519
  cost: {
1282
- input: 3,
1520
+ input: 5,
1283
1521
  output: 15,
1284
- cacheRead: 0.75,
1522
+ cacheRead: 5,
1285
1523
  cacheWrite: 0,
1286
1524
  },
1287
- contextWindow: 256000,
1288
- maxTokens: 64000,
1525
+ contextWindow: 131072,
1526
+ maxTokens: 4096,
1289
1527
  },
1290
- "grok-vision-beta": {
1291
- id: "grok-vision-beta",
1292
- name: "Grok Vision Beta",
1528
+ "grok-3-mini-latest": {
1529
+ id: "grok-3-mini-latest",
1530
+ name: "Grok 3 Mini Latest",
1293
1531
  api: "openai-completions",
1294
1532
  provider: "xai",
1295
1533
  baseUrl: "https://api.x.ai/v1",
1296
- reasoning: false,
1297
- input: ["text", "image"],
1534
+ reasoning: true,
1535
+ input: ["text"],
1298
1536
  cost: {
1299
- input: 5,
1300
- output: 15,
1301
- cacheRead: 5,
1537
+ input: 0.3,
1538
+ output: 0.5,
1539
+ cacheRead: 0.075,
1302
1540
  cacheWrite: 0,
1303
1541
  },
1304
- contextWindow: 8192,
1305
- maxTokens: 4096,
1542
+ contextWindow: 131072,
1543
+ maxTokens: 8192,
1306
1544
  },
1307
1545
  "grok-3-mini-fast": {
1308
1546
  id: "grok-3-mini-fast",
@@ -1321,25 +1559,42 @@ export const MODELS = {
1321
1559
  contextWindow: 131072,
1322
1560
  maxTokens: 8192,
1323
1561
  },
1324
- "grok-code-fast-1": {
1325
- id: "grok-code-fast-1",
1326
- name: "Grok Code Fast 1",
1327
- api: "openai-completions",
1328
- provider: "xai",
1329
- baseUrl: "https://api.x.ai/v1",
1330
- reasoning: false,
1562
+ },
1563
+ zai: {
1564
+ "glm-4.5-flash": {
1565
+ id: "glm-4.5-flash",
1566
+ name: "GLM-4.5-Flash",
1567
+ api: "anthropic-messages",
1568
+ provider: "zai",
1569
+ baseUrl: "https://api.z.ai/api/anthropic",
1570
+ reasoning: true,
1331
1571
  input: ["text"],
1332
1572
  cost: {
1333
- input: 0.2,
1334
- output: 1.5,
1335
- cacheRead: 0.02,
1573
+ input: 0,
1574
+ output: 0,
1575
+ cacheRead: 0,
1336
1576
  cacheWrite: 0,
1337
1577
  },
1338
- contextWindow: 32768,
1339
- maxTokens: 8192,
1578
+ contextWindow: 131072,
1579
+ maxTokens: 98304,
1580
+ },
1581
+ "glm-4.5": {
1582
+ id: "glm-4.5",
1583
+ name: "GLM-4.5",
1584
+ api: "anthropic-messages",
1585
+ provider: "zai",
1586
+ baseUrl: "https://api.z.ai/api/anthropic",
1587
+ reasoning: true,
1588
+ input: ["text"],
1589
+ cost: {
1590
+ input: 0.6,
1591
+ output: 2.2,
1592
+ cacheRead: 0.11,
1593
+ cacheWrite: 0,
1594
+ },
1595
+ contextWindow: 131072,
1596
+ maxTokens: 98304,
1340
1597
  },
1341
- },
1342
- zai: {
1343
1598
  "glm-4.5-air": {
1344
1599
  id: "glm-4.5-air",
1345
1600
  name: "GLM-4.5-Air",
@@ -1374,75 +1629,109 @@ export const MODELS = {
1374
1629
  contextWindow: 64000,
1375
1630
  maxTokens: 16384,
1376
1631
  },
1377
- "glm-4.5-flash": {
1378
- id: "glm-4.5-flash",
1379
- name: "GLM-4.5-Flash",
1632
+ "glm-4.6": {
1633
+ id: "glm-4.6",
1634
+ name: "GLM-4.6",
1380
1635
  api: "anthropic-messages",
1381
1636
  provider: "zai",
1382
1637
  baseUrl: "https://api.z.ai/api/anthropic",
1383
1638
  reasoning: true,
1384
1639
  input: ["text"],
1385
1640
  cost: {
1386
- input: 0,
1387
- output: 0,
1388
- cacheRead: 0,
1641
+ input: 0.6,
1642
+ output: 2.2,
1643
+ cacheRead: 0.11,
1389
1644
  cacheWrite: 0,
1390
1645
  },
1391
- contextWindow: 131072,
1392
- maxTokens: 98304,
1646
+ contextWindow: 204800,
1647
+ maxTokens: 131072,
1393
1648
  },
1394
- "glm-4.5": {
1395
- id: "glm-4.5",
1396
- name: "GLM-4.5",
1397
- api: "anthropic-messages",
1398
- provider: "zai",
1399
- baseUrl: "https://api.z.ai/api/anthropic",
1649
+ },
1650
+ openrouter: {
1651
+ "z-ai/glm-4.6": {
1652
+ id: "z-ai/glm-4.6",
1653
+ name: "Z.AI: GLM 4.6",
1654
+ api: "openai-completions",
1655
+ provider: "openrouter",
1656
+ baseUrl: "https://openrouter.ai/api/v1",
1400
1657
  reasoning: true,
1401
1658
  input: ["text"],
1402
1659
  cost: {
1403
- input: 0.6,
1404
- output: 2.2,
1405
- cacheRead: 0.11,
1660
+ input: 0.5,
1661
+ output: 1.75,
1662
+ cacheRead: 0,
1406
1663
  cacheWrite: 0,
1407
1664
  },
1408
- contextWindow: 131072,
1409
- maxTokens: 98304,
1665
+ contextWindow: 202752,
1666
+ maxTokens: 202752,
1410
1667
  },
1411
- },
1412
- openrouter: {
1413
- "alibaba/tongyi-deepresearch-30b-a3b": {
1414
- id: "alibaba/tongyi-deepresearch-30b-a3b",
1415
- name: "Tongyi DeepResearch 30B A3B",
1668
+ "deepseek/deepseek-v3.2-exp": {
1669
+ id: "deepseek/deepseek-v3.2-exp",
1670
+ name: "DeepSeek: DeepSeek V3.2 Exp",
1416
1671
  api: "openai-completions",
1417
1672
  provider: "openrouter",
1418
1673
  baseUrl: "https://openrouter.ai/api/v1",
1419
1674
  reasoning: true,
1420
1675
  input: ["text"],
1421
1676
  cost: {
1422
- input: 0.09,
1423
- output: 0.44999999999999996,
1677
+ input: 0.27,
1678
+ output: 0.39999999999999997,
1679
+ cacheRead: 0,
1680
+ cacheWrite: 0,
1681
+ },
1682
+ contextWindow: 163840,
1683
+ maxTokens: 4096,
1684
+ },
1685
+ "qwen/qwen3-vl-235b-a22b-thinking": {
1686
+ id: "qwen/qwen3-vl-235b-a22b-thinking",
1687
+ name: "Qwen: Qwen3 VL 235B A22B Thinking",
1688
+ api: "openai-completions",
1689
+ provider: "openrouter",
1690
+ baseUrl: "https://openrouter.ai/api/v1",
1691
+ reasoning: true,
1692
+ input: ["text", "image"],
1693
+ cost: {
1694
+ input: 0.5,
1695
+ output: 3.5,
1696
+ cacheRead: 0,
1697
+ cacheWrite: 0,
1698
+ },
1699
+ contextWindow: 65536,
1700
+ maxTokens: 65536,
1701
+ },
1702
+ "qwen/qwen3-vl-235b-a22b-instruct": {
1703
+ id: "qwen/qwen3-vl-235b-a22b-instruct",
1704
+ name: "Qwen: Qwen3 VL 235B A22B Instruct",
1705
+ api: "openai-completions",
1706
+ provider: "openrouter",
1707
+ baseUrl: "https://openrouter.ai/api/v1",
1708
+ reasoning: true,
1709
+ input: ["text", "image"],
1710
+ cost: {
1711
+ input: 0.3,
1712
+ output: 1.5,
1424
1713
  cacheRead: 0,
1425
1714
  cacheWrite: 0,
1426
1715
  },
1427
1716
  contextWindow: 131072,
1428
- maxTokens: 131072,
1717
+ maxTokens: 32768,
1429
1718
  },
1430
- "qwen/qwen3-coder-flash": {
1431
- id: "qwen/qwen3-coder-flash",
1432
- name: "Qwen: Qwen3 Coder Flash",
1719
+ "qwen/qwen3-max": {
1720
+ id: "qwen/qwen3-max",
1721
+ name: "Qwen: Qwen3 Max",
1433
1722
  api: "openai-completions",
1434
1723
  provider: "openrouter",
1435
1724
  baseUrl: "https://openrouter.ai/api/v1",
1436
1725
  reasoning: false,
1437
1726
  input: ["text"],
1438
1727
  cost: {
1439
- input: 0.3,
1440
- output: 1.5,
1441
- cacheRead: 0.08,
1728
+ input: 1.2,
1729
+ output: 6,
1730
+ cacheRead: 0.24,
1442
1731
  cacheWrite: 0,
1443
1732
  },
1444
- contextWindow: 128000,
1445
- maxTokens: 65536,
1733
+ contextWindow: 256000,
1734
+ maxTokens: 32768,
1446
1735
  },
1447
1736
  "qwen/qwen3-coder-plus": {
1448
1737
  id: "qwen/qwen3-coder-plus",
@@ -1461,6 +1750,74 @@ export const MODELS = {
1461
1750
  contextWindow: 128000,
1462
1751
  maxTokens: 65536,
1463
1752
  },
1753
+ "deepseek/deepseek-v3.1-terminus": {
1754
+ id: "deepseek/deepseek-v3.1-terminus",
1755
+ name: "DeepSeek: DeepSeek V3.1 Terminus",
1756
+ api: "openai-completions",
1757
+ provider: "openrouter",
1758
+ baseUrl: "https://openrouter.ai/api/v1",
1759
+ reasoning: true,
1760
+ input: ["text"],
1761
+ cost: {
1762
+ input: 0.22999999999999998,
1763
+ output: 0.8999999999999999,
1764
+ cacheRead: 0,
1765
+ cacheWrite: 0,
1766
+ },
1767
+ contextWindow: 163840,
1768
+ maxTokens: 163840,
1769
+ },
1770
+ "alibaba/tongyi-deepresearch-30b-a3b:free": {
1771
+ id: "alibaba/tongyi-deepresearch-30b-a3b:free",
1772
+ name: "Tongyi DeepResearch 30B A3B (free)",
1773
+ api: "openai-completions",
1774
+ provider: "openrouter",
1775
+ baseUrl: "https://openrouter.ai/api/v1",
1776
+ reasoning: false,
1777
+ input: ["text"],
1778
+ cost: {
1779
+ input: 0,
1780
+ output: 0,
1781
+ cacheRead: 0,
1782
+ cacheWrite: 0,
1783
+ },
1784
+ contextWindow: 131072,
1785
+ maxTokens: 131072,
1786
+ },
1787
+ "alibaba/tongyi-deepresearch-30b-a3b": {
1788
+ id: "alibaba/tongyi-deepresearch-30b-a3b",
1789
+ name: "Tongyi DeepResearch 30B A3B",
1790
+ api: "openai-completions",
1791
+ provider: "openrouter",
1792
+ baseUrl: "https://openrouter.ai/api/v1",
1793
+ reasoning: true,
1794
+ input: ["text"],
1795
+ cost: {
1796
+ input: 0.09,
1797
+ output: 0.39999999999999997,
1798
+ cacheRead: 0,
1799
+ cacheWrite: 0,
1800
+ },
1801
+ contextWindow: 131072,
1802
+ maxTokens: 131072,
1803
+ },
1804
+ "qwen/qwen3-coder-flash": {
1805
+ id: "qwen/qwen3-coder-flash",
1806
+ name: "Qwen: Qwen3 Coder Flash",
1807
+ api: "openai-completions",
1808
+ provider: "openrouter",
1809
+ baseUrl: "https://openrouter.ai/api/v1",
1810
+ reasoning: false,
1811
+ input: ["text"],
1812
+ cost: {
1813
+ input: 0.3,
1814
+ output: 1.5,
1815
+ cacheRead: 0.08,
1816
+ cacheWrite: 0,
1817
+ },
1818
+ contextWindow: 128000,
1819
+ maxTokens: 65536,
1820
+ },
1464
1821
  "qwen/qwen3-next-80b-a3b-thinking": {
1465
1822
  id: "qwen/qwen3-next-80b-a3b-thinking",
1466
1823
  name: "Qwen: Qwen3 Next 80B A3B Thinking",
@@ -1476,7 +1833,7 @@ export const MODELS = {
1476
1833
  cacheWrite: 0,
1477
1834
  },
1478
1835
  contextWindow: 262144,
1479
- maxTokens: 4096,
1836
+ maxTokens: 262144,
1480
1837
  },
1481
1838
  "qwen/qwen3-next-80b-a3b-instruct": {
1482
1839
  id: "qwen/qwen3-next-80b-a3b-instruct",
@@ -1493,24 +1850,24 @@ export const MODELS = {
1493
1850
  cacheWrite: 0,
1494
1851
  },
1495
1852
  contextWindow: 262144,
1496
- maxTokens: 4096,
1853
+ maxTokens: 262144,
1497
1854
  },
1498
- "meituan/longcat-flash-chat": {
1499
- id: "meituan/longcat-flash-chat",
1500
- name: "Meituan: LongCat Flash Chat",
1855
+ "meituan/longcat-flash-chat:free": {
1856
+ id: "meituan/longcat-flash-chat:free",
1857
+ name: "Meituan: LongCat Flash Chat (free)",
1501
1858
  api: "openai-completions",
1502
1859
  provider: "openrouter",
1503
1860
  baseUrl: "https://openrouter.ai/api/v1",
1504
1861
  reasoning: false,
1505
1862
  input: ["text"],
1506
1863
  cost: {
1507
- input: 0.12,
1508
- output: 0.6,
1864
+ input: 0,
1865
+ output: 0,
1509
1866
  cacheRead: 0,
1510
1867
  cacheWrite: 0,
1511
1868
  },
1512
1869
  contextWindow: 131072,
1513
- maxTokens: 4096,
1870
+ maxTokens: 131072,
1514
1871
  },
1515
1872
  "qwen/qwen-plus-2025-07-28": {
1516
1873
  id: "qwen/qwen-plus-2025-07-28",
@@ -1580,57 +1937,6 @@ export const MODELS = {
1580
1937
  contextWindow: 131072,
1581
1938
  maxTokens: 4096,
1582
1939
  },
1583
- "openrouter/sonoma-dusk-alpha": {
1584
- id: "openrouter/sonoma-dusk-alpha",
1585
- name: "Sonoma Dusk Alpha",
1586
- api: "openai-completions",
1587
- provider: "openrouter",
1588
- baseUrl: "https://openrouter.ai/api/v1",
1589
- reasoning: false,
1590
- input: ["text", "image"],
1591
- cost: {
1592
- input: 0,
1593
- output: 0,
1594
- cacheRead: 0,
1595
- cacheWrite: 0,
1596
- },
1597
- contextWindow: 2000000,
1598
- maxTokens: 4096,
1599
- },
1600
- "openrouter/sonoma-sky-alpha": {
1601
- id: "openrouter/sonoma-sky-alpha",
1602
- name: "Sonoma Sky Alpha",
1603
- api: "openai-completions",
1604
- provider: "openrouter",
1605
- baseUrl: "https://openrouter.ai/api/v1",
1606
- reasoning: true,
1607
- input: ["text", "image"],
1608
- cost: {
1609
- input: 0,
1610
- output: 0,
1611
- cacheRead: 0,
1612
- cacheWrite: 0,
1613
- },
1614
- contextWindow: 2000000,
1615
- maxTokens: 4096,
1616
- },
1617
- "qwen/qwen3-max": {
1618
- id: "qwen/qwen3-max",
1619
- name: "Qwen: Qwen3 Max",
1620
- api: "openai-completions",
1621
- provider: "openrouter",
1622
- baseUrl: "https://openrouter.ai/api/v1",
1623
- reasoning: false,
1624
- input: ["text"],
1625
- cost: {
1626
- input: 1.2,
1627
- output: 6,
1628
- cacheRead: 0.24,
1629
- cacheWrite: 0,
1630
- },
1631
- contextWindow: 256000,
1632
- maxTokens: 32768,
1633
- },
1634
1940
  "moonshotai/kimi-k2-0905": {
1635
1941
  id: "moonshotai/kimi-k2-0905",
1636
1942
  name: "MoonshotAI: Kimi K2 0905",
@@ -1640,13 +1946,13 @@ export const MODELS = {
1640
1946
  reasoning: false,
1641
1947
  input: ["text"],
1642
1948
  cost: {
1643
- input: 0.38,
1644
- output: 1.52,
1949
+ input: 0.39,
1950
+ output: 1.9,
1645
1951
  cacheRead: 0,
1646
1952
  cacheWrite: 0,
1647
1953
  },
1648
1954
  contextWindow: 262144,
1649
- maxTokens: 4096,
1955
+ maxTokens: 262144,
1650
1956
  },
1651
1957
  "deepcogito/cogito-v2-preview-llama-109b-moe": {
1652
1958
  id: "deepcogito/cogito-v2-preview-llama-109b-moe",
@@ -1714,7 +2020,7 @@ export const MODELS = {
1714
2020
  cacheWrite: 0,
1715
2021
  },
1716
2022
  contextWindow: 131072,
1717
- maxTokens: 4096,
2023
+ maxTokens: 131072,
1718
2024
  },
1719
2025
  "nousresearch/hermes-4-405b": {
1720
2026
  id: "nousresearch/hermes-4-405b",
@@ -1725,13 +2031,13 @@ export const MODELS = {
1725
2031
  reasoning: true,
1726
2032
  input: ["text"],
1727
2033
  cost: {
1728
- input: 0.24999987999999998,
1729
- output: 0.999999888,
2034
+ input: 0.3,
2035
+ output: 1.2,
1730
2036
  cacheRead: 0,
1731
2037
  cacheWrite: 0,
1732
2038
  },
1733
2039
  contextWindow: 131072,
1734
- maxTokens: 4096,
2040
+ maxTokens: 131072,
1735
2041
  },
1736
2042
  "deepseek/deepseek-chat-v3.1:free": {
1737
2043
  id: "deepseek/deepseek-chat-v3.1:free",
@@ -1747,7 +2053,7 @@ export const MODELS = {
1747
2053
  cacheRead: 0,
1748
2054
  cacheWrite: 0,
1749
2055
  },
1750
- contextWindow: 163840,
2056
+ contextWindow: 163800,
1751
2057
  maxTokens: 4096,
1752
2058
  },
1753
2059
  "deepseek/deepseek-chat-v3.1": {
@@ -1759,13 +2065,13 @@ export const MODELS = {
1759
2065
  reasoning: true,
1760
2066
  input: ["text"],
1761
2067
  cost: {
1762
- input: 0.24999987999999998,
1763
- output: 0.999999888,
2068
+ input: 0.19999999999999998,
2069
+ output: 0.7999999999999999,
1764
2070
  cacheRead: 0,
1765
2071
  cacheWrite: 0,
1766
2072
  },
1767
2073
  contextWindow: 163840,
1768
- maxTokens: 4096,
2074
+ maxTokens: 163840,
1769
2075
  },
1770
2076
  "mistralai/mistral-medium-3.1": {
1771
2077
  id: "mistralai/mistral-medium-3.1",
@@ -1793,13 +2099,13 @@ export const MODELS = {
1793
2099
  reasoning: true,
1794
2100
  input: ["text", "image"],
1795
2101
  cost: {
1796
- input: 0.5,
2102
+ input: 0.6,
1797
2103
  output: 1.7999999999999998,
1798
2104
  cacheRead: 0,
1799
2105
  cacheWrite: 0,
1800
2106
  },
1801
2107
  contextWindow: 65536,
1802
- maxTokens: 65536,
2108
+ maxTokens: 16384,
1803
2109
  },
1804
2110
  "ai21/jamba-mini-1.7": {
1805
2111
  id: "ai21/jamba-mini-1.7",
@@ -1861,13 +2167,13 @@ export const MODELS = {
1861
2167
  reasoning: false,
1862
2168
  input: ["text"],
1863
2169
  cost: {
1864
- input: 0.07,
1865
- output: 0.28,
2170
+ input: 0.06,
2171
+ output: 0.25,
1866
2172
  cacheRead: 0,
1867
2173
  cacheWrite: 0,
1868
2174
  },
1869
2175
  contextWindow: 262144,
1870
- maxTokens: 4096,
2176
+ maxTokens: 262144,
1871
2177
  },
1872
2178
  "qwen/qwen3-30b-a3b-instruct-2507": {
1873
2179
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -1878,13 +2184,13 @@ export const MODELS = {
1878
2184
  reasoning: false,
1879
2185
  input: ["text"],
1880
2186
  cost: {
1881
- input: 0.07,
1882
- output: 0.28,
2187
+ input: 0.08,
2188
+ output: 0.33,
1883
2189
  cacheRead: 0,
1884
2190
  cacheWrite: 0,
1885
2191
  },
1886
2192
  contextWindow: 262144,
1887
- maxTokens: 4096,
2193
+ maxTokens: 262144,
1888
2194
  },
1889
2195
  "z-ai/glm-4.5": {
1890
2196
  id: "z-ai/glm-4.5",
@@ -1895,13 +2201,13 @@ export const MODELS = {
1895
2201
  reasoning: true,
1896
2202
  input: ["text"],
1897
2203
  cost: {
1898
- input: 0.41,
1899
- output: 1.6500000000000001,
2204
+ input: 0.35,
2205
+ output: 1.55,
1900
2206
  cacheRead: 0,
1901
2207
  cacheWrite: 0,
1902
2208
  },
1903
2209
  contextWindow: 131072,
1904
- maxTokens: 4096,
2210
+ maxTokens: 131072,
1905
2211
  },
1906
2212
  "z-ai/glm-4.5-air:free": {
1907
2213
  id: "z-ai/glm-4.5-air:free",
@@ -1918,7 +2224,7 @@ export const MODELS = {
1918
2224
  cacheWrite: 0,
1919
2225
  },
1920
2226
  contextWindow: 131072,
1921
- maxTokens: 4096,
2227
+ maxTokens: 131072,
1922
2228
  },
1923
2229
  "z-ai/glm-4.5-air": {
1924
2230
  id: "z-ai/glm-4.5-air",
@@ -1946,13 +2252,13 @@ export const MODELS = {
1946
2252
  reasoning: true,
1947
2253
  input: ["text"],
1948
2254
  cost: {
1949
- input: 0.09999999999999999,
1950
- output: 0.39,
2255
+ input: 0.11,
2256
+ output: 0.6,
1951
2257
  cacheRead: 0,
1952
2258
  cacheWrite: 0,
1953
2259
  },
1954
2260
  contextWindow: 262144,
1955
- maxTokens: 4096,
2261
+ maxTokens: 262144,
1956
2262
  },
1957
2263
  "z-ai/glm-4-32b": {
1958
2264
  id: "z-ai/glm-4-32b",
@@ -2003,7 +2309,7 @@ export const MODELS = {
2003
2309
  cacheWrite: 0,
2004
2310
  },
2005
2311
  contextWindow: 262144,
2006
- maxTokens: 4096,
2312
+ maxTokens: 262144,
2007
2313
  },
2008
2314
  "qwen/qwen3-235b-a22b-2507": {
2009
2315
  id: "qwen/qwen3-235b-a22b-2507",
@@ -2014,31 +2320,14 @@ export const MODELS = {
2014
2320
  reasoning: false,
2015
2321
  input: ["text"],
2016
2322
  cost: {
2017
- input: 0.09999999999999999,
2018
- output: 0.09999999999999999,
2323
+ input: 0.08,
2324
+ output: 0.55,
2019
2325
  cacheRead: 0,
2020
2326
  cacheWrite: 0,
2021
2327
  },
2022
2328
  contextWindow: 262144,
2023
2329
  maxTokens: 262144,
2024
2330
  },
2025
- "moonshotai/kimi-k2:free": {
2026
- id: "moonshotai/kimi-k2:free",
2027
- name: "MoonshotAI: Kimi K2 0711 (free)",
2028
- api: "openai-completions",
2029
- provider: "openrouter",
2030
- baseUrl: "https://openrouter.ai/api/v1",
2031
- reasoning: false,
2032
- input: ["text"],
2033
- cost: {
2034
- input: 0,
2035
- output: 0,
2036
- cacheRead: 0,
2037
- cacheWrite: 0,
2038
- },
2039
- contextWindow: 32768,
2040
- maxTokens: 4096,
2041
- },
2042
2331
  "moonshotai/kimi-k2": {
2043
2332
  id: "moonshotai/kimi-k2",
2044
2333
  name: "MoonshotAI: Kimi K2 0711",
@@ -2133,13 +2422,13 @@ export const MODELS = {
2133
2422
  reasoning: false,
2134
2423
  input: ["text", "image"],
2135
2424
  cost: {
2136
- input: 0.075,
2137
- output: 0.19999999999999998,
2425
+ input: 0.06,
2426
+ output: 0.18,
2138
2427
  cacheRead: 0,
2139
2428
  cacheWrite: 0,
2140
2429
  },
2141
- contextWindow: 128000,
2142
- maxTokens: 4096,
2430
+ contextWindow: 131072,
2431
+ maxTokens: 131072,
2143
2432
  },
2144
2433
  "minimax/minimax-m1": {
2145
2434
  id: "minimax/minimax-m1",
@@ -2224,7 +2513,7 @@ export const MODELS = {
2224
2513
  cacheWrite: 0,
2225
2514
  },
2226
2515
  contextWindow: 163840,
2227
- maxTokens: 4096,
2516
+ maxTokens: 163840,
2228
2517
  },
2229
2518
  "mistralai/devstral-small-2505:free": {
2230
2519
  id: "mistralai/devstral-small-2505:free",
@@ -2252,13 +2541,13 @@ export const MODELS = {
2252
2541
  reasoning: false,
2253
2542
  input: ["text"],
2254
2543
  cost: {
2255
- input: 0.04,
2256
- output: 0.14,
2544
+ input: 0.049999999999999996,
2545
+ output: 0.22,
2257
2546
  cacheRead: 0,
2258
2547
  cacheWrite: 0,
2259
2548
  },
2260
2549
  contextWindow: 131072,
2261
- maxTokens: 4096,
2550
+ maxTokens: 131072,
2262
2551
  },
2263
2552
  "meta-llama/llama-3.3-8b-instruct:free": {
2264
2553
  id: "meta-llama/llama-3.3-8b-instruct:free",
@@ -2360,7 +2649,7 @@ export const MODELS = {
2360
2649
  cacheWrite: 0,
2361
2650
  },
2362
2651
  contextWindow: 40960,
2363
- maxTokens: 4096,
2652
+ maxTokens: 40960,
2364
2653
  },
2365
2654
  "qwen/qwen3-14b": {
2366
2655
  id: "qwen/qwen3-14b",
@@ -2371,8 +2660,8 @@ export const MODELS = {
2371
2660
  reasoning: true,
2372
2661
  input: ["text"],
2373
2662
  cost: {
2374
- input: 0.06,
2375
- output: 0.24,
2663
+ input: 0.049999999999999996,
2664
+ output: 0.22,
2376
2665
  cacheRead: 0,
2377
2666
  cacheWrite: 0,
2378
2667
  },
@@ -2388,13 +2677,13 @@ export const MODELS = {
2388
2677
  reasoning: true,
2389
2678
  input: ["text"],
2390
2679
  cost: {
2391
- input: 0.03,
2392
- output: 0.13,
2680
+ input: 0.049999999999999996,
2681
+ output: 0.19999999999999998,
2393
2682
  cacheRead: 0,
2394
2683
  cacheWrite: 0,
2395
2684
  },
2396
2685
  contextWindow: 40960,
2397
- maxTokens: 4096,
2686
+ maxTokens: 40960,
2398
2687
  },
2399
2688
  "qwen/qwen3-235b-a22b:free": {
2400
2689
  id: "qwen/qwen3-235b-a22b:free",
@@ -2495,8 +2784,8 @@ export const MODELS = {
2495
2784
  cacheRead: 0,
2496
2785
  cacheWrite: 0,
2497
2786
  },
2498
- contextWindow: 1048576,
2499
- maxTokens: 1048576,
2787
+ contextWindow: 327680,
2788
+ maxTokens: 16384,
2500
2789
  },
2501
2790
  "deepseek/deepseek-chat-v3-0324:free": {
2502
2791
  id: "deepseek/deepseek-chat-v3-0324:free",
@@ -2524,13 +2813,13 @@ export const MODELS = {
2524
2813
  reasoning: false,
2525
2814
  input: ["text"],
2526
2815
  cost: {
2527
- input: 0.24999987999999998,
2528
- output: 0.999999888,
2816
+ input: 0.24,
2817
+ output: 0.84,
2529
2818
  cacheRead: 0,
2530
2819
  cacheWrite: 0,
2531
2820
  },
2532
2821
  contextWindow: 163840,
2533
- maxTokens: 4096,
2822
+ maxTokens: 163840,
2534
2823
  },
2535
2824
  "mistralai/mistral-small-3.1-24b-instruct:free": {
2536
2825
  id: "mistralai/mistral-small-3.1-24b-instruct:free",
@@ -2558,13 +2847,13 @@ export const MODELS = {
2558
2847
  reasoning: false,
2559
2848
  input: ["text", "image"],
2560
2849
  cost: {
2561
- input: 0.04,
2562
- output: 0.15,
2850
+ input: 0.049999999999999996,
2851
+ output: 0.09999999999999999,
2563
2852
  cacheRead: 0,
2564
2853
  cacheWrite: 0,
2565
2854
  },
2566
- contextWindow: 131072,
2567
- maxTokens: 96000,
2855
+ contextWindow: 128000,
2856
+ maxTokens: 4096,
2568
2857
  },
2569
2858
  "microsoft/phi-4-multimodal-instruct": {
2570
2859
  id: "microsoft/phi-4-multimodal-instruct",
@@ -2677,13 +2966,13 @@ export const MODELS = {
2677
2966
  reasoning: false,
2678
2967
  input: ["text"],
2679
2968
  cost: {
2680
- input: 0.04,
2681
- output: 0.15,
2969
+ input: 0.049999999999999996,
2970
+ output: 0.08,
2682
2971
  cacheRead: 0,
2683
2972
  cacheWrite: 0,
2684
2973
  },
2685
2974
  contextWindow: 32768,
2686
- maxTokens: 4096,
2975
+ maxTokens: 16384,
2687
2976
  },
2688
2977
  "deepseek/deepseek-r1-distill-llama-70b": {
2689
2978
  id: "deepseek/deepseek-r1-distill-llama-70b",
@@ -2700,7 +2989,7 @@ export const MODELS = {
2700
2989
  cacheWrite: 0,
2701
2990
  },
2702
2991
  contextWindow: 131072,
2703
- maxTokens: 4096,
2992
+ maxTokens: 131072,
2704
2993
  },
2705
2994
  "deepseek/deepseek-r1": {
2706
2995
  id: "deepseek/deepseek-r1",
@@ -2745,13 +3034,13 @@ export const MODELS = {
2745
3034
  reasoning: false,
2746
3035
  input: ["text"],
2747
3036
  cost: {
2748
- input: 0.24999987999999998,
2749
- output: 0.999999888,
3037
+ input: 0.3,
3038
+ output: 0.85,
2750
3039
  cacheRead: 0,
2751
3040
  cacheWrite: 0,
2752
3041
  },
2753
3042
  contextWindow: 163840,
2754
- maxTokens: 4096,
3043
+ maxTokens: 163840,
2755
3044
  },
2756
3045
  "meta-llama/llama-3.3-70b-instruct:free": {
2757
3046
  id: "meta-llama/llama-3.3-70b-instruct:free",
@@ -2779,13 +3068,13 @@ export const MODELS = {
2779
3068
  reasoning: false,
2780
3069
  input: ["text"],
2781
3070
  cost: {
2782
- input: 0.012,
2783
- output: 0.036,
3071
+ input: 0.04,
3072
+ output: 0.12,
2784
3073
  cacheRead: 0,
2785
3074
  cacheWrite: 0,
2786
3075
  },
2787
3076
  contextWindow: 131072,
2788
- maxTokens: 8192,
3077
+ maxTokens: 131072,
2789
3078
  },
2790
3079
  "amazon/nova-lite-v1": {
2791
3080
  id: "amazon/nova-lite-v1",
@@ -2989,7 +3278,7 @@ export const MODELS = {
2989
3278
  cacheWrite: 0,
2990
3279
  },
2991
3280
  contextWindow: 32768,
2992
- maxTokens: 4096,
3281
+ maxTokens: 32768,
2993
3282
  },
2994
3283
  "mistralai/pixtral-12b": {
2995
3284
  id: "mistralai/pixtral-12b",
@@ -3068,46 +3357,46 @@ export const MODELS = {
3068
3357
  reasoning: false,
3069
3358
  input: ["text"],
3070
3359
  cost: {
3071
- input: 0.12,
3360
+ input: 0.3,
3072
3361
  output: 0.3,
3073
3362
  cacheRead: 0,
3074
3363
  cacheWrite: 0,
3075
3364
  },
3076
- contextWindow: 131072,
3077
- maxTokens: 131072,
3365
+ contextWindow: 65000,
3366
+ maxTokens: 4096,
3078
3367
  },
3079
- "meta-llama/llama-3.1-8b-instruct": {
3080
- id: "meta-llama/llama-3.1-8b-instruct",
3081
- name: "Meta: Llama 3.1 8B Instruct",
3368
+ "meta-llama/llama-3.1-405b-instruct": {
3369
+ id: "meta-llama/llama-3.1-405b-instruct",
3370
+ name: "Meta: Llama 3.1 405B Instruct",
3082
3371
  api: "openai-completions",
3083
3372
  provider: "openrouter",
3084
3373
  baseUrl: "https://openrouter.ai/api/v1",
3085
3374
  reasoning: false,
3086
3375
  input: ["text"],
3087
3376
  cost: {
3088
- input: 0.02,
3089
- output: 0.03,
3377
+ input: 0.7999999999999999,
3378
+ output: 0.7999999999999999,
3090
3379
  cacheRead: 0,
3091
3380
  cacheWrite: 0,
3092
3381
  },
3093
- contextWindow: 16384,
3382
+ contextWindow: 32768,
3094
3383
  maxTokens: 16384,
3095
3384
  },
3096
- "meta-llama/llama-3.1-405b-instruct": {
3097
- id: "meta-llama/llama-3.1-405b-instruct",
3098
- name: "Meta: Llama 3.1 405B Instruct",
3385
+ "meta-llama/llama-3.1-8b-instruct": {
3386
+ id: "meta-llama/llama-3.1-8b-instruct",
3387
+ name: "Meta: Llama 3.1 8B Instruct",
3099
3388
  api: "openai-completions",
3100
3389
  provider: "openrouter",
3101
3390
  baseUrl: "https://openrouter.ai/api/v1",
3102
3391
  reasoning: false,
3103
3392
  input: ["text"],
3104
3393
  cost: {
3105
- input: 0.7999999999999999,
3106
- output: 0.7999999999999999,
3394
+ input: 0.02,
3395
+ output: 0.03,
3107
3396
  cacheRead: 0,
3108
3397
  cacheWrite: 0,
3109
3398
  },
3110
- contextWindow: 32768,
3399
+ contextWindow: 16384,
3111
3400
  maxTokens: 16384,
3112
3401
  },
3113
3402
  "meta-llama/llama-3.1-70b-instruct": {
@@ -3119,13 +3408,13 @@ export const MODELS = {
3119
3408
  reasoning: false,
3120
3409
  input: ["text"],
3121
3410
  cost: {
3122
- input: 0.09999999999999999,
3123
- output: 0.28,
3411
+ input: 0.39999999999999997,
3412
+ output: 0.39999999999999997,
3124
3413
  cacheRead: 0,
3125
3414
  cacheWrite: 0,
3126
3415
  },
3127
3416
  contextWindow: 131072,
3128
- maxTokens: 16384,
3417
+ maxTokens: 4096,
3129
3418
  },
3130
3419
  "mistralai/mistral-nemo": {
3131
3420
  id: "mistralai/mistral-nemo",
@@ -3280,74 +3569,6 @@ export const MODELS = {
3280
3569
  contextWindow: 65536,
3281
3570
  maxTokens: 4096,
3282
3571
  },
3283
- "cohere/command-r-plus": {
3284
- id: "cohere/command-r-plus",
3285
- name: "Cohere: Command R+",
3286
- api: "openai-completions",
3287
- provider: "openrouter",
3288
- baseUrl: "https://openrouter.ai/api/v1",
3289
- reasoning: false,
3290
- input: ["text"],
3291
- cost: {
3292
- input: 3,
3293
- output: 15,
3294
- cacheRead: 0,
3295
- cacheWrite: 0,
3296
- },
3297
- contextWindow: 128000,
3298
- maxTokens: 4000,
3299
- },
3300
- "cohere/command-r-plus-04-2024": {
3301
- id: "cohere/command-r-plus-04-2024",
3302
- name: "Cohere: Command R+ (04-2024)",
3303
- api: "openai-completions",
3304
- provider: "openrouter",
3305
- baseUrl: "https://openrouter.ai/api/v1",
3306
- reasoning: false,
3307
- input: ["text"],
3308
- cost: {
3309
- input: 3,
3310
- output: 15,
3311
- cacheRead: 0,
3312
- cacheWrite: 0,
3313
- },
3314
- contextWindow: 128000,
3315
- maxTokens: 4000,
3316
- },
3317
- "cohere/command-r": {
3318
- id: "cohere/command-r",
3319
- name: "Cohere: Command R",
3320
- api: "openai-completions",
3321
- provider: "openrouter",
3322
- baseUrl: "https://openrouter.ai/api/v1",
3323
- reasoning: false,
3324
- input: ["text"],
3325
- cost: {
3326
- input: 0.5,
3327
- output: 1.5,
3328
- cacheRead: 0,
3329
- cacheWrite: 0,
3330
- },
3331
- contextWindow: 128000,
3332
- maxTokens: 4000,
3333
- },
3334
- "cohere/command-r-03-2024": {
3335
- id: "cohere/command-r-03-2024",
3336
- name: "Cohere: Command R (03-2024)",
3337
- api: "openai-completions",
3338
- provider: "openrouter",
3339
- baseUrl: "https://openrouter.ai/api/v1",
3340
- reasoning: false,
3341
- input: ["text"],
3342
- cost: {
3343
- input: 0.5,
3344
- output: 1.5,
3345
- cacheRead: 0,
3346
- cacheWrite: 0,
3347
- },
3348
- contextWindow: 128000,
3349
- maxTokens: 4000,
3350
- },
3351
3572
  "mistralai/mistral-large": {
3352
3573
  id: "mistralai/mistral-large",
3353
3574
  name: "Mistral Large",
@@ -3408,8 +3629,8 @@ export const MODELS = {
3408
3629
  reasoning: false,
3409
3630
  input: ["text"],
3410
3631
  cost: {
3411
- input: 0.39999999999999997,
3412
- output: 0.39999999999999997,
3632
+ input: 0.54,
3633
+ output: 0.54,
3413
3634
  cacheRead: 0,
3414
3635
  cacheWrite: 0,
3415
3636
  },