@mariozechner/pi-ai 0.5.43 → 0.5.45

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 (39) hide show
  1. package/dist/agent/agent-loop.d.ts +2 -2
  2. package/dist/agent/agent-loop.d.ts.map +1 -1
  3. package/dist/agent/agent-loop.js +18 -2
  4. package/dist/agent/agent-loop.js.map +1 -1
  5. package/dist/agent/index.d.ts +1 -1
  6. package/dist/agent/index.d.ts.map +1 -1
  7. package/dist/agent/index.js.map +1 -1
  8. package/dist/agent/types.d.ts +6 -1
  9. package/dist/agent/types.d.ts.map +1 -1
  10. package/dist/agent/types.js.map +1 -1
  11. package/dist/models.generated.d.ts +933 -270
  12. package/dist/models.generated.d.ts.map +1 -1
  13. package/dist/models.generated.js +1500 -837
  14. package/dist/models.generated.js.map +1 -1
  15. package/dist/providers/anthropic.d.ts.map +1 -1
  16. package/dist/providers/anthropic.js +32 -8
  17. package/dist/providers/anthropic.js.map +1 -1
  18. package/dist/providers/google.d.ts.map +1 -1
  19. package/dist/providers/google.js +8 -7
  20. package/dist/providers/google.js.map +1 -1
  21. package/dist/providers/openai-completions.d.ts.map +1 -1
  22. package/dist/providers/openai-completions.js +6 -5
  23. package/dist/providers/openai-completions.js.map +1 -1
  24. package/dist/providers/openai-responses.d.ts.map +1 -1
  25. package/dist/providers/openai-responses.js +6 -5
  26. package/dist/providers/openai-responses.js.map +1 -1
  27. package/dist/providers/transorm-messages.d.ts.map +1 -1
  28. package/dist/providers/transorm-messages.js +46 -1
  29. package/dist/providers/transorm-messages.js.map +1 -1
  30. package/dist/utils/event-stream.js +3 -7
  31. package/dist/utils/event-stream.js.map +1 -1
  32. package/dist/utils/sanitize-unicode.d.ts +22 -0
  33. package/dist/utils/sanitize-unicode.d.ts.map +1 -0
  34. package/dist/utils/sanitize-unicode.js +26 -0
  35. package/dist/utils/sanitize-unicode.js.map +1 -0
  36. package/dist/utils/validation.d.ts.map +1 -1
  37. package/dist/utils/validation.js +25 -3
  38. package/dist/utils/validation.js.map +1 -1
  39. package/package.json +3 -3
package/dist/models.generated.js CHANGED
@@ -2,9 +2,60 @@
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
+ api: "anthropic-messages",
9
+ provider: "anthropic",
10
+ baseUrl: "https://api.anthropic.com",
11
+ reasoning: false,
12
+ input: ["text", "image"],
13
+ cost: {
14
+ input: 3,
15
+ output: 15,
16
+ cacheRead: 0.3,
17
+ cacheWrite: 3.75,
18
+ },
19
+ contextWindow: 200000,
20
+ maxTokens: 8192,
21
+ },
22
+ "claude-3-5-sonnet-20240620": {
23
+ id: "claude-3-5-sonnet-20240620",
24
+ name: "Claude Sonnet 3.5",
25
+ api: "anthropic-messages",
26
+ provider: "anthropic",
27
+ baseUrl: "https://api.anthropic.com",
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,
46
+ input: ["text", "image"],
47
+ cost: {
48
+ input: 15,
49
+ output: 75,
50
+ cacheRead: 1.5,
51
+ cacheWrite: 18.75,
52
+ },
53
+ contextWindow: 200000,
54
+ maxTokens: 4096,
55
+ },
56
+ "claude-sonnet-4-5-20250929": {
57
+ id: "claude-sonnet-4-5-20250929",
58
+ name: "Claude Sonnet 4.5",
8
59
  api: "anthropic-messages",
9
60
  provider: "anthropic",
10
61
  baseUrl: "https://api.anthropic.com",
@@ -19,9 +70,26 @@ export const MODELS = {
19
70
  contextWindow: 200000,
20
71
  maxTokens: 64000,
21
72
  },
22
- "claude-opus-4-1-20250805": {
23
- id: "claude-opus-4-1-20250805",
24
- name: "Claude Opus 4.1",
73
+ "claude-sonnet-4-20250514": {
74
+ id: "claude-sonnet-4-20250514",
75
+ name: "Claude Sonnet 4",
76
+ api: "anthropic-messages",
77
+ provider: "anthropic",
78
+ baseUrl: "https://api.anthropic.com",
79
+ reasoning: true,
80
+ input: ["text", "image"],
81
+ cost: {
82
+ input: 3,
83
+ output: 15,
84
+ cacheRead: 0.3,
85
+ cacheWrite: 3.75,
86
+ },
87
+ contextWindow: 200000,
88
+ maxTokens: 64000,
89
+ },
90
+ "claude-opus-4-20250514": {
91
+ id: "claude-opus-4-20250514",
92
+ name: "Claude Opus 4",
25
93
  api: "anthropic-messages",
26
94
  provider: "anthropic",
27
95
  baseUrl: "https://api.anthropic.com",
@@ -36,6 +104,23 @@ export const MODELS = {
36
104
  contextWindow: 200000,
37
105
  maxTokens: 32000,
38
106
  },
107
+ "claude-3-5-haiku-20241022": {
108
+ id: "claude-3-5-haiku-20241022",
109
+ name: "Claude Haiku 3.5",
110
+ api: "anthropic-messages",
111
+ provider: "anthropic",
112
+ baseUrl: "https://api.anthropic.com",
113
+ reasoning: false,
114
+ input: ["text", "image"],
115
+ cost: {
116
+ input: 0.8,
117
+ output: 4,
118
+ cacheRead: 0.08,
119
+ cacheWrite: 1,
120
+ },
121
+ contextWindow: 200000,
122
+ maxTokens: 8192,
123
+ },
39
124
  "claude-3-haiku-20240307": {
40
125
  id: "claude-3-haiku-20240307",
41
126
  name: "Claude Haiku 3",
@@ -53,26 +138,26 @@ export const MODELS = {
53
138
  contextWindow: 200000,
54
139
  maxTokens: 4096,
55
140
  },
56
- "claude-3-5-haiku-20241022": {
57
- id: "claude-3-5-haiku-20241022",
58
- name: "Claude Haiku 3.5",
141
+ "claude-3-7-sonnet-20250219": {
142
+ id: "claude-3-7-sonnet-20250219",
143
+ name: "Claude Sonnet 3.7",
59
144
  api: "anthropic-messages",
60
145
  provider: "anthropic",
61
146
  baseUrl: "https://api.anthropic.com",
62
- reasoning: false,
147
+ reasoning: true,
63
148
  input: ["text", "image"],
64
149
  cost: {
65
- input: 0.8,
66
- output: 4,
67
- cacheRead: 0.08,
68
- cacheWrite: 1,
150
+ input: 3,
151
+ output: 15,
152
+ cacheRead: 0.3,
153
+ cacheWrite: 3.75,
69
154
  },
70
155
  contextWindow: 200000,
71
- maxTokens: 8192,
156
+ maxTokens: 64000,
72
157
  },
73
- "claude-opus-4-20250514": {
74
- id: "claude-opus-4-20250514",
75
- name: "Claude Opus 4",
158
+ "claude-opus-4-1-20250805": {
159
+ id: "claude-opus-4-1-20250805",
160
+ name: "Claude Opus 4.1",
76
161
  api: "anthropic-messages",
77
162
  provider: "anthropic",
78
163
  baseUrl: "https://api.anthropic.com",
@@ -87,9 +172,9 @@ export const MODELS = {
87
172
  contextWindow: 200000,
88
173
  maxTokens: 32000,
89
174
  },
90
- "claude-3-5-sonnet-20241022": {
91
- id: "claude-3-5-sonnet-20241022",
92
- name: "Claude Sonnet 3.5 v2",
175
+ "claude-3-sonnet-20240229": {
176
+ id: "claude-3-sonnet-20240229",
177
+ name: "Claude Sonnet 3",
93
178
  api: "anthropic-messages",
94
179
  provider: "anthropic",
95
180
  baseUrl: "https://api.anthropic.com",
@@ -99,18 +184,35 @@ export const MODELS = {
99
184
  input: 3,
100
185
  output: 15,
101
186
  cacheRead: 0.3,
102
- cacheWrite: 3.75,
187
+ cacheWrite: 0.3,
103
188
  },
104
189
  contextWindow: 200000,
105
- maxTokens: 8192,
190
+ maxTokens: 4096,
106
191
  },
107
- "claude-3-5-sonnet-20240620": {
108
- id: "claude-3-5-sonnet-20240620",
109
- name: "Claude Sonnet 3.5",
192
+ "claude-haiku-4-5-20251001": {
193
+ id: "claude-haiku-4-5-20251001",
194
+ name: "Claude Haiku 4.5",
110
195
  api: "anthropic-messages",
111
196
  provider: "anthropic",
112
197
  baseUrl: "https://api.anthropic.com",
113
- reasoning: false,
198
+ reasoning: true,
199
+ input: ["text", "image"],
200
+ cost: {
201
+ input: 1,
202
+ output: 5,
203
+ cacheRead: 0.1,
204
+ cacheWrite: 1.25,
205
+ },
206
+ contextWindow: 200000,
207
+ maxTokens: 64000,
208
+ },
209
+ "claude-sonnet-4-0": {
210
+ id: "claude-sonnet-4-0",
211
+ name: "Claude Sonnet 4",
212
+ api: "anthropic-messages",
213
+ provider: "anthropic",
214
+ baseUrl: "https://api.anthropic.com",
215
+ reasoning: true,
114
216
  input: ["text", "image"],
115
217
  cost: {
116
218
  input: 3,
@@ -119,28 +221,28 @@ export const MODELS = {
119
221
  cacheWrite: 3.75,
120
222
  },
121
223
  contextWindow: 200000,
122
- maxTokens: 8192,
224
+ maxTokens: 64000,
123
225
  },
124
- "claude-3-sonnet-20240229": {
125
- id: "claude-3-sonnet-20240229",
126
- name: "Claude Sonnet 3",
226
+ "claude-3-7-sonnet-latest": {
227
+ id: "claude-3-7-sonnet-latest",
228
+ name: "Claude Sonnet 3.7",
127
229
  api: "anthropic-messages",
128
230
  provider: "anthropic",
129
231
  baseUrl: "https://api.anthropic.com",
130
- reasoning: false,
232
+ reasoning: true,
131
233
  input: ["text", "image"],
132
234
  cost: {
133
235
  input: 3,
134
236
  output: 15,
135
237
  cacheRead: 0.3,
136
- cacheWrite: 0.3,
238
+ cacheWrite: 3.75,
137
239
  },
138
240
  contextWindow: 200000,
139
- maxTokens: 4096,
241
+ maxTokens: 64000,
140
242
  },
141
- "claude-sonnet-4-20250514": {
142
- id: "claude-sonnet-4-20250514",
143
- name: "Claude Sonnet 4",
243
+ "claude-sonnet-4-5": {
244
+ id: "claude-sonnet-4-5",
245
+ name: "Claude Sonnet 4.5",
144
246
  api: "anthropic-messages",
145
247
  provider: "anthropic",
146
248
  baseUrl: "https://api.anthropic.com",
@@ -155,14 +257,48 @@ export const MODELS = {
155
257
  contextWindow: 200000,
156
258
  maxTokens: 64000,
157
259
  },
158
- "claude-3-opus-20240229": {
159
- id: "claude-3-opus-20240229",
160
- name: "Claude Opus 3",
260
+ "claude-3-5-haiku-latest": {
261
+ id: "claude-3-5-haiku-latest",
262
+ name: "Claude Haiku 3.5",
161
263
  api: "anthropic-messages",
162
264
  provider: "anthropic",
163
265
  baseUrl: "https://api.anthropic.com",
164
266
  reasoning: false,
165
267
  input: ["text", "image"],
268
+ cost: {
269
+ input: 0.8,
270
+ output: 4,
271
+ cacheRead: 0.08,
272
+ cacheWrite: 1,
273
+ },
274
+ contextWindow: 200000,
275
+ maxTokens: 8192,
276
+ },
277
+ "claude-haiku-4-5": {
278
+ id: "claude-haiku-4-5",
279
+ name: "Claude Haiku 4.5",
280
+ api: "anthropic-messages",
281
+ provider: "anthropic",
282
+ baseUrl: "https://api.anthropic.com",
283
+ reasoning: true,
284
+ input: ["text", "image"],
285
+ cost: {
286
+ input: 1,
287
+ output: 5,
288
+ cacheRead: 0.1,
289
+ cacheWrite: 1.25,
290
+ },
291
+ contextWindow: 200000,
292
+ maxTokens: 64000,
293
+ },
294
+ "claude-opus-4-1": {
295
+ id: "claude-opus-4-1",
296
+ name: "Claude Opus 4.1",
297
+ api: "anthropic-messages",
298
+ provider: "anthropic",
299
+ baseUrl: "https://api.anthropic.com",
300
+ reasoning: true,
301
+ input: ["text", "image"],
166
302
  cost: {
167
303
  input: 15,
168
304
  output: 75,
@@ -170,7 +306,24 @@ export const MODELS = {
170
306
  cacheWrite: 18.75,
171
307
  },
172
308
  contextWindow: 200000,
173
- maxTokens: 4096,
309
+ maxTokens: 32000,
310
+ },
311
+ "claude-opus-4-0": {
312
+ id: "claude-opus-4-0",
313
+ name: "Claude Opus 4",
314
+ api: "anthropic-messages",
315
+ provider: "anthropic",
316
+ baseUrl: "https://api.anthropic.com",
317
+ reasoning: true,
318
+ input: ["text", "image"],
319
+ cost: {
320
+ input: 15,
321
+ output: 75,
322
+ cacheRead: 1.5,
323
+ cacheWrite: 18.75,
324
+ },
325
+ contextWindow: 200000,
326
+ maxTokens: 32000,
174
327
  },
175
328
  },
176
329
  google: {
@@ -191,103 +344,137 @@ export const MODELS = {
191
344
  contextWindow: 1048576,
192
345
  maxTokens: 65536,
193
346
  },
194
- "gemini-2.5-pro": {
195
- id: "gemini-2.5-pro",
196
- name: "Gemini 2.5 Pro",
347
+ "gemini-flash-lite-latest": {
348
+ id: "gemini-flash-lite-latest",
349
+ name: "Gemini Flash-Lite Latest",
197
350
  api: "google-generative-ai",
198
351
  provider: "google",
199
352
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
200
353
  reasoning: true,
201
354
  input: ["text", "image"],
202
355
  cost: {
203
- input: 1.25,
204
- output: 10,
205
- cacheRead: 0.31,
356
+ input: 0.1,
357
+ output: 0.4,
358
+ cacheRead: 0.025,
206
359
  cacheWrite: 0,
207
360
  },
208
361
  contextWindow: 1048576,
209
362
  maxTokens: 65536,
210
363
  },
211
- "gemini-1.5-flash": {
212
- id: "gemini-1.5-flash",
213
- name: "Gemini 1.5 Flash",
364
+ "gemini-2.5-flash": {
365
+ id: "gemini-2.5-flash",
366
+ name: "Gemini 2.5 Flash",
214
367
  api: "google-generative-ai",
215
368
  provider: "google",
216
369
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
217
- reasoning: false,
370
+ reasoning: true,
218
371
  input: ["text", "image"],
219
372
  cost: {
220
- input: 0.075,
221
- output: 0.3,
222
- cacheRead: 0.01875,
373
+ input: 0.3,
374
+ output: 2.5,
375
+ cacheRead: 0.075,
223
376
  cacheWrite: 0,
224
377
  },
225
- contextWindow: 1000000,
226
- maxTokens: 8192,
378
+ contextWindow: 1048576,
379
+ maxTokens: 65536,
227
380
  },
228
- "gemini-2.0-flash-lite": {
229
- id: "gemini-2.0-flash-lite",
230
- name: "Gemini 2.0 Flash Lite",
381
+ "gemini-flash-latest": {
382
+ id: "gemini-flash-latest",
383
+ name: "Gemini Flash Latest",
231
384
  api: "google-generative-ai",
232
385
  provider: "google",
233
386
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
234
- reasoning: false,
387
+ reasoning: true,
235
388
  input: ["text", "image"],
236
389
  cost: {
237
- input: 0.075,
238
- output: 0.3,
239
- cacheRead: 0,
390
+ input: 0.3,
391
+ output: 2.5,
392
+ cacheRead: 0.075,
240
393
  cacheWrite: 0,
241
394
  },
242
395
  contextWindow: 1048576,
243
- maxTokens: 8192,
396
+ maxTokens: 65536,
244
397
  },
245
- "gemini-1.5-pro": {
246
- id: "gemini-1.5-pro",
247
- name: "Gemini 1.5 Pro",
248
- api: "google-generative-ai",
249
- provider: "google",
398
+ "gemini-2.5-pro-preview-05-06": {
399
+ id: "gemini-2.5-pro-preview-05-06",
400
+ name: "Gemini 2.5 Pro Preview 05-06",
401
+ api: "google-generative-ai",
402
+ provider: "google",
250
403
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
251
- reasoning: false,
404
+ reasoning: true,
252
405
  input: ["text", "image"],
253
406
  cost: {
254
407
  input: 1.25,
255
- output: 5,
256
- cacheRead: 0.3125,
408
+ output: 10,
409
+ cacheRead: 0.31,
257
410
  cacheWrite: 0,
258
411
  },
259
- contextWindow: 1000000,
412
+ contextWindow: 1048576,
413
+ maxTokens: 65536,
414
+ },
415
+ "gemini-2.0-flash-lite": {
416
+ id: "gemini-2.0-flash-lite",
417
+ name: "Gemini 2.0 Flash Lite",
418
+ api: "google-generative-ai",
419
+ provider: "google",
420
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
421
+ reasoning: false,
422
+ input: ["text", "image"],
423
+ cost: {
424
+ input: 0.075,
425
+ output: 0.3,
426
+ cacheRead: 0,
427
+ cacheWrite: 0,
428
+ },
429
+ contextWindow: 1048576,
260
430
  maxTokens: 8192,
261
431
  },
262
- "gemini-1.5-flash-8b": {
263
- id: "gemini-1.5-flash-8b",
264
- name: "Gemini 1.5 Flash-8B",
432
+ "gemini-live-2.5-flash-preview-native-audio": {
433
+ id: "gemini-live-2.5-flash-preview-native-audio",
434
+ name: "Gemini Live 2.5 Flash Preview Native Audio",
435
+ api: "google-generative-ai",
436
+ provider: "google",
437
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
438
+ reasoning: true,
439
+ input: ["text"],
440
+ cost: {
441
+ input: 0.5,
442
+ output: 2,
443
+ cacheRead: 0,
444
+ cacheWrite: 0,
445
+ },
446
+ contextWindow: 131072,
447
+ maxTokens: 65536,
448
+ },
449
+ "gemini-2.0-flash": {
450
+ id: "gemini-2.0-flash",
451
+ name: "Gemini 2.0 Flash",
265
452
  api: "google-generative-ai",
266
453
  provider: "google",
267
454
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
268
455
  reasoning: false,
269
456
  input: ["text", "image"],
270
457
  cost: {
271
- input: 0.0375,
272
- output: 0.15,
273
- cacheRead: 0.01,
458
+ input: 0.1,
459
+ output: 0.4,
460
+ cacheRead: 0.025,
274
461
  cacheWrite: 0,
275
462
  },
276
- contextWindow: 1000000,
463
+ contextWindow: 1048576,
277
464
  maxTokens: 8192,
278
465
  },
279
- "gemini-2.5-flash": {
280
- id: "gemini-2.5-flash",
281
- name: "Gemini 2.5 Flash",
466
+ "gemini-2.5-flash-lite": {
467
+ id: "gemini-2.5-flash-lite",
468
+ name: "Gemini 2.5 Flash Lite",
282
469
  api: "google-generative-ai",
283
470
  provider: "google",
284
471
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
285
472
  reasoning: true,
286
473
  input: ["text", "image"],
287
474
  cost: {
288
- input: 0.3,
289
- output: 2.5,
290
- cacheRead: 0.075,
475
+ input: 0.1,
476
+ output: 0.4,
477
+ cacheRead: 0.025,
291
478
  cacheWrite: 0,
292
479
  },
293
480
  contextWindow: 1048576,
@@ -310,30 +497,30 @@ export const MODELS = {
310
497
  contextWindow: 1048576,
311
498
  maxTokens: 65536,
312
499
  },
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",
500
+ "gemini-live-2.5-flash": {
501
+ id: "gemini-live-2.5-flash",
502
+ name: "Gemini Live 2.5 Flash",
316
503
  api: "google-generative-ai",
317
504
  provider: "google",
318
505
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
319
506
  reasoning: true,
320
507
  input: ["text", "image"],
321
508
  cost: {
322
- input: 1.25,
323
- output: 10,
324
- cacheRead: 0.31,
509
+ input: 0.5,
510
+ output: 2,
511
+ cacheRead: 0,
325
512
  cacheWrite: 0,
326
513
  },
327
- contextWindow: 1048576,
328
- maxTokens: 65536,
514
+ contextWindow: 128000,
515
+ maxTokens: 8000,
329
516
  },
330
- "gemini-2.0-flash": {
331
- id: "gemini-2.0-flash",
332
- name: "Gemini 2.0 Flash",
517
+ "gemini-2.5-flash-lite-preview-06-17": {
518
+ id: "gemini-2.5-flash-lite-preview-06-17",
519
+ name: "Gemini 2.5 Flash Lite Preview 06-17",
333
520
  api: "google-generative-ai",
334
521
  provider: "google",
335
522
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
336
- reasoning: false,
523
+ reasoning: true,
337
524
  input: ["text", "image"],
338
525
  cost: {
339
526
  input: 0.1,
@@ -342,23 +529,23 @@ export const MODELS = {
342
529
  cacheWrite: 0,
343
530
  },
344
531
  contextWindow: 1048576,
345
- maxTokens: 8192,
532
+ maxTokens: 65536,
346
533
  },
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",
534
+ "gemini-2.5-flash-preview-09-2025": {
535
+ id: "gemini-2.5-flash-preview-09-2025",
536
+ name: "Gemini 2.5 Flash Preview 09-25",
350
537
  api: "google-generative-ai",
351
538
  provider: "google",
352
539
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
353
540
  reasoning: true,
354
541
  input: ["text", "image"],
355
542
  cost: {
356
- input: 0.1,
357
- output: 0.4,
358
- cacheRead: 0.025,
543
+ input: 0.3,
544
+ output: 2.5,
545
+ cacheRead: 0.075,
359
546
  cacheWrite: 0,
360
547
  },
361
- contextWindow: 65536,
548
+ contextWindow: 1048576,
362
549
  maxTokens: 65536,
363
550
  },
364
551
  "gemini-2.5-flash-preview-04-17": {
@@ -378,262 +565,296 @@ export const MODELS = {
378
565
  contextWindow: 1048576,
379
566
  maxTokens: 65536,
380
567
  },
381
- },
382
- openai: {
383
- "gpt-5-nano": {
384
- id: "gpt-5-nano",
385
- name: "GPT-5 Nano",
386
- api: "openai-responses",
387
- provider: "openai",
388
- baseUrl: "https://api.openai.com/v1",
568
+ "gemini-2.5-pro": {
569
+ id: "gemini-2.5-pro",
570
+ name: "Gemini 2.5 Pro",
571
+ api: "google-generative-ai",
572
+ provider: "google",
573
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
389
574
  reasoning: true,
390
575
  input: ["text", "image"],
391
576
  cost: {
392
- input: 0.05,
393
- output: 0.4,
577
+ input: 1.25,
578
+ output: 10,
579
+ cacheRead: 0.31,
580
+ cacheWrite: 0,
581
+ },
582
+ contextWindow: 1048576,
583
+ maxTokens: 65536,
584
+ },
585
+ "gemini-1.5-flash": {
586
+ id: "gemini-1.5-flash",
587
+ name: "Gemini 1.5 Flash",
588
+ api: "google-generative-ai",
589
+ provider: "google",
590
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
591
+ reasoning: false,
592
+ input: ["text", "image"],
593
+ cost: {
594
+ input: 0.075,
595
+ output: 0.3,
596
+ cacheRead: 0.01875,
597
+ cacheWrite: 0,
598
+ },
599
+ contextWindow: 1000000,
600
+ maxTokens: 8192,
601
+ },
602
+ "gemini-1.5-flash-8b": {
603
+ id: "gemini-1.5-flash-8b",
604
+ name: "Gemini 1.5 Flash-8B",
605
+ api: "google-generative-ai",
606
+ provider: "google",
607
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
608
+ reasoning: false,
609
+ input: ["text", "image"],
610
+ cost: {
611
+ input: 0.0375,
612
+ output: 0.15,
394
613
  cacheRead: 0.01,
395
614
  cacheWrite: 0,
396
615
  },
397
- contextWindow: 400000,
398
- maxTokens: 128000,
616
+ contextWindow: 1000000,
617
+ maxTokens: 8192,
399
618
  },
400
- "o3-pro": {
401
- id: "o3-pro",
402
- name: "o3-pro",
403
- api: "openai-responses",
404
- provider: "openai",
405
- baseUrl: "https://api.openai.com/v1",
619
+ "gemini-2.5-flash-lite-preview-09-2025": {
620
+ id: "gemini-2.5-flash-lite-preview-09-2025",
621
+ name: "Gemini 2.5 Flash Lite Preview 09-25",
622
+ api: "google-generative-ai",
623
+ provider: "google",
624
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
406
625
  reasoning: true,
407
626
  input: ["text", "image"],
408
627
  cost: {
409
- input: 20,
410
- output: 80,
411
- cacheRead: 0,
628
+ input: 0.1,
629
+ output: 0.4,
630
+ cacheRead: 0.025,
412
631
  cacheWrite: 0,
413
632
  },
414
- contextWindow: 200000,
415
- maxTokens: 100000,
633
+ contextWindow: 1048576,
634
+ maxTokens: 65536,
416
635
  },
417
- "codex-mini-latest": {
418
- id: "codex-mini-latest",
419
- name: "Codex Mini",
420
- api: "openai-responses",
421
- provider: "openai",
422
- baseUrl: "https://api.openai.com/v1",
423
- reasoning: true,
424
- input: ["text"],
636
+ "gemini-1.5-pro": {
637
+ id: "gemini-1.5-pro",
638
+ name: "Gemini 1.5 Pro",
639
+ api: "google-generative-ai",
640
+ provider: "google",
641
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
642
+ reasoning: false,
643
+ input: ["text", "image"],
425
644
  cost: {
426
- input: 1.5,
427
- output: 6,
428
- cacheRead: 0.375,
645
+ input: 1.25,
646
+ output: 5,
647
+ cacheRead: 0.3125,
429
648
  cacheWrite: 0,
430
649
  },
431
- contextWindow: 200000,
432
- maxTokens: 100000,
650
+ contextWindow: 1000000,
651
+ maxTokens: 8192,
433
652
  },
434
- "gpt-4.1": {
435
- id: "gpt-4.1",
436
- name: "GPT-4.1",
653
+ },
654
+ openai: {
655
+ "gpt-4.1-nano": {
656
+ id: "gpt-4.1-nano",
657
+ name: "GPT-4.1 nano",
437
658
  api: "openai-responses",
438
659
  provider: "openai",
439
660
  baseUrl: "https://api.openai.com/v1",
440
661
  reasoning: false,
441
662
  input: ["text", "image"],
442
663
  cost: {
443
- input: 2,
444
- output: 8,
445
- cacheRead: 0.5,
664
+ input: 0.1,
665
+ output: 0.4,
666
+ cacheRead: 0.03,
446
667
  cacheWrite: 0,
447
668
  },
448
669
  contextWindow: 1047576,
449
670
  maxTokens: 32768,
450
671
  },
451
- "gpt-4-turbo": {
452
- id: "gpt-4-turbo",
453
- name: "GPT-4 Turbo",
672
+ "gpt-4": {
673
+ id: "gpt-4",
674
+ name: "GPT-4",
454
675
  api: "openai-responses",
455
676
  provider: "openai",
456
677
  baseUrl: "https://api.openai.com/v1",
457
678
  reasoning: false,
458
- input: ["text", "image"],
679
+ input: ["text"],
459
680
  cost: {
460
- input: 10,
461
- output: 30,
681
+ input: 30,
682
+ output: 60,
462
683
  cacheRead: 0,
463
684
  cacheWrite: 0,
464
685
  },
465
- contextWindow: 128000,
466
- maxTokens: 4096,
686
+ contextWindow: 8192,
687
+ maxTokens: 8192,
467
688
  },
468
- "o1": {
469
- id: "o1",
470
- name: "o1",
689
+ "o1-pro": {
690
+ id: "o1-pro",
691
+ name: "o1-pro",
471
692
  api: "openai-responses",
472
693
  provider: "openai",
473
694
  baseUrl: "https://api.openai.com/v1",
474
695
  reasoning: true,
475
696
  input: ["text", "image"],
476
697
  cost: {
477
- input: 15,
478
- output: 60,
479
- cacheRead: 7.5,
698
+ input: 150,
699
+ output: 600,
700
+ cacheRead: 0,
480
701
  cacheWrite: 0,
481
702
  },
482
703
  contextWindow: 200000,
483
704
  maxTokens: 100000,
484
705
  },
485
- "o3-deep-research": {
486
- id: "o3-deep-research",
487
- name: "o3-deep-research",
706
+ "gpt-4o-2024-05-13": {
707
+ id: "gpt-4o-2024-05-13",
708
+ name: "GPT-4o (2024-05-13)",
488
709
  api: "openai-responses",
489
710
  provider: "openai",
490
711
  baseUrl: "https://api.openai.com/v1",
491
- reasoning: true,
712
+ reasoning: false,
492
713
  input: ["text", "image"],
493
714
  cost: {
494
- input: 10,
495
- output: 40,
496
- cacheRead: 2.5,
715
+ input: 5,
716
+ output: 15,
717
+ cacheRead: 0,
497
718
  cacheWrite: 0,
498
719
  },
499
- contextWindow: 200000,
500
- maxTokens: 100000,
720
+ contextWindow: 128000,
721
+ maxTokens: 4096,
501
722
  },
502
- "gpt-5": {
503
- id: "gpt-5",
504
- name: "GPT-5",
723
+ "gpt-4o-2024-08-06": {
724
+ id: "gpt-4o-2024-08-06",
725
+ name: "GPT-4o (2024-08-06)",
505
726
  api: "openai-responses",
506
727
  provider: "openai",
507
728
  baseUrl: "https://api.openai.com/v1",
508
- reasoning: true,
729
+ reasoning: false,
509
730
  input: ["text", "image"],
510
731
  cost: {
511
- input: 1.25,
732
+ input: 2.5,
512
733
  output: 10,
513
- cacheRead: 0.13,
734
+ cacheRead: 1.25,
514
735
  cacheWrite: 0,
515
736
  },
516
- contextWindow: 400000,
517
- maxTokens: 128000,
737
+ contextWindow: 128000,
738
+ maxTokens: 16384,
518
739
  },
519
- "o1-pro": {
520
- id: "o1-pro",
521
- name: "o1-pro",
740
+ "gpt-4.1-mini": {
741
+ id: "gpt-4.1-mini",
742
+ name: "GPT-4.1 mini",
522
743
  api: "openai-responses",
523
744
  provider: "openai",
524
745
  baseUrl: "https://api.openai.com/v1",
525
- reasoning: true,
746
+ reasoning: false,
526
747
  input: ["text", "image"],
527
748
  cost: {
528
- input: 150,
529
- output: 600,
530
- cacheRead: 0,
749
+ input: 0.4,
750
+ output: 1.6,
751
+ cacheRead: 0.1,
531
752
  cacheWrite: 0,
532
753
  },
533
- contextWindow: 200000,
534
- maxTokens: 100000,
754
+ contextWindow: 1047576,
755
+ maxTokens: 32768,
535
756
  },
536
- "o3": {
537
- id: "o3",
538
- name: "o3",
757
+ "o3-deep-research": {
758
+ id: "o3-deep-research",
759
+ name: "o3-deep-research",
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: 10,
767
+ output: 40,
768
+ cacheRead: 2.5,
548
769
  cacheWrite: 0,
549
770
  },
550
771
  contextWindow: 200000,
551
772
  maxTokens: 100000,
552
773
  },
553
- "gpt-5-mini": {
554
- id: "gpt-5-mini",
555
- name: "GPT-5 Mini",
774
+ "gpt-4-turbo": {
775
+ id: "gpt-4-turbo",
776
+ name: "GPT-4 Turbo",
556
777
  api: "openai-responses",
557
778
  provider: "openai",
558
779
  baseUrl: "https://api.openai.com/v1",
559
- reasoning: true,
780
+ reasoning: false,
560
781
  input: ["text", "image"],
561
782
  cost: {
562
- input: 0.25,
563
- output: 2,
564
- cacheRead: 0.03,
783
+ input: 10,
784
+ output: 30,
785
+ cacheRead: 0,
565
786
  cacheWrite: 0,
566
787
  },
567
- contextWindow: 400000,
568
- maxTokens: 128000,
788
+ contextWindow: 128000,
789
+ maxTokens: 4096,
569
790
  },
570
- "o4-mini-deep-research": {
571
- id: "o4-mini-deep-research",
572
- name: "o4-mini-deep-research",
791
+ "o3-mini": {
792
+ id: "o3-mini",
793
+ name: "o3-mini",
573
794
  api: "openai-responses",
574
795
  provider: "openai",
575
796
  baseUrl: "https://api.openai.com/v1",
576
797
  reasoning: true,
577
- input: ["text", "image"],
798
+ input: ["text"],
578
799
  cost: {
579
- input: 2,
580
- output: 8,
581
- cacheRead: 0.5,
800
+ input: 1.1,
801
+ output: 4.4,
802
+ cacheRead: 0.55,
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
+ "codex-mini-latest": {
809
+ id: "codex-mini-latest",
810
+ name: "Codex Mini",
590
811
  api: "openai-responses",
591
812
  provider: "openai",
592
813
  baseUrl: "https://api.openai.com/v1",
593
- reasoning: false,
594
- input: ["text", "image"],
814
+ reasoning: true,
815
+ input: ["text"],
595
816
  cost: {
596
- input: 0.15,
597
- output: 0.6,
598
- cacheRead: 0.08,
817
+ input: 1.5,
818
+ output: 6,
819
+ cacheRead: 0.375,
599
820
  cacheWrite: 0,
600
821
  },
601
- contextWindow: 128000,
602
- maxTokens: 16384,
822
+ contextWindow: 200000,
823
+ maxTokens: 100000,
603
824
  },
604
- "gpt-4.1-nano": {
605
- id: "gpt-4.1-nano",
606
- name: "GPT-4.1 nano",
825
+ "gpt-5-nano": {
826
+ id: "gpt-5-nano",
827
+ name: "GPT-5 Nano",
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,
834
+ input: 0.05,
614
835
  output: 0.4,
615
- cacheRead: 0.03,
836
+ cacheRead: 0.01,
616
837
  cacheWrite: 0,
617
838
  },
618
- contextWindow: 1047576,
619
- maxTokens: 32768,
839
+ contextWindow: 400000,
840
+ maxTokens: 128000,
620
841
  },
621
- "gpt-4.1-mini": {
622
- id: "gpt-4.1-mini",
623
- name: "GPT-4.1 mini",
842
+ "gpt-5-codex": {
843
+ id: "gpt-5-codex",
844
+ name: "GPT-5-Codex",
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: 1.25,
852
+ output: 10,
853
+ cacheRead: 0.125,
633
854
  cacheWrite: 0,
634
855
  },
635
- contextWindow: 1047576,
636
- maxTokens: 32768,
856
+ contextWindow: 400000,
857
+ maxTokens: 128000,
637
858
  },
638
859
  "gpt-4o": {
639
860
  id: "gpt-4o",
@@ -652,22 +873,22 @@ export const MODELS = {
652
873
  contextWindow: 128000,
653
874
  maxTokens: 16384,
654
875
  },
655
- "gpt-4": {
656
- id: "gpt-4",
657
- name: "GPT-4",
876
+ "gpt-4.1": {
877
+ id: "gpt-4.1",
878
+ name: "GPT-4.1",
658
879
  api: "openai-responses",
659
880
  provider: "openai",
660
881
  baseUrl: "https://api.openai.com/v1",
661
882
  reasoning: false,
662
- input: ["text"],
883
+ input: ["text", "image"],
663
884
  cost: {
664
- input: 30,
665
- output: 60,
666
- cacheRead: 0,
885
+ input: 2,
886
+ output: 8,
887
+ cacheRead: 0.5,
667
888
  cacheWrite: 0,
668
889
  },
669
- contextWindow: 8192,
670
- maxTokens: 8192,
890
+ contextWindow: 1047576,
891
+ maxTokens: 32768,
671
892
  },
672
893
  "o4-mini": {
673
894
  id: "o4-mini",
@@ -686,110 +907,195 @@ export const MODELS = {
686
907
  contextWindow: 200000,
687
908
  maxTokens: 100000,
688
909
  },
689
- "o3-mini": {
690
- id: "o3-mini",
691
- name: "o3-mini",
910
+ "o1": {
911
+ id: "o1",
912
+ name: "o1",
692
913
  api: "openai-responses",
693
914
  provider: "openai",
694
915
  baseUrl: "https://api.openai.com/v1",
695
916
  reasoning: true,
696
- input: ["text"],
917
+ input: ["text", "image"],
697
918
  cost: {
698
- input: 1.1,
699
- output: 4.4,
700
- cacheRead: 0.55,
919
+ input: 15,
920
+ output: 60,
921
+ cacheRead: 7.5,
701
922
  cacheWrite: 0,
702
923
  },
703
924
  contextWindow: 200000,
704
925
  maxTokens: 100000,
705
926
  },
706
- "gpt-5-chat-latest": {
707
- id: "gpt-5-chat-latest",
708
- name: "GPT-5 Chat Latest",
927
+ "gpt-5-mini": {
928
+ id: "gpt-5-mini",
929
+ name: "GPT-5 Mini",
709
930
  api: "openai-responses",
710
931
  provider: "openai",
711
932
  baseUrl: "https://api.openai.com/v1",
712
- reasoning: false,
933
+ reasoning: true,
713
934
  input: ["text", "image"],
714
935
  cost: {
715
- input: 1.25,
716
- output: 10,
717
- cacheRead: 0.125,
936
+ input: 0.25,
937
+ output: 2,
938
+ cacheRead: 0.03,
718
939
  cacheWrite: 0,
719
940
  },
720
- contextWindow: 128000,
721
- maxTokens: 16384,
941
+ contextWindow: 400000,
942
+ maxTokens: 128000,
722
943
  },
723
- },
724
- groq: {
725
- "llama-3.1-8b-instant": {
726
- id: "llama-3.1-8b-instant",
727
- name: "Llama 3.1 8B Instant",
728
- api: "openai-completions",
729
- provider: "groq",
730
- baseUrl: "https://api.groq.com/openai/v1",
731
- reasoning: false,
732
- input: ["text"],
944
+ "o3-pro": {
945
+ id: "o3-pro",
946
+ name: "o3-pro",
947
+ api: "openai-responses",
948
+ provider: "openai",
949
+ baseUrl: "https://api.openai.com/v1",
950
+ reasoning: true,
951
+ input: ["text", "image"],
733
952
  cost: {
734
- input: 0.05,
735
- output: 0.08,
953
+ input: 20,
954
+ output: 80,
736
955
  cacheRead: 0,
737
956
  cacheWrite: 0,
738
957
  },
739
- contextWindow: 131072,
740
- maxTokens: 8192,
958
+ contextWindow: 200000,
959
+ maxTokens: 100000,
741
960
  },
742
- "qwen-qwq-32b": {
743
- id: "qwen-qwq-32b",
744
- name: "Qwen QwQ 32B",
745
- api: "openai-completions",
746
- provider: "groq",
747
- baseUrl: "https://api.groq.com/openai/v1",
748
- reasoning: true,
749
- input: ["text"],
961
+ "gpt-4o-2024-11-20": {
962
+ id: "gpt-4o-2024-11-20",
963
+ name: "GPT-4o (2024-11-20)",
964
+ api: "openai-responses",
965
+ provider: "openai",
966
+ baseUrl: "https://api.openai.com/v1",
967
+ reasoning: false,
968
+ input: ["text", "image"],
750
969
  cost: {
751
- input: 0.29,
752
- output: 0.39,
753
- cacheRead: 0,
970
+ input: 2.5,
971
+ output: 10,
972
+ cacheRead: 1.25,
754
973
  cacheWrite: 0,
755
974
  },
756
- contextWindow: 131072,
975
+ contextWindow: 128000,
757
976
  maxTokens: 16384,
758
977
  },
759
- "llama3-70b-8192": {
760
- id: "llama3-70b-8192",
761
- name: "Llama 3 70B",
762
- api: "openai-completions",
763
- provider: "groq",
764
- baseUrl: "https://api.groq.com/openai/v1",
765
- reasoning: false,
766
- input: ["text"],
978
+ "o3": {
979
+ id: "o3",
980
+ name: "o3",
981
+ api: "openai-responses",
982
+ provider: "openai",
983
+ baseUrl: "https://api.openai.com/v1",
984
+ reasoning: true,
985
+ input: ["text", "image"],
767
986
  cost: {
768
- input: 0.59,
769
- output: 0.79,
770
- cacheRead: 0,
987
+ input: 2,
988
+ output: 8,
989
+ cacheRead: 0.5,
771
990
  cacheWrite: 0,
772
991
  },
773
- contextWindow: 8192,
774
- maxTokens: 8192,
992
+ contextWindow: 200000,
993
+ maxTokens: 100000,
775
994
  },
776
- "deepseek-r1-distill-llama-70b": {
777
- id: "deepseek-r1-distill-llama-70b",
778
- name: "DeepSeek R1 Distill Llama 70B",
779
- api: "openai-completions",
780
- provider: "groq",
781
- baseUrl: "https://api.groq.com/openai/v1",
995
+ "o4-mini-deep-research": {
996
+ id: "o4-mini-deep-research",
997
+ name: "o4-mini-deep-research",
998
+ api: "openai-responses",
999
+ provider: "openai",
1000
+ baseUrl: "https://api.openai.com/v1",
1001
+ reasoning: true,
1002
+ input: ["text", "image"],
1003
+ cost: {
1004
+ input: 2,
1005
+ output: 8,
1006
+ cacheRead: 0.5,
1007
+ cacheWrite: 0,
1008
+ },
1009
+ contextWindow: 200000,
1010
+ maxTokens: 100000,
1011
+ },
1012
+ "gpt-4o-mini": {
1013
+ id: "gpt-4o-mini",
1014
+ name: "GPT-4o mini",
1015
+ api: "openai-responses",
1016
+ provider: "openai",
1017
+ baseUrl: "https://api.openai.com/v1",
1018
+ reasoning: false,
1019
+ input: ["text", "image"],
1020
+ cost: {
1021
+ input: 0.15,
1022
+ output: 0.6,
1023
+ cacheRead: 0.08,
1024
+ cacheWrite: 0,
1025
+ },
1026
+ contextWindow: 128000,
1027
+ maxTokens: 16384,
1028
+ },
1029
+ "gpt-5": {
1030
+ id: "gpt-5",
1031
+ name: "GPT-5",
1032
+ api: "openai-responses",
1033
+ provider: "openai",
1034
+ baseUrl: "https://api.openai.com/v1",
782
1035
  reasoning: true,
1036
+ input: ["text", "image"],
1037
+ cost: {
1038
+ input: 1.25,
1039
+ output: 10,
1040
+ cacheRead: 0.13,
1041
+ cacheWrite: 0,
1042
+ },
1043
+ contextWindow: 400000,
1044
+ maxTokens: 128000,
1045
+ },
1046
+ "gpt-5-chat-latest": {
1047
+ id: "gpt-5-chat-latest",
1048
+ name: "GPT-5 Chat Latest",
1049
+ api: "openai-responses",
1050
+ provider: "openai",
1051
+ baseUrl: "https://api.openai.com/v1",
1052
+ reasoning: false,
1053
+ input: ["text", "image"],
1054
+ cost: {
1055
+ input: 1.25,
1056
+ output: 10,
1057
+ cacheRead: 0.125,
1058
+ cacheWrite: 0,
1059
+ },
1060
+ contextWindow: 128000,
1061
+ maxTokens: 16384,
1062
+ },
1063
+ },
1064
+ groq: {
1065
+ "llama-3.1-8b-instant": {
1066
+ id: "llama-3.1-8b-instant",
1067
+ name: "Llama 3.1 8B Instant",
1068
+ api: "openai-completions",
1069
+ provider: "groq",
1070
+ baseUrl: "https://api.groq.com/openai/v1",
1071
+ reasoning: false,
783
1072
  input: ["text"],
784
1073
  cost: {
785
- input: 0.75,
786
- output: 0.99,
1074
+ input: 0.05,
1075
+ output: 0.08,
787
1076
  cacheRead: 0,
788
1077
  cacheWrite: 0,
789
1078
  },
790
1079
  contextWindow: 131072,
791
1080
  maxTokens: 8192,
792
1081
  },
1082
+ "mistral-saba-24b": {
1083
+ id: "mistral-saba-24b",
1084
+ name: "Mistral Saba 24B",
1085
+ api: "openai-completions",
1086
+ provider: "groq",
1087
+ baseUrl: "https://api.groq.com/openai/v1",
1088
+ reasoning: false,
1089
+ input: ["text"],
1090
+ cost: {
1091
+ input: 0.79,
1092
+ output: 0.79,
1093
+ cacheRead: 0,
1094
+ cacheWrite: 0,
1095
+ },
1096
+ contextWindow: 32768,
1097
+ maxTokens: 32768,
1098
+ },
793
1099
  "llama3-8b-8192": {
794
1100
  id: "llama3-8b-8192",
795
1101
  name: "Llama 3 8B",
@@ -807,6 +1113,57 @@ export const MODELS = {
807
1113
  contextWindow: 8192,
808
1114
  maxTokens: 8192,
809
1115
  },
1116
+ "qwen-qwq-32b": {
1117
+ id: "qwen-qwq-32b",
1118
+ name: "Qwen QwQ 32B",
1119
+ api: "openai-completions",
1120
+ provider: "groq",
1121
+ baseUrl: "https://api.groq.com/openai/v1",
1122
+ reasoning: true,
1123
+ input: ["text"],
1124
+ cost: {
1125
+ input: 0.29,
1126
+ output: 0.39,
1127
+ cacheRead: 0,
1128
+ cacheWrite: 0,
1129
+ },
1130
+ contextWindow: 131072,
1131
+ maxTokens: 16384,
1132
+ },
1133
+ "llama3-70b-8192": {
1134
+ id: "llama3-70b-8192",
1135
+ name: "Llama 3 70B",
1136
+ api: "openai-completions",
1137
+ provider: "groq",
1138
+ baseUrl: "https://api.groq.com/openai/v1",
1139
+ reasoning: false,
1140
+ input: ["text"],
1141
+ cost: {
1142
+ input: 0.59,
1143
+ output: 0.79,
1144
+ cacheRead: 0,
1145
+ cacheWrite: 0,
1146
+ },
1147
+ contextWindow: 8192,
1148
+ maxTokens: 8192,
1149
+ },
1150
+ "deepseek-r1-distill-llama-70b": {
1151
+ id: "deepseek-r1-distill-llama-70b",
1152
+ name: "DeepSeek R1 Distill Llama 70B",
1153
+ api: "openai-completions",
1154
+ provider: "groq",
1155
+ baseUrl: "https://api.groq.com/openai/v1",
1156
+ reasoning: true,
1157
+ input: ["text"],
1158
+ cost: {
1159
+ input: 0.75,
1160
+ output: 0.99,
1161
+ cacheRead: 0,
1162
+ cacheWrite: 0,
1163
+ },
1164
+ contextWindow: 131072,
1165
+ maxTokens: 8192,
1166
+ },
810
1167
  "gemma2-9b-it": {
811
1168
  id: "gemma2-9b-it",
812
1169
  name: "Gemma 2 9B",
@@ -841,22 +1198,39 @@ export const MODELS = {
841
1198
  contextWindow: 131072,
842
1199
  maxTokens: 32768,
843
1200
  },
844
- "mistral-saba-24b": {
845
- id: "mistral-saba-24b",
846
- name: "Mistral Saba 24B",
1201
+ "moonshotai/kimi-k2-instruct-0905": {
1202
+ id: "moonshotai/kimi-k2-instruct-0905",
1203
+ name: "Kimi K2 Instruct 0905",
847
1204
  api: "openai-completions",
848
1205
  provider: "groq",
849
1206
  baseUrl: "https://api.groq.com/openai/v1",
850
1207
  reasoning: false,
851
1208
  input: ["text"],
852
1209
  cost: {
853
- input: 0.79,
854
- output: 0.79,
1210
+ input: 1,
1211
+ output: 3,
855
1212
  cacheRead: 0,
856
1213
  cacheWrite: 0,
857
1214
  },
858
- contextWindow: 32768,
859
- maxTokens: 32768,
1215
+ contextWindow: 262144,
1216
+ maxTokens: 16384,
1217
+ },
1218
+ "moonshotai/kimi-k2-instruct": {
1219
+ id: "moonshotai/kimi-k2-instruct",
1220
+ name: "Kimi K2 Instruct",
1221
+ api: "openai-completions",
1222
+ provider: "groq",
1223
+ baseUrl: "https://api.groq.com/openai/v1",
1224
+ reasoning: false,
1225
+ input: ["text"],
1226
+ cost: {
1227
+ input: 1,
1228
+ output: 3,
1229
+ cacheRead: 0,
1230
+ cacheWrite: 0,
1231
+ },
1232
+ contextWindow: 131072,
1233
+ maxTokens: 16384,
860
1234
  },
861
1235
  "openai/gpt-oss-20b": {
862
1236
  id: "openai/gpt-oss-20b",
@@ -892,40 +1266,6 @@ export const MODELS = {
892
1266
  contextWindow: 131072,
893
1267
  maxTokens: 32768,
894
1268
  },
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",
915
- api: "openai-completions",
916
- provider: "groq",
917
- baseUrl: "https://api.groq.com/openai/v1",
918
- reasoning: false,
919
- input: ["text", "image"],
920
- cost: {
921
- input: 0.11,
922
- output: 0.34,
923
- cacheRead: 0,
924
- cacheWrite: 0,
925
- },
926
- contextWindow: 131072,
927
- maxTokens: 8192,
928
- },
929
1269
  "qwen/qwen3-32b": {
930
1270
  id: "qwen/qwen3-32b",
931
1271
  name: "Qwen3 32B",
@@ -943,39 +1283,39 @@ export const MODELS = {
943
1283
  contextWindow: 131072,
944
1284
  maxTokens: 16384,
945
1285
  },
946
- "moonshotai/kimi-k2-instruct-0905": {
947
- id: "moonshotai/kimi-k2-instruct-0905",
948
- name: "Kimi K2 Instruct 0905",
1286
+ "meta-llama/llama-4-scout-17b-16e-instruct": {
1287
+ id: "meta-llama/llama-4-scout-17b-16e-instruct",
1288
+ name: "Llama 4 Scout 17B",
949
1289
  api: "openai-completions",
950
1290
  provider: "groq",
951
1291
  baseUrl: "https://api.groq.com/openai/v1",
952
1292
  reasoning: false,
953
- input: ["text"],
1293
+ input: ["text", "image"],
954
1294
  cost: {
955
- input: 1,
956
- output: 3,
1295
+ input: 0.11,
1296
+ output: 0.34,
957
1297
  cacheRead: 0,
958
1298
  cacheWrite: 0,
959
1299
  },
960
- contextWindow: 262144,
961
- maxTokens: 16384,
1300
+ contextWindow: 131072,
1301
+ maxTokens: 8192,
962
1302
  },
963
- "moonshotai/kimi-k2-instruct": {
964
- id: "moonshotai/kimi-k2-instruct",
965
- name: "Kimi K2 Instruct",
1303
+ "meta-llama/llama-4-maverick-17b-128e-instruct": {
1304
+ id: "meta-llama/llama-4-maverick-17b-128e-instruct",
1305
+ name: "Llama 4 Maverick 17B",
966
1306
  api: "openai-completions",
967
1307
  provider: "groq",
968
1308
  baseUrl: "https://api.groq.com/openai/v1",
969
1309
  reasoning: false,
970
- input: ["text"],
1310
+ input: ["text", "image"],
971
1311
  cost: {
972
- input: 1,
973
- output: 3,
1312
+ input: 0.2,
1313
+ output: 0.6,
974
1314
  cacheRead: 0,
975
1315
  cacheWrite: 0,
976
1316
  },
977
1317
  contextWindow: 131072,
978
- maxTokens: 16384,
1318
+ maxTokens: 8192,
979
1319
  },
980
1320
  },
981
1321
  cerebras: {
@@ -996,139 +1336,156 @@ export const MODELS = {
996
1336
  contextWindow: 131000,
997
1337
  maxTokens: 32000,
998
1338
  },
999
- "gpt-oss-120b": {
1000
- id: "gpt-oss-120b",
1001
- name: "GPT OSS 120B",
1339
+ "qwen-3-coder-480b": {
1340
+ id: "qwen-3-coder-480b",
1341
+ name: "Qwen 3 Coder 480B",
1002
1342
  api: "openai-completions",
1003
1343
  provider: "cerebras",
1004
1344
  baseUrl: "https://api.cerebras.ai/v1",
1005
- reasoning: true,
1345
+ reasoning: false,
1006
1346
  input: ["text"],
1007
1347
  cost: {
1008
- input: 0.25,
1009
- output: 0.69,
1348
+ input: 2,
1349
+ output: 2,
1010
1350
  cacheRead: 0,
1011
1351
  cacheWrite: 0,
1012
1352
  },
1013
- contextWindow: 131072,
1014
- maxTokens: 32768,
1353
+ contextWindow: 131000,
1354
+ maxTokens: 32000,
1015
1355
  },
1016
- "qwen-3-coder-480b": {
1017
- id: "qwen-3-coder-480b",
1018
- name: "Qwen 3 Coder 480B",
1356
+ "gpt-oss-120b": {
1357
+ id: "gpt-oss-120b",
1358
+ name: "GPT OSS 120B",
1019
1359
  api: "openai-completions",
1020
1360
  provider: "cerebras",
1021
1361
  baseUrl: "https://api.cerebras.ai/v1",
1022
- reasoning: false,
1362
+ reasoning: true,
1023
1363
  input: ["text"],
1024
1364
  cost: {
1025
- input: 2,
1026
- output: 2,
1365
+ input: 0.25,
1366
+ output: 0.69,
1027
1367
  cacheRead: 0,
1028
1368
  cacheWrite: 0,
1029
1369
  },
1030
- contextWindow: 131000,
1031
- maxTokens: 32000,
1370
+ contextWindow: 131072,
1371
+ maxTokens: 32768,
1032
1372
  },
1033
1373
  },
1034
1374
  xai: {
1035
- "grok-3-mini-fast-latest": {
1036
- id: "grok-3-mini-fast-latest",
1037
- name: "Grok 3 Mini Fast Latest",
1375
+ "grok-4-fast-non-reasoning": {
1376
+ id: "grok-4-fast-non-reasoning",
1377
+ name: "Grok 4 Fast (Non-Reasoning)",
1038
1378
  api: "openai-completions",
1039
1379
  provider: "xai",
1040
1380
  baseUrl: "https://api.x.ai/v1",
1041
- reasoning: true,
1381
+ reasoning: false,
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-3-fast": {
1393
+ id: "grok-3-fast",
1394
+ name: "Grok 3 Fast",
1395
+ api: "openai-completions",
1396
+ provider: "xai",
1397
+ baseUrl: "https://api.x.ai/v1",
1398
+ reasoning: false,
1042
1399
  input: ["text"],
1043
1400
  cost: {
1044
- input: 0.6,
1045
- output: 4,
1046
- cacheRead: 0.15,
1401
+ input: 5,
1402
+ output: 25,
1403
+ cacheRead: 1.25,
1047
1404
  cacheWrite: 0,
1048
1405
  },
1049
1406
  contextWindow: 131072,
1050
1407
  maxTokens: 8192,
1051
1408
  },
1052
- "grok-3-mini-latest": {
1053
- id: "grok-3-mini-latest",
1054
- name: "Grok 3 Mini Latest",
1409
+ "grok-4": {
1410
+ id: "grok-4",
1411
+ name: "Grok 4",
1055
1412
  api: "openai-completions",
1056
1413
  provider: "xai",
1057
1414
  baseUrl: "https://api.x.ai/v1",
1058
1415
  reasoning: true,
1059
1416
  input: ["text"],
1060
1417
  cost: {
1061
- input: 0.3,
1062
- output: 0.5,
1063
- cacheRead: 0.075,
1418
+ input: 3,
1419
+ output: 15,
1420
+ cacheRead: 0.75,
1064
1421
  cacheWrite: 0,
1065
1422
  },
1066
- contextWindow: 131072,
1067
- maxTokens: 8192,
1423
+ contextWindow: 256000,
1424
+ maxTokens: 64000,
1068
1425
  },
1069
- "grok-beta": {
1070
- id: "grok-beta",
1071
- name: "Grok Beta",
1426
+ "grok-2-vision": {
1427
+ id: "grok-2-vision",
1428
+ name: "Grok 2 Vision",
1072
1429
  api: "openai-completions",
1073
1430
  provider: "xai",
1074
1431
  baseUrl: "https://api.x.ai/v1",
1075
1432
  reasoning: false,
1076
- input: ["text"],
1433
+ input: ["text", "image"],
1077
1434
  cost: {
1078
- input: 5,
1079
- output: 15,
1080
- cacheRead: 5,
1435
+ input: 2,
1436
+ output: 10,
1437
+ cacheRead: 2,
1081
1438
  cacheWrite: 0,
1082
1439
  },
1083
- contextWindow: 131072,
1440
+ contextWindow: 8192,
1084
1441
  maxTokens: 4096,
1085
1442
  },
1086
- "grok-3-fast-latest": {
1087
- id: "grok-3-fast-latest",
1088
- name: "Grok 3 Fast Latest",
1443
+ "grok-code-fast-1": {
1444
+ id: "grok-code-fast-1",
1445
+ name: "Grok Code Fast 1",
1089
1446
  api: "openai-completions",
1090
1447
  provider: "xai",
1091
1448
  baseUrl: "https://api.x.ai/v1",
1092
- reasoning: false,
1449
+ reasoning: true,
1093
1450
  input: ["text"],
1094
1451
  cost: {
1095
- input: 5,
1096
- output: 25,
1097
- cacheRead: 1.25,
1452
+ input: 0.2,
1453
+ output: 1.5,
1454
+ cacheRead: 0.02,
1098
1455
  cacheWrite: 0,
1099
1456
  },
1100
- contextWindow: 131072,
1101
- maxTokens: 8192,
1457
+ contextWindow: 256000,
1458
+ maxTokens: 10000,
1102
1459
  },
1103
- "grok-3": {
1104
- id: "grok-3",
1105
- name: "Grok 3",
1460
+ "grok-2": {
1461
+ id: "grok-2",
1462
+ name: "Grok 2",
1106
1463
  api: "openai-completions",
1107
1464
  provider: "xai",
1108
1465
  baseUrl: "https://api.x.ai/v1",
1109
1466
  reasoning: false,
1110
1467
  input: ["text"],
1111
1468
  cost: {
1112
- input: 3,
1113
- output: 15,
1114
- cacheRead: 0.75,
1469
+ input: 2,
1470
+ output: 10,
1471
+ cacheRead: 2,
1115
1472
  cacheWrite: 0,
1116
1473
  },
1117
1474
  contextWindow: 131072,
1118
1475
  maxTokens: 8192,
1119
1476
  },
1120
- "grok-3-mini": {
1121
- id: "grok-3-mini",
1122
- name: "Grok 3 Mini",
1477
+ "grok-3-mini-fast-latest": {
1478
+ id: "grok-3-mini-fast-latest",
1479
+ name: "Grok 3 Mini Fast Latest",
1123
1480
  api: "openai-completions",
1124
1481
  provider: "xai",
1125
1482
  baseUrl: "https://api.x.ai/v1",
1126
1483
  reasoning: true,
1127
1484
  input: ["text"],
1128
1485
  cost: {
1129
- input: 0.3,
1130
- output: 0.5,
1131
- cacheRead: 0.075,
1486
+ input: 0.6,
1487
+ output: 4,
1488
+ cacheRead: 0.15,
1132
1489
  cacheWrite: 0,
1133
1490
  },
1134
1491
  contextWindow: 131072,
@@ -1151,77 +1508,60 @@ export const MODELS = {
1151
1508
  contextWindow: 8192,
1152
1509
  maxTokens: 4096,
1153
1510
  },
1154
- "grok-2": {
1155
- id: "grok-2",
1156
- name: "Grok 2",
1511
+ "grok-3": {
1512
+ id: "grok-3",
1513
+ name: "Grok 3",
1157
1514
  api: "openai-completions",
1158
1515
  provider: "xai",
1159
1516
  baseUrl: "https://api.x.ai/v1",
1160
1517
  reasoning: false,
1161
1518
  input: ["text"],
1162
1519
  cost: {
1163
- input: 2,
1164
- output: 10,
1165
- cacheRead: 2,
1520
+ input: 3,
1521
+ output: 15,
1522
+ cacheRead: 0.75,
1166
1523
  cacheWrite: 0,
1167
1524
  },
1168
1525
  contextWindow: 131072,
1169
1526
  maxTokens: 8192,
1170
1527
  },
1171
- "grok-2-vision-latest": {
1172
- id: "grok-2-vision-latest",
1173
- name: "Grok 2 Vision Latest",
1528
+ "grok-4-fast": {
1529
+ id: "grok-4-fast",
1530
+ name: "Grok 4 Fast",
1174
1531
  api: "openai-completions",
1175
1532
  provider: "xai",
1176
1533
  baseUrl: "https://api.x.ai/v1",
1177
- reasoning: false,
1534
+ reasoning: true,
1178
1535
  input: ["text", "image"],
1179
1536
  cost: {
1180
- input: 2,
1181
- output: 10,
1182
- cacheRead: 2,
1537
+ input: 0.2,
1538
+ output: 0.5,
1539
+ cacheRead: 0.05,
1183
1540
  cacheWrite: 0,
1184
1541
  },
1185
- contextWindow: 8192,
1186
- maxTokens: 4096,
1542
+ contextWindow: 2000000,
1543
+ maxTokens: 30000,
1187
1544
  },
1188
- "grok-3-latest": {
1189
- id: "grok-3-latest",
1190
- name: "Grok 3 Latest",
1545
+ "grok-2-latest": {
1546
+ id: "grok-2-latest",
1547
+ name: "Grok 2 Latest",
1191
1548
  api: "openai-completions",
1192
1549
  provider: "xai",
1193
1550
  baseUrl: "https://api.x.ai/v1",
1194
1551
  reasoning: false,
1195
1552
  input: ["text"],
1196
- cost: {
1197
- input: 3,
1198
- output: 15,
1199
- cacheRead: 0.75,
1200
- cacheWrite: 0,
1201
- },
1202
- contextWindow: 131072,
1203
- maxTokens: 8192,
1204
- },
1205
- "grok-2-vision": {
1206
- id: "grok-2-vision",
1207
- name: "Grok 2 Vision",
1208
- api: "openai-completions",
1209
- provider: "xai",
1210
- baseUrl: "https://api.x.ai/v1",
1211
- reasoning: false,
1212
- input: ["text", "image"],
1213
1553
  cost: {
1214
1554
  input: 2,
1215
1555
  output: 10,
1216
1556
  cacheRead: 2,
1217
1557
  cacheWrite: 0,
1218
1558
  },
1219
- contextWindow: 8192,
1220
- maxTokens: 4096,
1559
+ contextWindow: 131072,
1560
+ maxTokens: 8192,
1221
1561
  },
1222
- "grok-2-latest": {
1223
- id: "grok-2-latest",
1224
- name: "Grok 2 Latest",
1562
+ "grok-2-1212": {
1563
+ id: "grok-2-1212",
1564
+ name: "Grok 2 (1212)",
1225
1565
  api: "openai-completions",
1226
1566
  provider: "xai",
1227
1567
  baseUrl: "https://api.x.ai/v1",
@@ -1236,9 +1576,9 @@ export const MODELS = {
1236
1576
  contextWindow: 131072,
1237
1577
  maxTokens: 8192,
1238
1578
  },
1239
- "grok-3-fast": {
1240
- id: "grok-3-fast",
1241
- name: "Grok 3 Fast",
1579
+ "grok-3-fast-latest": {
1580
+ id: "grok-3-fast-latest",
1581
+ name: "Grok 3 Fast Latest",
1242
1582
  api: "openai-completions",
1243
1583
  provider: "xai",
1244
1584
  baseUrl: "https://api.x.ai/v1",
@@ -1253,39 +1593,39 @@ export const MODELS = {
1253
1593
  contextWindow: 131072,
1254
1594
  maxTokens: 8192,
1255
1595
  },
1256
- "grok-2-1212": {
1257
- id: "grok-2-1212",
1258
- name: "Grok 2 (1212)",
1596
+ "grok-3-latest": {
1597
+ id: "grok-3-latest",
1598
+ name: "Grok 3 Latest",
1259
1599
  api: "openai-completions",
1260
1600
  provider: "xai",
1261
1601
  baseUrl: "https://api.x.ai/v1",
1262
1602
  reasoning: false,
1263
1603
  input: ["text"],
1264
1604
  cost: {
1265
- input: 2,
1266
- output: 10,
1267
- cacheRead: 2,
1605
+ input: 3,
1606
+ output: 15,
1607
+ cacheRead: 0.75,
1268
1608
  cacheWrite: 0,
1269
1609
  },
1270
1610
  contextWindow: 131072,
1271
1611
  maxTokens: 8192,
1272
1612
  },
1273
- "grok-4": {
1274
- id: "grok-4",
1275
- name: "Grok 4",
1613
+ "grok-2-vision-latest": {
1614
+ id: "grok-2-vision-latest",
1615
+ name: "Grok 2 Vision Latest",
1276
1616
  api: "openai-completions",
1277
1617
  provider: "xai",
1278
1618
  baseUrl: "https://api.x.ai/v1",
1279
- reasoning: true,
1280
- input: ["text"],
1619
+ reasoning: false,
1620
+ input: ["text", "image"],
1281
1621
  cost: {
1282
- input: 3,
1283
- output: 15,
1284
- cacheRead: 0.75,
1622
+ input: 2,
1623
+ output: 10,
1624
+ cacheRead: 2,
1285
1625
  cacheWrite: 0,
1286
1626
  },
1287
- contextWindow: 256000,
1288
- maxTokens: 64000,
1627
+ contextWindow: 8192,
1628
+ maxTokens: 4096,
1289
1629
  },
1290
1630
  "grok-vision-beta": {
1291
1631
  id: "grok-vision-beta",
@@ -1304,53 +1644,121 @@ export const MODELS = {
1304
1644
  contextWindow: 8192,
1305
1645
  maxTokens: 4096,
1306
1646
  },
1307
- "grok-3-mini-fast": {
1308
- id: "grok-3-mini-fast",
1309
- name: "Grok 3 Mini Fast",
1647
+ "grok-3-mini": {
1648
+ id: "grok-3-mini",
1649
+ name: "Grok 3 Mini",
1310
1650
  api: "openai-completions",
1311
1651
  provider: "xai",
1312
1652
  baseUrl: "https://api.x.ai/v1",
1313
1653
  reasoning: true,
1314
1654
  input: ["text"],
1315
1655
  cost: {
1316
- input: 0.6,
1317
- output: 4,
1318
- cacheRead: 0.15,
1656
+ input: 0.3,
1657
+ output: 0.5,
1658
+ cacheRead: 0.075,
1319
1659
  cacheWrite: 0,
1320
1660
  },
1321
1661
  contextWindow: 131072,
1322
1662
  maxTokens: 8192,
1323
1663
  },
1324
- "grok-code-fast-1": {
1325
- id: "grok-code-fast-1",
1326
- name: "Grok Code Fast 1",
1664
+ "grok-beta": {
1665
+ id: "grok-beta",
1666
+ name: "Grok Beta",
1327
1667
  api: "openai-completions",
1328
1668
  provider: "xai",
1329
1669
  baseUrl: "https://api.x.ai/v1",
1330
1670
  reasoning: false,
1331
1671
  input: ["text"],
1332
1672
  cost: {
1333
- input: 0.2,
1334
- output: 1.5,
1335
- cacheRead: 0.02,
1673
+ input: 5,
1674
+ output: 15,
1675
+ cacheRead: 5,
1336
1676
  cacheWrite: 0,
1337
1677
  },
1338
- contextWindow: 32768,
1339
- maxTokens: 8192,
1678
+ contextWindow: 131072,
1679
+ maxTokens: 4096,
1340
1680
  },
1341
- },
1342
- zai: {
1343
- "glm-4.5-air": {
1344
- id: "glm-4.5-air",
1345
- name: "GLM-4.5-Air",
1346
- api: "anthropic-messages",
1347
- provider: "zai",
1348
- baseUrl: "https://api.z.ai/api/anthropic",
1349
- reasoning: true,
1350
- input: ["text"],
1351
- cost: {
1352
- input: 0.2,
1353
- output: 1.1,
1681
+ "grok-3-mini-latest": {
1682
+ id: "grok-3-mini-latest",
1683
+ name: "Grok 3 Mini Latest",
1684
+ api: "openai-completions",
1685
+ provider: "xai",
1686
+ baseUrl: "https://api.x.ai/v1",
1687
+ reasoning: true,
1688
+ input: ["text"],
1689
+ cost: {
1690
+ input: 0.3,
1691
+ output: 0.5,
1692
+ cacheRead: 0.075,
1693
+ cacheWrite: 0,
1694
+ },
1695
+ contextWindow: 131072,
1696
+ maxTokens: 8192,
1697
+ },
1698
+ "grok-3-mini-fast": {
1699
+ id: "grok-3-mini-fast",
1700
+ name: "Grok 3 Mini Fast",
1701
+ api: "openai-completions",
1702
+ provider: "xai",
1703
+ baseUrl: "https://api.x.ai/v1",
1704
+ reasoning: true,
1705
+ input: ["text"],
1706
+ cost: {
1707
+ input: 0.6,
1708
+ output: 4,
1709
+ cacheRead: 0.15,
1710
+ cacheWrite: 0,
1711
+ },
1712
+ contextWindow: 131072,
1713
+ maxTokens: 8192,
1714
+ },
1715
+ },
1716
+ zai: {
1717
+ "glm-4.5-flash": {
1718
+ id: "glm-4.5-flash",
1719
+ name: "GLM-4.5-Flash",
1720
+ api: "anthropic-messages",
1721
+ provider: "zai",
1722
+ baseUrl: "https://api.z.ai/api/anthropic",
1723
+ reasoning: true,
1724
+ input: ["text"],
1725
+ cost: {
1726
+ input: 0,
1727
+ output: 0,
1728
+ cacheRead: 0,
1729
+ cacheWrite: 0,
1730
+ },
1731
+ contextWindow: 131072,
1732
+ maxTokens: 98304,
1733
+ },
1734
+ "glm-4.5": {
1735
+ id: "glm-4.5",
1736
+ name: "GLM-4.5",
1737
+ api: "anthropic-messages",
1738
+ provider: "zai",
1739
+ baseUrl: "https://api.z.ai/api/anthropic",
1740
+ reasoning: true,
1741
+ input: ["text"],
1742
+ cost: {
1743
+ input: 0.6,
1744
+ output: 2.2,
1745
+ cacheRead: 0.11,
1746
+ cacheWrite: 0,
1747
+ },
1748
+ contextWindow: 131072,
1749
+ maxTokens: 98304,
1750
+ },
1751
+ "glm-4.5-air": {
1752
+ id: "glm-4.5-air",
1753
+ name: "GLM-4.5-Air",
1754
+ api: "anthropic-messages",
1755
+ provider: "zai",
1756
+ baseUrl: "https://api.z.ai/api/anthropic",
1757
+ reasoning: true,
1758
+ input: ["text"],
1759
+ cost: {
1760
+ input: 0.2,
1761
+ output: 1.1,
1354
1762
  cacheRead: 0.03,
1355
1763
  cacheWrite: 0,
1356
1764
  },
@@ -1366,50 +1774,322 @@ export const MODELS = {
1366
1774
  reasoning: true,
1367
1775
  input: ["text", "image"],
1368
1776
  cost: {
1369
- input: 0.6,
1370
- output: 1.8,
1371
- cacheRead: 0,
1777
+ input: 0.6,
1778
+ output: 1.8,
1779
+ cacheRead: 0,
1780
+ cacheWrite: 0,
1781
+ },
1782
+ contextWindow: 64000,
1783
+ maxTokens: 16384,
1784
+ },
1785
+ "glm-4.6": {
1786
+ id: "glm-4.6",
1787
+ name: "GLM-4.6",
1788
+ api: "anthropic-messages",
1789
+ provider: "zai",
1790
+ baseUrl: "https://api.z.ai/api/anthropic",
1791
+ reasoning: true,
1792
+ input: ["text"],
1793
+ cost: {
1794
+ input: 0.6,
1795
+ output: 2.2,
1796
+ cacheRead: 0.11,
1797
+ cacheWrite: 0,
1798
+ },
1799
+ contextWindow: 204800,
1800
+ maxTokens: 131072,
1801
+ },
1802
+ },
1803
+ openrouter: {
1804
+ "openrouter/andromeda-alpha": {
1805
+ id: "openrouter/andromeda-alpha",
1806
+ name: "Andromeda Alpha",
1807
+ api: "openai-completions",
1808
+ provider: "openrouter",
1809
+ baseUrl: "https://openrouter.ai/api/v1",
1810
+ reasoning: true,
1811
+ input: ["text", "image"],
1812
+ cost: {
1813
+ input: 0,
1814
+ output: 0,
1815
+ cacheRead: 0,
1816
+ cacheWrite: 0,
1817
+ },
1818
+ contextWindow: 128000,
1819
+ maxTokens: 128000,
1820
+ },
1821
+ "deepcogito/cogito-v2-preview-llama-405b": {
1822
+ id: "deepcogito/cogito-v2-preview-llama-405b",
1823
+ name: "Deep Cogito: Cogito V2 Preview Llama 405B",
1824
+ api: "openai-completions",
1825
+ provider: "openrouter",
1826
+ baseUrl: "https://openrouter.ai/api/v1",
1827
+ reasoning: true,
1828
+ input: ["text"],
1829
+ cost: {
1830
+ input: 3.5,
1831
+ output: 3.5,
1832
+ cacheRead: 0,
1833
+ cacheWrite: 0,
1834
+ },
1835
+ contextWindow: 32768,
1836
+ maxTokens: 4096,
1837
+ },
1838
+ "qwen/qwen3-vl-8b-thinking": {
1839
+ id: "qwen/qwen3-vl-8b-thinking",
1840
+ name: "Qwen: Qwen3 VL 8B Thinking",
1841
+ api: "openai-completions",
1842
+ provider: "openrouter",
1843
+ baseUrl: "https://openrouter.ai/api/v1",
1844
+ reasoning: true,
1845
+ input: ["text", "image"],
1846
+ cost: {
1847
+ input: 0.18,
1848
+ output: 2.0999999999999996,
1849
+ cacheRead: 0,
1850
+ cacheWrite: 0,
1851
+ },
1852
+ contextWindow: 256000,
1853
+ maxTokens: 32768,
1854
+ },
1855
+ "qwen/qwen3-vl-8b-instruct": {
1856
+ id: "qwen/qwen3-vl-8b-instruct",
1857
+ name: "Qwen: Qwen3 VL 8B Instruct",
1858
+ api: "openai-completions",
1859
+ provider: "openrouter",
1860
+ baseUrl: "https://openrouter.ai/api/v1",
1861
+ reasoning: false,
1862
+ input: ["text", "image"],
1863
+ cost: {
1864
+ input: 0.08,
1865
+ output: 0.5,
1866
+ cacheRead: 0,
1867
+ cacheWrite: 0,
1868
+ },
1869
+ contextWindow: 131072,
1870
+ maxTokens: 32768,
1871
+ },
1872
+ "inclusionai/ring-1t": {
1873
+ id: "inclusionai/ring-1t",
1874
+ name: "inclusionAI: Ring 1T",
1875
+ api: "openai-completions",
1876
+ provider: "openrouter",
1877
+ baseUrl: "https://openrouter.ai/api/v1",
1878
+ reasoning: true,
1879
+ input: ["text"],
1880
+ cost: {
1881
+ input: 0.5700000000000001,
1882
+ output: 2.2800000000000002,
1883
+ cacheRead: 0,
1884
+ cacheWrite: 0,
1885
+ },
1886
+ contextWindow: 131072,
1887
+ maxTokens: 131072,
1888
+ },
1889
+ "inclusionai/ling-1t": {
1890
+ id: "inclusionai/ling-1t",
1891
+ name: "inclusionAI: Ling-1T",
1892
+ api: "openai-completions",
1893
+ provider: "openrouter",
1894
+ baseUrl: "https://openrouter.ai/api/v1",
1895
+ reasoning: false,
1896
+ input: ["text"],
1897
+ cost: {
1898
+ input: 0.39999999999999997,
1899
+ output: 2,
1900
+ cacheRead: 0,
1901
+ cacheWrite: 0,
1902
+ },
1903
+ contextWindow: 131072,
1904
+ maxTokens: 131072,
1905
+ },
1906
+ "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
1907
+ id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
1908
+ name: "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5",
1909
+ api: "openai-completions",
1910
+ provider: "openrouter",
1911
+ baseUrl: "https://openrouter.ai/api/v1",
1912
+ reasoning: true,
1913
+ input: ["text"],
1914
+ cost: {
1915
+ input: 0.09999999999999999,
1916
+ output: 0.39999999999999997,
1917
+ cacheRead: 0,
1918
+ cacheWrite: 0,
1919
+ },
1920
+ contextWindow: 131072,
1921
+ maxTokens: 4096,
1922
+ },
1923
+ "qwen/qwen3-vl-30b-a3b-thinking": {
1924
+ id: "qwen/qwen3-vl-30b-a3b-thinking",
1925
+ name: "Qwen: Qwen3 VL 30B A3B Thinking",
1926
+ api: "openai-completions",
1927
+ provider: "openrouter",
1928
+ baseUrl: "https://openrouter.ai/api/v1",
1929
+ reasoning: true,
1930
+ input: ["text", "image"],
1931
+ cost: {
1932
+ input: 0.19999999999999998,
1933
+ output: 1,
1934
+ cacheRead: 0,
1935
+ cacheWrite: 0,
1936
+ },
1937
+ contextWindow: 131072,
1938
+ maxTokens: 32768,
1939
+ },
1940
+ "qwen/qwen3-vl-30b-a3b-instruct": {
1941
+ id: "qwen/qwen3-vl-30b-a3b-instruct",
1942
+ name: "Qwen: Qwen3 VL 30B A3B Instruct",
1943
+ api: "openai-completions",
1944
+ provider: "openrouter",
1945
+ baseUrl: "https://openrouter.ai/api/v1",
1946
+ reasoning: false,
1947
+ input: ["text", "image"],
1948
+ cost: {
1949
+ input: 0.19999999999999998,
1950
+ output: 0.7,
1951
+ cacheRead: 0,
1952
+ cacheWrite: 0,
1953
+ },
1954
+ contextWindow: 131072,
1955
+ maxTokens: 32768,
1956
+ },
1957
+ "z-ai/glm-4.6": {
1958
+ id: "z-ai/glm-4.6",
1959
+ name: "Z.AI: GLM 4.6",
1960
+ api: "openai-completions",
1961
+ provider: "openrouter",
1962
+ baseUrl: "https://openrouter.ai/api/v1",
1963
+ reasoning: true,
1964
+ input: ["text"],
1965
+ cost: {
1966
+ input: 0.5,
1967
+ output: 1.75,
1968
+ cacheRead: 0,
1969
+ cacheWrite: 0,
1970
+ },
1971
+ contextWindow: 202752,
1972
+ maxTokens: 202752,
1973
+ },
1974
+ "deepseek/deepseek-v3.2-exp": {
1975
+ id: "deepseek/deepseek-v3.2-exp",
1976
+ name: "DeepSeek: DeepSeek V3.2 Exp",
1977
+ api: "openai-completions",
1978
+ provider: "openrouter",
1979
+ baseUrl: "https://openrouter.ai/api/v1",
1980
+ reasoning: true,
1981
+ input: ["text"],
1982
+ cost: {
1983
+ input: 0.27,
1984
+ output: 0.39999999999999997,
1985
+ cacheRead: 0,
1986
+ cacheWrite: 0,
1987
+ },
1988
+ contextWindow: 163840,
1989
+ maxTokens: 4096,
1990
+ },
1991
+ "qwen/qwen3-vl-235b-a22b-thinking": {
1992
+ id: "qwen/qwen3-vl-235b-a22b-thinking",
1993
+ name: "Qwen: Qwen3 VL 235B A22B Thinking",
1994
+ api: "openai-completions",
1995
+ provider: "openrouter",
1996
+ baseUrl: "https://openrouter.ai/api/v1",
1997
+ reasoning: true,
1998
+ input: ["text", "image"],
1999
+ cost: {
2000
+ input: 0.3,
2001
+ output: 1.2,
2002
+ cacheRead: 0,
2003
+ cacheWrite: 0,
2004
+ },
2005
+ contextWindow: 262144,
2006
+ maxTokens: 262144,
2007
+ },
2008
+ "qwen/qwen3-vl-235b-a22b-instruct": {
2009
+ id: "qwen/qwen3-vl-235b-a22b-instruct",
2010
+ name: "Qwen: Qwen3 VL 235B A22B Instruct",
2011
+ api: "openai-completions",
2012
+ provider: "openrouter",
2013
+ baseUrl: "https://openrouter.ai/api/v1",
2014
+ reasoning: true,
2015
+ input: ["text", "image"],
2016
+ cost: {
2017
+ input: 0.3,
2018
+ output: 1.2,
2019
+ cacheRead: 0,
2020
+ cacheWrite: 0,
2021
+ },
2022
+ contextWindow: 262144,
2023
+ maxTokens: 262144,
2024
+ },
2025
+ "qwen/qwen3-max": {
2026
+ id: "qwen/qwen3-max",
2027
+ name: "Qwen: Qwen3 Max",
2028
+ api: "openai-completions",
2029
+ provider: "openrouter",
2030
+ baseUrl: "https://openrouter.ai/api/v1",
2031
+ reasoning: false,
2032
+ input: ["text"],
2033
+ cost: {
2034
+ input: 1.2,
2035
+ output: 6,
2036
+ cacheRead: 0.24,
2037
+ cacheWrite: 0,
2038
+ },
2039
+ contextWindow: 256000,
2040
+ maxTokens: 32768,
2041
+ },
2042
+ "qwen/qwen3-coder-plus": {
2043
+ id: "qwen/qwen3-coder-plus",
2044
+ name: "Qwen: Qwen3 Coder Plus",
2045
+ api: "openai-completions",
2046
+ provider: "openrouter",
2047
+ baseUrl: "https://openrouter.ai/api/v1",
2048
+ reasoning: false,
2049
+ input: ["text"],
2050
+ cost: {
2051
+ input: 1,
2052
+ output: 5,
2053
+ cacheRead: 0.09999999999999999,
1372
2054
  cacheWrite: 0,
1373
2055
  },
1374
- contextWindow: 64000,
1375
- maxTokens: 16384,
2056
+ contextWindow: 128000,
2057
+ maxTokens: 65536,
1376
2058
  },
1377
- "glm-4.5-flash": {
1378
- id: "glm-4.5-flash",
1379
- name: "GLM-4.5-Flash",
1380
- api: "anthropic-messages",
1381
- provider: "zai",
1382
- baseUrl: "https://api.z.ai/api/anthropic",
2059
+ "deepseek/deepseek-v3.1-terminus": {
2060
+ id: "deepseek/deepseek-v3.1-terminus",
2061
+ name: "DeepSeek: DeepSeek V3.1 Terminus",
2062
+ api: "openai-completions",
2063
+ provider: "openrouter",
2064
+ baseUrl: "https://openrouter.ai/api/v1",
1383
2065
  reasoning: true,
1384
2066
  input: ["text"],
1385
2067
  cost: {
1386
- input: 0,
1387
- output: 0,
2068
+ input: 0.22999999999999998,
2069
+ output: 0.8999999999999999,
1388
2070
  cacheRead: 0,
1389
2071
  cacheWrite: 0,
1390
2072
  },
1391
- contextWindow: 131072,
1392
- maxTokens: 98304,
2073
+ contextWindow: 163840,
2074
+ maxTokens: 163840,
1393
2075
  },
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",
2076
+ "alibaba/tongyi-deepresearch-30b-a3b:free": {
2077
+ id: "alibaba/tongyi-deepresearch-30b-a3b:free",
2078
+ name: "Tongyi DeepResearch 30B A3B (free)",
2079
+ api: "openai-completions",
2080
+ provider: "openrouter",
2081
+ baseUrl: "https://openrouter.ai/api/v1",
1400
2082
  reasoning: true,
1401
2083
  input: ["text"],
1402
2084
  cost: {
1403
- input: 0.6,
1404
- output: 2.2,
1405
- cacheRead: 0.11,
2085
+ input: 0,
2086
+ output: 0,
2087
+ cacheRead: 0,
1406
2088
  cacheWrite: 0,
1407
2089
  },
1408
2090
  contextWindow: 131072,
1409
- maxTokens: 98304,
2091
+ maxTokens: 131072,
1410
2092
  },
1411
- },
1412
- openrouter: {
1413
2093
  "alibaba/tongyi-deepresearch-30b-a3b": {
1414
2094
  id: "alibaba/tongyi-deepresearch-30b-a3b",
1415
2095
  name: "Tongyi DeepResearch 30B A3B",
@@ -1420,7 +2100,7 @@ export const MODELS = {
1420
2100
  input: ["text"],
1421
2101
  cost: {
1422
2102
  input: 0.09,
1423
- output: 0.44999999999999996,
2103
+ output: 0.39999999999999997,
1424
2104
  cacheRead: 0,
1425
2105
  cacheWrite: 0,
1426
2106
  },
@@ -1444,23 +2124,6 @@ export const MODELS = {
1444
2124
  contextWindow: 128000,
1445
2125
  maxTokens: 65536,
1446
2126
  },
1447
- "qwen/qwen3-coder-plus": {
1448
- id: "qwen/qwen3-coder-plus",
1449
- name: "Qwen: Qwen3 Coder Plus",
1450
- api: "openai-completions",
1451
- provider: "openrouter",
1452
- baseUrl: "https://openrouter.ai/api/v1",
1453
- reasoning: false,
1454
- input: ["text"],
1455
- cost: {
1456
- input: 1,
1457
- output: 5,
1458
- cacheRead: 0.09999999999999999,
1459
- cacheWrite: 0,
1460
- },
1461
- contextWindow: 128000,
1462
- maxTokens: 65536,
1463
- },
1464
2127
  "qwen/qwen3-next-80b-a3b-thinking": {
1465
2128
  id: "qwen/qwen3-next-80b-a3b-thinking",
1466
2129
  name: "Qwen: Qwen3 Next 80B A3B Thinking",
@@ -1470,8 +2133,8 @@ export const MODELS = {
1470
2133
  reasoning: true,
1471
2134
  input: ["text"],
1472
2135
  cost: {
1473
- input: 0.09999999999999999,
1474
- output: 0.7999999999999999,
2136
+ input: 0.14,
2137
+ output: 1.2,
1475
2138
  cacheRead: 0,
1476
2139
  cacheWrite: 0,
1477
2140
  },
@@ -1493,24 +2156,24 @@ export const MODELS = {
1493
2156
  cacheWrite: 0,
1494
2157
  },
1495
2158
  contextWindow: 262144,
1496
- maxTokens: 4096,
2159
+ maxTokens: 262144,
1497
2160
  },
1498
- "meituan/longcat-flash-chat": {
1499
- id: "meituan/longcat-flash-chat",
1500
- name: "Meituan: LongCat Flash Chat",
2161
+ "meituan/longcat-flash-chat:free": {
2162
+ id: "meituan/longcat-flash-chat:free",
2163
+ name: "Meituan: LongCat Flash Chat (free)",
1501
2164
  api: "openai-completions",
1502
2165
  provider: "openrouter",
1503
2166
  baseUrl: "https://openrouter.ai/api/v1",
1504
2167
  reasoning: false,
1505
2168
  input: ["text"],
1506
2169
  cost: {
1507
- input: 0.12,
1508
- output: 0.6,
2170
+ input: 0,
2171
+ output: 0,
1509
2172
  cacheRead: 0,
1510
2173
  cacheWrite: 0,
1511
2174
  },
1512
2175
  contextWindow: 131072,
1513
- maxTokens: 4096,
2176
+ maxTokens: 131072,
1514
2177
  },
1515
2178
  "qwen/qwen-plus-2025-07-28": {
1516
2179
  id: "qwen/qwen-plus-2025-07-28",
@@ -1580,72 +2243,38 @@ export const MODELS = {
1580
2243
  contextWindow: 131072,
1581
2244
  maxTokens: 4096,
1582
2245
  },
1583
- "openrouter/sonoma-dusk-alpha": {
1584
- id: "openrouter/sonoma-dusk-alpha",
1585
- name: "Sonoma Dusk Alpha",
2246
+ "moonshotai/kimi-k2-0905": {
2247
+ id: "moonshotai/kimi-k2-0905",
2248
+ name: "MoonshotAI: Kimi K2 0905",
1586
2249
  api: "openai-completions",
1587
2250
  provider: "openrouter",
1588
2251
  baseUrl: "https://openrouter.ai/api/v1",
1589
2252
  reasoning: false,
1590
- input: ["text", "image"],
2253
+ input: ["text"],
1591
2254
  cost: {
1592
- input: 0,
1593
- output: 0,
2255
+ input: 0.39,
2256
+ output: 1.9,
1594
2257
  cacheRead: 0,
1595
2258
  cacheWrite: 0,
1596
2259
  },
1597
- contextWindow: 2000000,
1598
- maxTokens: 4096,
2260
+ contextWindow: 262144,
2261
+ maxTokens: 262144,
1599
2262
  },
1600
- "openrouter/sonoma-sky-alpha": {
1601
- id: "openrouter/sonoma-sky-alpha",
1602
- name: "Sonoma Sky Alpha",
2263
+ "deepcogito/cogito-v2-preview-llama-70b": {
2264
+ id: "deepcogito/cogito-v2-preview-llama-70b",
2265
+ name: "Deep Cogito: Cogito V2 Preview Llama 70B",
1603
2266
  api: "openai-completions",
1604
2267
  provider: "openrouter",
1605
2268
  baseUrl: "https://openrouter.ai/api/v1",
1606
2269
  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
- "moonshotai/kimi-k2-0905": {
1635
- id: "moonshotai/kimi-k2-0905",
1636
- name: "MoonshotAI: Kimi K2 0905",
1637
- api: "openai-completions",
1638
- provider: "openrouter",
1639
- baseUrl: "https://openrouter.ai/api/v1",
1640
- reasoning: false,
1641
2270
  input: ["text"],
1642
2271
  cost: {
1643
- input: 0.38,
1644
- output: 1.52,
2272
+ input: 0.88,
2273
+ output: 0.88,
1645
2274
  cacheRead: 0,
1646
2275
  cacheWrite: 0,
1647
2276
  },
1648
- contextWindow: 262144,
2277
+ contextWindow: 32768,
1649
2278
  maxTokens: 4096,
1650
2279
  },
1651
2280
  "deepcogito/cogito-v2-preview-llama-109b-moe": {
@@ -1714,7 +2343,7 @@ export const MODELS = {
1714
2343
  cacheWrite: 0,
1715
2344
  },
1716
2345
  contextWindow: 131072,
1717
- maxTokens: 4096,
2346
+ maxTokens: 131072,
1718
2347
  },
1719
2348
  "nousresearch/hermes-4-405b": {
1720
2349
  id: "nousresearch/hermes-4-405b",
@@ -1725,30 +2354,13 @@ export const MODELS = {
1725
2354
  reasoning: true,
1726
2355
  input: ["text"],
1727
2356
  cost: {
1728
- input: 0.24999987999999998,
1729
- output: 0.999999888,
2357
+ input: 0.3,
2358
+ output: 1.2,
1730
2359
  cacheRead: 0,
1731
2360
  cacheWrite: 0,
1732
2361
  },
1733
2362
  contextWindow: 131072,
1734
- maxTokens: 4096,
1735
- },
1736
- "deepseek/deepseek-chat-v3.1:free": {
1737
- id: "deepseek/deepseek-chat-v3.1:free",
1738
- name: "DeepSeek: DeepSeek V3.1 (free)",
1739
- api: "openai-completions",
1740
- provider: "openrouter",
1741
- baseUrl: "https://openrouter.ai/api/v1",
1742
- reasoning: true,
1743
- input: ["text"],
1744
- cost: {
1745
- input: 0,
1746
- output: 0,
1747
- cacheRead: 0,
1748
- cacheWrite: 0,
1749
- },
1750
- contextWindow: 163840,
1751
- maxTokens: 4096,
2363
+ maxTokens: 131072,
1752
2364
  },
1753
2365
  "deepseek/deepseek-chat-v3.1": {
1754
2366
  id: "deepseek/deepseek-chat-v3.1",
@@ -1759,13 +2371,13 @@ export const MODELS = {
1759
2371
  reasoning: true,
1760
2372
  input: ["text"],
1761
2373
  cost: {
1762
- input: 0.24999987999999998,
1763
- output: 0.999999888,
2374
+ input: 0.27,
2375
+ output: 1,
1764
2376
  cacheRead: 0,
1765
2377
  cacheWrite: 0,
1766
2378
  },
1767
- contextWindow: 163840,
1768
- maxTokens: 4096,
2379
+ contextWindow: 131072,
2380
+ maxTokens: 32768,
1769
2381
  },
1770
2382
  "mistralai/mistral-medium-3.1": {
1771
2383
  id: "mistralai/mistral-medium-3.1",
@@ -1784,6 +2396,40 @@ export const MODELS = {
1784
2396
  contextWindow: 131072,
1785
2397
  maxTokens: 4096,
1786
2398
  },
2399
+ "baidu/ernie-4.5-21b-a3b": {
2400
+ id: "baidu/ernie-4.5-21b-a3b",
2401
+ name: "Baidu: ERNIE 4.5 21B A3B",
2402
+ api: "openai-completions",
2403
+ provider: "openrouter",
2404
+ baseUrl: "https://openrouter.ai/api/v1",
2405
+ reasoning: false,
2406
+ input: ["text"],
2407
+ cost: {
2408
+ input: 0.07,
2409
+ output: 0.28,
2410
+ cacheRead: 0,
2411
+ cacheWrite: 0,
2412
+ },
2413
+ contextWindow: 120000,
2414
+ maxTokens: 8000,
2415
+ },
2416
+ "baidu/ernie-4.5-vl-28b-a3b": {
2417
+ id: "baidu/ernie-4.5-vl-28b-a3b",
2418
+ name: "Baidu: ERNIE 4.5 VL 28B A3B",
2419
+ api: "openai-completions",
2420
+ provider: "openrouter",
2421
+ baseUrl: "https://openrouter.ai/api/v1",
2422
+ reasoning: true,
2423
+ input: ["text", "image"],
2424
+ cost: {
2425
+ input: 0.14,
2426
+ output: 0.56,
2427
+ cacheRead: 0,
2428
+ cacheWrite: 0,
2429
+ },
2430
+ contextWindow: 30000,
2431
+ maxTokens: 8000,
2432
+ },
1787
2433
  "z-ai/glm-4.5v": {
1788
2434
  id: "z-ai/glm-4.5v",
1789
2435
  name: "Z.AI: GLM 4.5V",
@@ -1793,13 +2439,13 @@ export const MODELS = {
1793
2439
  reasoning: true,
1794
2440
  input: ["text", "image"],
1795
2441
  cost: {
1796
- input: 0.5,
2442
+ input: 0.6,
1797
2443
  output: 1.7999999999999998,
1798
- cacheRead: 0,
2444
+ cacheRead: 0.11,
1799
2445
  cacheWrite: 0,
1800
2446
  },
1801
2447
  contextWindow: 65536,
1802
- maxTokens: 65536,
2448
+ maxTokens: 16384,
1803
2449
  },
1804
2450
  "ai21/jamba-mini-1.7": {
1805
2451
  id: "ai21/jamba-mini-1.7",
@@ -1861,13 +2507,13 @@ export const MODELS = {
1861
2507
  reasoning: false,
1862
2508
  input: ["text"],
1863
2509
  cost: {
1864
- input: 0.07,
1865
- output: 0.28,
2510
+ input: 0.06,
2511
+ output: 0.25,
1866
2512
  cacheRead: 0,
1867
2513
  cacheWrite: 0,
1868
2514
  },
1869
2515
  contextWindow: 262144,
1870
- maxTokens: 4096,
2516
+ maxTokens: 262144,
1871
2517
  },
1872
2518
  "qwen/qwen3-30b-a3b-instruct-2507": {
1873
2519
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -1878,13 +2524,13 @@ export const MODELS = {
1878
2524
  reasoning: false,
1879
2525
  input: ["text"],
1880
2526
  cost: {
1881
- input: 0.07,
1882
- output: 0.28,
2527
+ input: 0.08,
2528
+ output: 0.33,
1883
2529
  cacheRead: 0,
1884
2530
  cacheWrite: 0,
1885
2531
  },
1886
2532
  contextWindow: 262144,
1887
- maxTokens: 4096,
2533
+ maxTokens: 262144,
1888
2534
  },
1889
2535
  "z-ai/glm-4.5": {
1890
2536
  id: "z-ai/glm-4.5",
@@ -1895,13 +2541,13 @@ export const MODELS = {
1895
2541
  reasoning: true,
1896
2542
  input: ["text"],
1897
2543
  cost: {
1898
- input: 0.41,
1899
- output: 1.6500000000000001,
2544
+ input: 0.35,
2545
+ output: 1.5,
1900
2546
  cacheRead: 0,
1901
2547
  cacheWrite: 0,
1902
2548
  },
1903
2549
  contextWindow: 131072,
1904
- maxTokens: 4096,
2550
+ maxTokens: 131072,
1905
2551
  },
1906
2552
  "z-ai/glm-4.5-air:free": {
1907
2553
  id: "z-ai/glm-4.5-air:free",
@@ -1918,7 +2564,7 @@ export const MODELS = {
1918
2564
  cacheWrite: 0,
1919
2565
  },
1920
2566
  contextWindow: 131072,
1921
- maxTokens: 4096,
2567
+ maxTokens: 131072,
1922
2568
  },
1923
2569
  "z-ai/glm-4.5-air": {
1924
2570
  id: "z-ai/glm-4.5-air",
@@ -1929,13 +2575,13 @@ export const MODELS = {
1929
2575
  reasoning: true,
1930
2576
  input: ["text"],
1931
2577
  cost: {
1932
- input: 0.14,
1933
- output: 0.86,
2578
+ input: 0.13,
2579
+ output: 0.85,
1934
2580
  cacheRead: 0,
1935
2581
  cacheWrite: 0,
1936
2582
  },
1937
2583
  contextWindow: 131072,
1938
- maxTokens: 131072,
2584
+ maxTokens: 98304,
1939
2585
  },
1940
2586
  "qwen/qwen3-235b-a22b-thinking-2507": {
1941
2587
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -1946,13 +2592,13 @@ export const MODELS = {
1946
2592
  reasoning: true,
1947
2593
  input: ["text"],
1948
2594
  cost: {
1949
- input: 0.09999999999999999,
1950
- output: 0.39,
2595
+ input: 0.11,
2596
+ output: 0.6,
1951
2597
  cacheRead: 0,
1952
2598
  cacheWrite: 0,
1953
2599
  },
1954
2600
  contextWindow: 262144,
1955
- maxTokens: 4096,
2601
+ maxTokens: 262144,
1956
2602
  },
1957
2603
  "z-ai/glm-4-32b": {
1958
2604
  id: "z-ai/glm-4-32b",
@@ -1985,8 +2631,8 @@ export const MODELS = {
1985
2631
  cacheRead: 0,
1986
2632
  cacheWrite: 0,
1987
2633
  },
1988
- contextWindow: 262144,
1989
- maxTokens: 4096,
2634
+ contextWindow: 262000,
2635
+ maxTokens: 262000,
1990
2636
  },
1991
2637
  "qwen/qwen3-coder": {
1992
2638
  id: "qwen/qwen3-coder",
@@ -1994,7 +2640,7 @@ export const MODELS = {
1994
2640
  api: "openai-completions",
1995
2641
  provider: "openrouter",
1996
2642
  baseUrl: "https://openrouter.ai/api/v1",
1997
- reasoning: false,
2643
+ reasoning: true,
1998
2644
  input: ["text"],
1999
2645
  cost: {
2000
2646
  input: 0.22,
@@ -2003,7 +2649,7 @@ export const MODELS = {
2003
2649
  cacheWrite: 0,
2004
2650
  },
2005
2651
  contextWindow: 262144,
2006
- maxTokens: 4096,
2652
+ maxTokens: 262144,
2007
2653
  },
2008
2654
  "qwen/qwen3-235b-a22b-2507": {
2009
2655
  id: "qwen/qwen3-235b-a22b-2507",
@@ -2011,34 +2657,17 @@ export const MODELS = {
2011
2657
  api: "openai-completions",
2012
2658
  provider: "openrouter",
2013
2659
  baseUrl: "https://openrouter.ai/api/v1",
2014
- reasoning: false,
2660
+ reasoning: true,
2015
2661
  input: ["text"],
2016
2662
  cost: {
2017
- input: 0.09999999999999999,
2018
- output: 0.09999999999999999,
2663
+ input: 0.08,
2664
+ output: 0.55,
2019
2665
  cacheRead: 0,
2020
2666
  cacheWrite: 0,
2021
2667
  },
2022
2668
  contextWindow: 262144,
2023
2669
  maxTokens: 262144,
2024
2670
  },
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
2671
  "moonshotai/kimi-k2": {
2043
2672
  id: "moonshotai/kimi-k2",
2044
2673
  name: "MoonshotAI: Kimi K2 0711",
@@ -2090,6 +2719,23 @@ export const MODELS = {
2090
2719
  contextWindow: 128000,
2091
2720
  maxTokens: 4096,
2092
2721
  },
2722
+ "tngtech/deepseek-r1t2-chimera": {
2723
+ id: "tngtech/deepseek-r1t2-chimera",
2724
+ name: "TNG: DeepSeek R1T2 Chimera",
2725
+ api: "openai-completions",
2726
+ provider: "openrouter",
2727
+ baseUrl: "https://openrouter.ai/api/v1",
2728
+ reasoning: true,
2729
+ input: ["text"],
2730
+ cost: {
2731
+ input: 0.3,
2732
+ output: 1.2,
2733
+ cacheRead: 0,
2734
+ cacheWrite: 0,
2735
+ },
2736
+ contextWindow: 163840,
2737
+ maxTokens: 163840,
2738
+ },
2093
2739
  "inception/mercury": {
2094
2740
  id: "inception/mercury",
2095
2741
  name: "Inception: Mercury",
@@ -2133,13 +2779,13 @@ export const MODELS = {
2133
2779
  reasoning: false,
2134
2780
  input: ["text", "image"],
2135
2781
  cost: {
2136
- input: 0.075,
2137
- output: 0.19999999999999998,
2782
+ input: 0.06,
2783
+ output: 0.18,
2138
2784
  cacheRead: 0,
2139
2785
  cacheWrite: 0,
2140
2786
  },
2141
- contextWindow: 128000,
2142
- maxTokens: 4096,
2787
+ contextWindow: 131072,
2788
+ maxTokens: 131072,
2143
2789
  },
2144
2790
  "minimax/minimax-m1": {
2145
2791
  id: "minimax/minimax-m1",
@@ -2150,8 +2796,8 @@ export const MODELS = {
2150
2796
  reasoning: true,
2151
2797
  input: ["text"],
2152
2798
  cost: {
2153
- input: 0.3,
2154
- output: 1.6500000000000001,
2799
+ input: 0.39999999999999997,
2800
+ output: 2.2,
2155
2801
  cacheRead: 0,
2156
2802
  cacheWrite: 0,
2157
2803
  },
@@ -2224,7 +2870,7 @@ export const MODELS = {
2224
2870
  cacheWrite: 0,
2225
2871
  },
2226
2872
  contextWindow: 163840,
2227
- maxTokens: 4096,
2873
+ maxTokens: 163840,
2228
2874
  },
2229
2875
  "mistralai/devstral-small-2505:free": {
2230
2876
  id: "mistralai/devstral-small-2505:free",
@@ -2252,13 +2898,13 @@ export const MODELS = {
2252
2898
  reasoning: false,
2253
2899
  input: ["text"],
2254
2900
  cost: {
2255
- input: 0.04,
2256
- output: 0.14,
2901
+ input: 0.049999999999999996,
2902
+ output: 0.22,
2257
2903
  cacheRead: 0,
2258
2904
  cacheWrite: 0,
2259
2905
  },
2260
2906
  contextWindow: 131072,
2261
- maxTokens: 4096,
2907
+ maxTokens: 131072,
2262
2908
  },
2263
2909
  "meta-llama/llama-3.3-8b-instruct:free": {
2264
2910
  id: "meta-llama/llama-3.3-8b-instruct:free",
@@ -2277,6 +2923,23 @@ export const MODELS = {
2277
2923
  contextWindow: 128000,
2278
2924
  maxTokens: 4028,
2279
2925
  },
2926
+ "nousresearch/deephermes-3-mistral-24b-preview": {
2927
+ id: "nousresearch/deephermes-3-mistral-24b-preview",
2928
+ name: "Nous: DeepHermes 3 Mistral 24B Preview",
2929
+ api: "openai-completions",
2930
+ provider: "openrouter",
2931
+ baseUrl: "https://openrouter.ai/api/v1",
2932
+ reasoning: true,
2933
+ input: ["text"],
2934
+ cost: {
2935
+ input: 0.15,
2936
+ output: 0.59,
2937
+ cacheRead: 0,
2938
+ cacheWrite: 0,
2939
+ },
2940
+ contextWindow: 32768,
2941
+ maxTokens: 32768,
2942
+ },
2280
2943
  "mistralai/mistral-medium-3": {
2281
2944
  id: "mistralai/mistral-medium-3",
2282
2945
  name: "Mistral: Mistral Medium 3",
@@ -2360,7 +3023,7 @@ export const MODELS = {
2360
3023
  cacheWrite: 0,
2361
3024
  },
2362
3025
  contextWindow: 40960,
2363
- maxTokens: 4096,
3026
+ maxTokens: 40960,
2364
3027
  },
2365
3028
  "qwen/qwen3-14b": {
2366
3029
  id: "qwen/qwen3-14b",
@@ -2371,8 +3034,8 @@ export const MODELS = {
2371
3034
  reasoning: true,
2372
3035
  input: ["text"],
2373
3036
  cost: {
2374
- input: 0.06,
2375
- output: 0.24,
3037
+ input: 0.049999999999999996,
3038
+ output: 0.22,
2376
3039
  cacheRead: 0,
2377
3040
  cacheWrite: 0,
2378
3041
  },
@@ -2388,13 +3051,13 @@ export const MODELS = {
2388
3051
  reasoning: true,
2389
3052
  input: ["text"],
2390
3053
  cost: {
2391
- input: 0.03,
2392
- output: 0.13,
3054
+ input: 0.049999999999999996,
3055
+ output: 0.19999999999999998,
2393
3056
  cacheRead: 0,
2394
3057
  cacheWrite: 0,
2395
3058
  },
2396
3059
  contextWindow: 40960,
2397
- maxTokens: 4096,
3060
+ maxTokens: 40960,
2398
3061
  },
2399
3062
  "qwen/qwen3-235b-a22b:free": {
2400
3063
  id: "qwen/qwen3-235b-a22b:free",
@@ -2410,7 +3073,7 @@ export const MODELS = {
2410
3073
  cacheRead: 0,
2411
3074
  cacheWrite: 0,
2412
3075
  },
2413
- contextWindow: 131072,
3076
+ contextWindow: 40960,
2414
3077
  maxTokens: 4096,
2415
3078
  },
2416
3079
  "qwen/qwen3-235b-a22b": {
@@ -2495,8 +3158,8 @@ export const MODELS = {
2495
3158
  cacheRead: 0,
2496
3159
  cacheWrite: 0,
2497
3160
  },
2498
- contextWindow: 1048576,
2499
- maxTokens: 1048576,
3161
+ contextWindow: 327680,
3162
+ maxTokens: 16384,
2500
3163
  },
2501
3164
  "deepseek/deepseek-chat-v3-0324:free": {
2502
3165
  id: "deepseek/deepseek-chat-v3-0324:free",
@@ -2521,16 +3184,16 @@ export const MODELS = {
2521
3184
  api: "openai-completions",
2522
3185
  provider: "openrouter",
2523
3186
  baseUrl: "https://openrouter.ai/api/v1",
2524
- reasoning: false,
3187
+ reasoning: true,
2525
3188
  input: ["text"],
2526
3189
  cost: {
2527
- input: 0.24999987999999998,
2528
- output: 0.999999888,
3190
+ input: 0.24,
3191
+ output: 0.84,
2529
3192
  cacheRead: 0,
2530
3193
  cacheWrite: 0,
2531
3194
  },
2532
3195
  contextWindow: 163840,
2533
- maxTokens: 4096,
3196
+ maxTokens: 163840,
2534
3197
  },
2535
3198
  "mistralai/mistral-small-3.1-24b-instruct:free": {
2536
3199
  id: "mistralai/mistral-small-3.1-24b-instruct:free",
@@ -2557,30 +3220,13 @@ export const MODELS = {
2557
3220
  baseUrl: "https://openrouter.ai/api/v1",
2558
3221
  reasoning: false,
2559
3222
  input: ["text", "image"],
2560
- cost: {
2561
- input: 0.04,
2562
- output: 0.15,
2563
- cacheRead: 0,
2564
- cacheWrite: 0,
2565
- },
2566
- contextWindow: 131072,
2567
- maxTokens: 96000,
2568
- },
2569
- "microsoft/phi-4-multimodal-instruct": {
2570
- id: "microsoft/phi-4-multimodal-instruct",
2571
- name: "Microsoft: Phi 4 Multimodal Instruct",
2572
- api: "openai-completions",
2573
- provider: "openrouter",
2574
- baseUrl: "https://openrouter.ai/api/v1",
2575
- reasoning: false,
2576
- input: ["text", "image"],
2577
3223
  cost: {
2578
3224
  input: 0.049999999999999996,
2579
3225
  output: 0.09999999999999999,
2580
3226
  cacheRead: 0,
2581
3227
  cacheWrite: 0,
2582
3228
  },
2583
- contextWindow: 131072,
3229
+ contextWindow: 128000,
2584
3230
  maxTokens: 4096,
2585
3231
  },
2586
3232
  "qwen/qwq-32b": {
@@ -2600,6 +3246,23 @@ export const MODELS = {
2600
3246
  contextWindow: 32768,
2601
3247
  maxTokens: 4096,
2602
3248
  },
3249
+ "nousresearch/deephermes-3-llama-3-8b-preview": {
3250
+ id: "nousresearch/deephermes-3-llama-3-8b-preview",
3251
+ name: "Nous: DeepHermes 3 Llama 3 8B Preview",
3252
+ api: "openai-completions",
3253
+ provider: "openrouter",
3254
+ baseUrl: "https://openrouter.ai/api/v1",
3255
+ reasoning: false,
3256
+ input: ["text"],
3257
+ cost: {
3258
+ input: 0.03,
3259
+ output: 0.11,
3260
+ cacheRead: 0,
3261
+ cacheWrite: 0,
3262
+ },
3263
+ contextWindow: 131072,
3264
+ maxTokens: 131072,
3265
+ },
2603
3266
  "mistralai/mistral-saba": {
2604
3267
  id: "mistralai/mistral-saba",
2605
3268
  name: "Mistral: Saba",
@@ -2617,6 +3280,23 @@ export const MODELS = {
2617
3280
  contextWindow: 32768,
2618
3281
  maxTokens: 4096,
2619
3282
  },
3283
+ "qwen/qwen-vl-max": {
3284
+ id: "qwen/qwen-vl-max",
3285
+ name: "Qwen: Qwen VL Max",
3286
+ api: "openai-completions",
3287
+ provider: "openrouter",
3288
+ baseUrl: "https://openrouter.ai/api/v1",
3289
+ reasoning: false,
3290
+ input: ["text", "image"],
3291
+ cost: {
3292
+ input: 0.7999999999999999,
3293
+ output: 3.1999999999999997,
3294
+ cacheRead: 0,
3295
+ cacheWrite: 0,
3296
+ },
3297
+ contextWindow: 131072,
3298
+ maxTokens: 8192,
3299
+ },
2620
3300
  "qwen/qwen-turbo": {
2621
3301
  id: "qwen/qwen-turbo",
2622
3302
  name: "Qwen: Qwen-Turbo",
@@ -2677,13 +3357,13 @@ export const MODELS = {
2677
3357
  reasoning: false,
2678
3358
  input: ["text"],
2679
3359
  cost: {
2680
- input: 0.04,
2681
- output: 0.15,
3360
+ input: 0.049999999999999996,
3361
+ output: 0.08,
2682
3362
  cacheRead: 0,
2683
3363
  cacheWrite: 0,
2684
3364
  },
2685
3365
  contextWindow: 32768,
2686
- maxTokens: 4096,
3366
+ maxTokens: 16384,
2687
3367
  },
2688
3368
  "deepseek/deepseek-r1-distill-llama-70b": {
2689
3369
  id: "deepseek/deepseek-r1-distill-llama-70b",
@@ -2700,7 +3380,7 @@ export const MODELS = {
2700
3380
  cacheWrite: 0,
2701
3381
  },
2702
3382
  contextWindow: 131072,
2703
- maxTokens: 4096,
3383
+ maxTokens: 131072,
2704
3384
  },
2705
3385
  "deepseek/deepseek-r1": {
2706
3386
  id: "deepseek/deepseek-r1",
@@ -2745,13 +3425,13 @@ export const MODELS = {
2745
3425
  reasoning: false,
2746
3426
  input: ["text"],
2747
3427
  cost: {
2748
- input: 0.24999987999999998,
2749
- output: 0.999999888,
3428
+ input: 0.3,
3429
+ output: 0.85,
2750
3430
  cacheRead: 0,
2751
3431
  cacheWrite: 0,
2752
3432
  },
2753
3433
  contextWindow: 163840,
2754
- maxTokens: 4096,
3434
+ maxTokens: 163840,
2755
3435
  },
2756
3436
  "meta-llama/llama-3.3-70b-instruct:free": {
2757
3437
  id: "meta-llama/llama-3.3-70b-instruct:free",
@@ -2767,8 +3447,8 @@ export const MODELS = {
2767
3447
  cacheRead: 0,
2768
3448
  cacheWrite: 0,
2769
3449
  },
2770
- contextWindow: 65536,
2771
- maxTokens: 4096,
3450
+ contextWindow: 131072,
3451
+ maxTokens: 2048,
2772
3452
  },
2773
3453
  "meta-llama/llama-3.3-70b-instruct": {
2774
3454
  id: "meta-llama/llama-3.3-70b-instruct",
@@ -2779,13 +3459,13 @@ export const MODELS = {
2779
3459
  reasoning: false,
2780
3460
  input: ["text"],
2781
3461
  cost: {
2782
- input: 0.012,
2783
- output: 0.036,
3462
+ input: 0.13,
3463
+ output: 0.38,
2784
3464
  cacheRead: 0,
2785
3465
  cacheWrite: 0,
2786
3466
  },
2787
3467
  contextWindow: 131072,
2788
- maxTokens: 8192,
3468
+ maxTokens: 16384,
2789
3469
  },
2790
3470
  "amazon/nova-lite-v1": {
2791
3471
  id: "amazon/nova-lite-v1",
@@ -2923,6 +3603,23 @@ export const MODELS = {
2923
3603
  contextWindow: 128000,
2924
3604
  maxTokens: 4096,
2925
3605
  },
3606
+ "qwen/qwen-2.5-7b-instruct": {
3607
+ id: "qwen/qwen-2.5-7b-instruct",
3608
+ name: "Qwen: Qwen2.5 7B Instruct",
3609
+ api: "openai-completions",
3610
+ provider: "openrouter",
3611
+ baseUrl: "https://openrouter.ai/api/v1",
3612
+ reasoning: false,
3613
+ input: ["text"],
3614
+ cost: {
3615
+ input: 0.04,
3616
+ output: 0.09999999999999999,
3617
+ cacheRead: 0,
3618
+ cacheWrite: 0,
3619
+ },
3620
+ contextWindow: 32768,
3621
+ maxTokens: 16384,
3622
+ },
2926
3623
  "nvidia/llama-3.1-nemotron-70b-instruct": {
2927
3624
  id: "nvidia/llama-3.1-nemotron-70b-instruct",
2928
3625
  name: "NVIDIA: Llama 3.1 Nemotron 70B Instruct",
@@ -2989,7 +3686,7 @@ export const MODELS = {
2989
3686
  cacheWrite: 0,
2990
3687
  },
2991
3688
  contextWindow: 32768,
2992
- maxTokens: 4096,
3689
+ maxTokens: 32768,
2993
3690
  },
2994
3691
  "mistralai/pixtral-12b": {
2995
3692
  id: "mistralai/pixtral-12b",
@@ -3008,6 +3705,23 @@ export const MODELS = {
3008
3705
  contextWindow: 32768,
3009
3706
  maxTokens: 4096,
3010
3707
  },
3708
+ "cohere/command-r-plus-08-2024": {
3709
+ id: "cohere/command-r-plus-08-2024",
3710
+ name: "Cohere: Command R+ (08-2024)",
3711
+ api: "openai-completions",
3712
+ provider: "openrouter",
3713
+ baseUrl: "https://openrouter.ai/api/v1",
3714
+ reasoning: false,
3715
+ input: ["text"],
3716
+ cost: {
3717
+ input: 2.5,
3718
+ output: 10,
3719
+ cacheRead: 0,
3720
+ cacheWrite: 0,
3721
+ },
3722
+ contextWindow: 128000,
3723
+ maxTokens: 4000,
3724
+ },
3011
3725
  "cohere/command-r-08-2024": {
3012
3726
  id: "cohere/command-r-08-2024",
3013
3727
  name: "Cohere: Command R (08-2024)",
@@ -3025,22 +3739,22 @@ export const MODELS = {
3025
3739
  contextWindow: 128000,
3026
3740
  maxTokens: 4000,
3027
3741
  },
3028
- "cohere/command-r-plus-08-2024": {
3029
- id: "cohere/command-r-plus-08-2024",
3030
- name: "Cohere: Command R+ (08-2024)",
3742
+ "sao10k/l3.1-euryale-70b": {
3743
+ id: "sao10k/l3.1-euryale-70b",
3744
+ name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
3031
3745
  api: "openai-completions",
3032
3746
  provider: "openrouter",
3033
3747
  baseUrl: "https://openrouter.ai/api/v1",
3034
3748
  reasoning: false,
3035
3749
  input: ["text"],
3036
3750
  cost: {
3037
- input: 2.5,
3038
- output: 10,
3751
+ input: 0.65,
3752
+ output: 0.75,
3039
3753
  cacheRead: 0,
3040
3754
  cacheWrite: 0,
3041
3755
  },
3042
- contextWindow: 128000,
3043
- maxTokens: 4000,
3756
+ contextWindow: 32768,
3757
+ maxTokens: 4096,
3044
3758
  },
3045
3759
  "microsoft/phi-3.5-mini-128k-instruct": {
3046
3760
  id: "microsoft/phi-3.5-mini-128k-instruct",
@@ -3068,13 +3782,13 @@ export const MODELS = {
3068
3782
  reasoning: false,
3069
3783
  input: ["text"],
3070
3784
  cost: {
3071
- input: 0.12,
3785
+ input: 0.3,
3072
3786
  output: 0.3,
3073
3787
  cacheRead: 0,
3074
3788
  cacheWrite: 0,
3075
3789
  },
3076
- contextWindow: 131072,
3077
- maxTokens: 131072,
3790
+ contextWindow: 65536,
3791
+ maxTokens: 4096,
3078
3792
  },
3079
3793
  "meta-llama/llama-3.1-8b-instruct": {
3080
3794
  id: "meta-llama/llama-3.1-8b-instruct",
@@ -3119,13 +3833,13 @@ export const MODELS = {
3119
3833
  reasoning: false,
3120
3834
  input: ["text"],
3121
3835
  cost: {
3122
- input: 0.09999999999999999,
3123
- output: 0.28,
3836
+ input: 0.39999999999999997,
3837
+ output: 0.39999999999999997,
3124
3838
  cacheRead: 0,
3125
3839
  cacheWrite: 0,
3126
3840
  },
3127
3841
  contextWindow: 131072,
3128
- maxTokens: 16384,
3842
+ maxTokens: 4096,
3129
3843
  },
3130
3844
  "mistralai/mistral-nemo": {
3131
3845
  id: "mistralai/mistral-nemo",
@@ -3144,22 +3858,22 @@ export const MODELS = {
3144
3858
  contextWindow: 131072,
3145
3859
  maxTokens: 16384,
3146
3860
  },
3147
- "mistralai/mistral-7b-instruct-v0.3": {
3148
- id: "mistralai/mistral-7b-instruct-v0.3",
3149
- name: "Mistral: Mistral 7B Instruct v0.3",
3861
+ "sao10k/l3-euryale-70b": {
3862
+ id: "sao10k/l3-euryale-70b",
3863
+ name: "Sao10k: Llama 3 Euryale 70B v2.1",
3150
3864
  api: "openai-completions",
3151
3865
  provider: "openrouter",
3152
3866
  baseUrl: "https://openrouter.ai/api/v1",
3153
3867
  reasoning: false,
3154
3868
  input: ["text"],
3155
3869
  cost: {
3156
- input: 0.028,
3157
- output: 0.054,
3870
+ input: 1.48,
3871
+ output: 1.48,
3158
3872
  cacheRead: 0,
3159
3873
  cacheWrite: 0,
3160
3874
  },
3161
- contextWindow: 32768,
3162
- maxTokens: 16384,
3875
+ contextWindow: 8192,
3876
+ maxTokens: 8192,
3163
3877
  },
3164
3878
  "mistralai/mistral-7b-instruct:free": {
3165
3879
  id: "mistralai/mistral-7b-instruct:free",
@@ -3195,6 +3909,23 @@ export const MODELS = {
3195
3909
  contextWindow: 32768,
3196
3910
  maxTokens: 16384,
3197
3911
  },
3912
+ "mistralai/mistral-7b-instruct-v0.3": {
3913
+ id: "mistralai/mistral-7b-instruct-v0.3",
3914
+ name: "Mistral: Mistral 7B Instruct v0.3",
3915
+ api: "openai-completions",
3916
+ provider: "openrouter",
3917
+ baseUrl: "https://openrouter.ai/api/v1",
3918
+ reasoning: false,
3919
+ input: ["text"],
3920
+ cost: {
3921
+ input: 0.028,
3922
+ output: 0.054,
3923
+ cacheRead: 0,
3924
+ cacheWrite: 0,
3925
+ },
3926
+ contextWindow: 32768,
3927
+ maxTokens: 16384,
3928
+ },
3198
3929
  "microsoft/phi-3-mini-128k-instruct": {
3199
3930
  id: "microsoft/phi-3-mini-128k-instruct",
3200
3931
  name: "Microsoft: Phi-3 Mini 128K Instruct",
@@ -3272,82 +4003,14 @@ export const MODELS = {
3272
4003
  reasoning: false,
3273
4004
  input: ["text"],
3274
4005
  cost: {
3275
- input: 0.8999999999999999,
3276
- output: 0.8999999999999999,
4006
+ input: 2,
4007
+ output: 6,
3277
4008
  cacheRead: 0,
3278
4009
  cacheWrite: 0,
3279
4010
  },
3280
4011
  contextWindow: 65536,
3281
4012
  maxTokens: 4096,
3282
4013
  },
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
4014
  "mistralai/mistral-large": {
3352
4015
  id: "mistralai/mistral-large",
3353
4016
  name: "Mistral Large",
@@ -3365,34 +4028,34 @@ export const MODELS = {
3365
4028
  contextWindow: 128000,
3366
4029
  maxTokens: 4096,
3367
4030
  },
3368
- "mistralai/mistral-tiny": {
3369
- id: "mistralai/mistral-tiny",
3370
- name: "Mistral Tiny",
4031
+ "mistralai/mistral-small": {
4032
+ id: "mistralai/mistral-small",
4033
+ name: "Mistral Small",
3371
4034
  api: "openai-completions",
3372
4035
  provider: "openrouter",
3373
4036
  baseUrl: "https://openrouter.ai/api/v1",
3374
4037
  reasoning: false,
3375
4038
  input: ["text"],
3376
4039
  cost: {
3377
- input: 0.25,
3378
- output: 0.25,
4040
+ input: 0.19999999999999998,
4041
+ output: 0.6,
3379
4042
  cacheRead: 0,
3380
4043
  cacheWrite: 0,
3381
4044
  },
3382
4045
  contextWindow: 32768,
3383
4046
  maxTokens: 4096,
3384
4047
  },
3385
- "mistralai/mistral-small": {
3386
- id: "mistralai/mistral-small",
3387
- name: "Mistral Small",
4048
+ "mistralai/mistral-tiny": {
4049
+ id: "mistralai/mistral-tiny",
4050
+ name: "Mistral Tiny",
3388
4051
  api: "openai-completions",
3389
4052
  provider: "openrouter",
3390
4053
  baseUrl: "https://openrouter.ai/api/v1",
3391
4054
  reasoning: false,
3392
4055
  input: ["text"],
3393
4056
  cost: {
3394
- input: 0.19999999999999998,
3395
- output: 0.6,
4057
+ input: 0.25,
4058
+ output: 0.25,
3396
4059
  cacheRead: 0,
3397
4060
  cacheWrite: 0,
3398
4061
  },
@@ -3408,8 +4071,8 @@ export const MODELS = {
3408
4071
  reasoning: false,
3409
4072
  input: ["text"],
3410
4073
  cost: {
3411
- input: 0.39999999999999997,
3412
- output: 0.39999999999999997,
4074
+ input: 0.54,
4075
+ output: 0.54,
3413
4076
  cacheRead: 0,
3414
4077
  cacheWrite: 0,
3415
4078
  },