@oh-my-pi/pi-ai 5.3.1 → 5.4.2

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 (2) hide show
  1. package/package.json +1 -1
  2. package/src/models.generated.ts +2685 -606
@@ -5,47 +5,47 @@ import type { Model } from "./types";
5
5
 
6
6
  export const MODELS = {
7
7
  "amazon-bedrock": {
8
- "eu.anthropic.claude-haiku-4-5-20251001-v1:0": {
9
- id: "eu.anthropic.claude-haiku-4-5-20251001-v1:0",
10
- name: "Claude Haiku 4.5 (EU)",
8
+ "anthropic.claude-3-5-haiku-20241022-v1:0": {
9
+ id: "anthropic.claude-3-5-haiku-20241022-v1:0",
10
+ name: "Claude Haiku 3.5",
11
11
  api: "bedrock-converse-stream",
12
12
  provider: "amazon-bedrock",
13
13
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
14
- reasoning: true,
14
+ reasoning: false,
15
15
  input: ["text", "image"],
16
16
  cost: {
17
- input: 1,
18
- output: 5,
19
- cacheRead: 0.1,
20
- cacheWrite: 1.25,
17
+ input: 0.8,
18
+ output: 4,
19
+ cacheRead: 0.08,
20
+ cacheWrite: 1,
21
21
  },
22
22
  contextWindow: 200000,
23
- maxTokens: 64000,
23
+ maxTokens: 8192,
24
24
  } satisfies Model<"bedrock-converse-stream">,
25
- "eu.anthropic.claude-opus-4-5-20251101-v1:0": {
26
- id: "eu.anthropic.claude-opus-4-5-20251101-v1:0",
27
- name: "Claude Opus 4.5 (EU)",
25
+ "anthropic.claude-3-5-sonnet-20240620-v1:0": {
26
+ id: "anthropic.claude-3-5-sonnet-20240620-v1:0",
27
+ name: "Claude Sonnet 3.5",
28
28
  api: "bedrock-converse-stream",
29
29
  provider: "amazon-bedrock",
30
30
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
31
- reasoning: true,
31
+ reasoning: false,
32
32
  input: ["text", "image"],
33
33
  cost: {
34
- input: 5,
35
- output: 25,
36
- cacheRead: 0.5,
37
- cacheWrite: 6.25,
34
+ input: 3,
35
+ output: 15,
36
+ cacheRead: 0.3,
37
+ cacheWrite: 3.75,
38
38
  },
39
39
  contextWindow: 200000,
40
- maxTokens: 64000,
40
+ maxTokens: 8192,
41
41
  } satisfies Model<"bedrock-converse-stream">,
42
- "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": {
43
- id: "eu.anthropic.claude-sonnet-4-5-20250929-v1:0",
44
- name: "Claude Sonnet 4.5 (EU)",
42
+ "anthropic.claude-3-5-sonnet-20241022-v2:0": {
43
+ id: "anthropic.claude-3-5-sonnet-20241022-v2:0",
44
+ name: "Claude Sonnet 3.5 v2",
45
45
  api: "bedrock-converse-stream",
46
46
  provider: "amazon-bedrock",
47
47
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
48
- reasoning: true,
48
+ reasoning: false,
49
49
  input: ["text", "image"],
50
50
  cost: {
51
51
  input: 3,
@@ -54,202 +54,219 @@ export const MODELS = {
54
54
  cacheWrite: 3.75,
55
55
  },
56
56
  contextWindow: 200000,
57
- maxTokens: 64000,
57
+ maxTokens: 8192,
58
58
  } satisfies Model<"bedrock-converse-stream">,
59
- "us.ai21.jamba-1-5-large-v1:0": {
60
- id: "us.ai21.jamba-1-5-large-v1:0",
61
- name: "Jamba 1.5 Large",
59
+ "anthropic.claude-3-haiku-20240307-v1:0": {
60
+ id: "anthropic.claude-3-haiku-20240307-v1:0",
61
+ name: "Claude Haiku 3",
62
62
  api: "bedrock-converse-stream",
63
63
  provider: "amazon-bedrock",
64
64
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
65
65
  reasoning: false,
66
- input: ["text"],
66
+ input: ["text", "image"],
67
67
  cost: {
68
- input: 2,
69
- output: 8,
68
+ input: 0.25,
69
+ output: 1.25,
70
70
  cacheRead: 0,
71
71
  cacheWrite: 0,
72
72
  },
73
- contextWindow: 256000,
73
+ contextWindow: 200000,
74
74
  maxTokens: 4096,
75
75
  } satisfies Model<"bedrock-converse-stream">,
76
- "us.ai21.jamba-1-5-mini-v1:0": {
77
- id: "us.ai21.jamba-1-5-mini-v1:0",
78
- name: "Jamba 1.5 Mini",
76
+ "anthropic.claude-3-opus-20240229-v1:0": {
77
+ id: "anthropic.claude-3-opus-20240229-v1:0",
78
+ name: "Claude Opus 3",
79
79
  api: "bedrock-converse-stream",
80
80
  provider: "amazon-bedrock",
81
81
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
82
82
  reasoning: false,
83
- input: ["text"],
83
+ input: ["text", "image"],
84
84
  cost: {
85
- input: 0.2,
86
- output: 0.4,
85
+ input: 15,
86
+ output: 75,
87
87
  cacheRead: 0,
88
88
  cacheWrite: 0,
89
89
  },
90
- contextWindow: 256000,
90
+ contextWindow: 200000,
91
91
  maxTokens: 4096,
92
92
  } satisfies Model<"bedrock-converse-stream">,
93
- "us.amazon.nova-2-lite-v1:0": {
94
- id: "us.amazon.nova-2-lite-v1:0",
95
- name: "Nova 2 Lite",
93
+ "anthropic.claude-3-sonnet-20240229-v1:0": {
94
+ id: "anthropic.claude-3-sonnet-20240229-v1:0",
95
+ name: "Claude Sonnet 3",
96
96
  api: "bedrock-converse-stream",
97
97
  provider: "amazon-bedrock",
98
98
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
99
99
  reasoning: false,
100
100
  input: ["text", "image"],
101
101
  cost: {
102
- input: 0.33,
103
- output: 2.75,
102
+ input: 3,
103
+ output: 15,
104
104
  cacheRead: 0,
105
105
  cacheWrite: 0,
106
106
  },
107
- contextWindow: 128000,
107
+ contextWindow: 200000,
108
108
  maxTokens: 4096,
109
109
  } satisfies Model<"bedrock-converse-stream">,
110
- "us.amazon.nova-lite-v1:0": {
111
- id: "us.amazon.nova-lite-v1:0",
112
- name: "Nova Lite",
110
+ "cohere.command-r-plus-v1:0": {
111
+ id: "cohere.command-r-plus-v1:0",
112
+ name: "Command R+",
113
113
  api: "bedrock-converse-stream",
114
114
  provider: "amazon-bedrock",
115
115
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
116
116
  reasoning: false,
117
- input: ["text", "image"],
117
+ input: ["text"],
118
118
  cost: {
119
- input: 0.06,
120
- output: 0.24,
121
- cacheRead: 0.015,
119
+ input: 3,
120
+ output: 15,
121
+ cacheRead: 0,
122
122
  cacheWrite: 0,
123
123
  },
124
- contextWindow: 300000,
125
- maxTokens: 8192,
124
+ contextWindow: 128000,
125
+ maxTokens: 4096,
126
126
  } satisfies Model<"bedrock-converse-stream">,
127
- "us.amazon.nova-micro-v1:0": {
128
- id: "us.amazon.nova-micro-v1:0",
129
- name: "Nova Micro",
127
+ "cohere.command-r-v1:0": {
128
+ id: "cohere.command-r-v1:0",
129
+ name: "Command R",
130
130
  api: "bedrock-converse-stream",
131
131
  provider: "amazon-bedrock",
132
132
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
133
133
  reasoning: false,
134
134
  input: ["text"],
135
135
  cost: {
136
- input: 0.035,
137
- output: 0.14,
138
- cacheRead: 0.00875,
136
+ input: 0.5,
137
+ output: 1.5,
138
+ cacheRead: 0,
139
139
  cacheWrite: 0,
140
140
  },
141
141
  contextWindow: 128000,
142
- maxTokens: 8192,
142
+ maxTokens: 4096,
143
143
  } satisfies Model<"bedrock-converse-stream">,
144
- "us.amazon.nova-premier-v1:0": {
145
- id: "us.amazon.nova-premier-v1:0",
146
- name: "Nova Premier",
144
+ "deepseek.v3-v1:0": {
145
+ id: "deepseek.v3-v1:0",
146
+ name: "DeepSeek-V3.1",
147
147
  api: "bedrock-converse-stream",
148
148
  provider: "amazon-bedrock",
149
149
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
150
150
  reasoning: true,
151
- input: ["text", "image"],
151
+ input: ["text"],
152
152
  cost: {
153
- input: 2.5,
154
- output: 12.5,
153
+ input: 0.58,
154
+ output: 1.68,
155
155
  cacheRead: 0,
156
156
  cacheWrite: 0,
157
157
  },
158
- contextWindow: 1000000,
159
- maxTokens: 16384,
158
+ contextWindow: 163840,
159
+ maxTokens: 81920,
160
160
  } satisfies Model<"bedrock-converse-stream">,
161
- "us.amazon.nova-pro-v1:0": {
162
- id: "us.amazon.nova-pro-v1:0",
163
- name: "Nova Pro",
161
+ "eu.anthropic.claude-haiku-4-5-20251001-v1:0": {
162
+ id: "eu.anthropic.claude-haiku-4-5-20251001-v1:0",
163
+ name: "Claude Haiku 4.5 (EU)",
164
164
  api: "bedrock-converse-stream",
165
165
  provider: "amazon-bedrock",
166
166
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
167
- reasoning: false,
167
+ reasoning: true,
168
168
  input: ["text", "image"],
169
169
  cost: {
170
- input: 0.8,
171
- output: 3.2,
172
- cacheRead: 0.2,
173
- cacheWrite: 0,
170
+ input: 1,
171
+ output: 5,
172
+ cacheRead: 0.1,
173
+ cacheWrite: 1.25,
174
174
  },
175
- contextWindow: 300000,
176
- maxTokens: 8192,
175
+ contextWindow: 200000,
176
+ maxTokens: 64000,
177
177
  } satisfies Model<"bedrock-converse-stream">,
178
- "us.amazon.titan-text-express-v1": {
179
- id: "us.amazon.titan-text-express-v1",
180
- name: "Titan Text G1 - Express",
178
+ "eu.anthropic.claude-opus-4-5-20251101-v1:0": {
179
+ id: "eu.anthropic.claude-opus-4-5-20251101-v1:0",
180
+ name: "Claude Opus 4.5 (EU)",
181
181
  api: "bedrock-converse-stream",
182
182
  provider: "amazon-bedrock",
183
183
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
184
- reasoning: false,
185
- input: ["text"],
184
+ reasoning: true,
185
+ input: ["text", "image"],
186
186
  cost: {
187
- input: 0.2,
188
- output: 0.6,
189
- cacheRead: 0,
190
- cacheWrite: 0,
187
+ input: 5,
188
+ output: 25,
189
+ cacheRead: 0.5,
190
+ cacheWrite: 6.25,
191
191
  },
192
- contextWindow: 128000,
193
- maxTokens: 4096,
192
+ contextWindow: 200000,
193
+ maxTokens: 64000,
194
+ } satisfies Model<"bedrock-converse-stream">,
195
+ "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": {
196
+ id: "eu.anthropic.claude-sonnet-4-5-20250929-v1:0",
197
+ name: "Claude Sonnet 4.5 (EU)",
198
+ api: "bedrock-converse-stream",
199
+ provider: "amazon-bedrock",
200
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
201
+ reasoning: true,
202
+ input: ["text", "image"],
203
+ cost: {
204
+ input: 3,
205
+ output: 15,
206
+ cacheRead: 0.3,
207
+ cacheWrite: 3.75,
208
+ },
209
+ contextWindow: 200000,
210
+ maxTokens: 64000,
194
211
  } satisfies Model<"bedrock-converse-stream">,
195
- "us.amazon.titan-text-express-v1:0:8k": {
196
- id: "us.amazon.titan-text-express-v1:0:8k",
197
- name: "Titan Text G1 - Express",
212
+ "global.amazon.nova-2-lite-v1:0": {
213
+ id: "global.amazon.nova-2-lite-v1:0",
214
+ name: "Nova 2 Lite",
198
215
  api: "bedrock-converse-stream",
199
216
  provider: "amazon-bedrock",
200
217
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
201
218
  reasoning: false,
202
- input: ["text"],
219
+ input: ["text", "image"],
203
220
  cost: {
204
- input: 0.2,
205
- output: 0.6,
221
+ input: 0.33,
222
+ output: 2.75,
206
223
  cacheRead: 0,
207
224
  cacheWrite: 0,
208
225
  },
209
226
  contextWindow: 128000,
210
227
  maxTokens: 4096,
211
228
  } satisfies Model<"bedrock-converse-stream">,
212
- "us.anthropic.claude-3-5-haiku-20241022-v1:0": {
213
- id: "us.anthropic.claude-3-5-haiku-20241022-v1:0",
214
- name: "Claude Haiku 3.5",
229
+ "global.anthropic.claude-haiku-4-5-20251001-v1:0": {
230
+ id: "global.anthropic.claude-haiku-4-5-20251001-v1:0",
231
+ name: "Claude Haiku 4.5",
215
232
  api: "bedrock-converse-stream",
216
233
  provider: "amazon-bedrock",
217
234
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
218
- reasoning: false,
235
+ reasoning: true,
219
236
  input: ["text", "image"],
220
237
  cost: {
221
- input: 0.8,
222
- output: 4,
223
- cacheRead: 0.08,
224
- cacheWrite: 1,
238
+ input: 1,
239
+ output: 5,
240
+ cacheRead: 0.1,
241
+ cacheWrite: 1.25,
225
242
  },
226
243
  contextWindow: 200000,
227
- maxTokens: 8192,
244
+ maxTokens: 64000,
228
245
  } satisfies Model<"bedrock-converse-stream">,
229
- "us.anthropic.claude-3-5-sonnet-20240620-v1:0": {
230
- id: "us.anthropic.claude-3-5-sonnet-20240620-v1:0",
231
- name: "Claude Sonnet 3.5",
246
+ "global.anthropic.claude-opus-4-5-20251101-v1:0": {
247
+ id: "global.anthropic.claude-opus-4-5-20251101-v1:0",
248
+ name: "Claude Opus 4.5",
232
249
  api: "bedrock-converse-stream",
233
250
  provider: "amazon-bedrock",
234
251
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
235
- reasoning: false,
252
+ reasoning: true,
236
253
  input: ["text", "image"],
237
254
  cost: {
238
- input: 3,
239
- output: 15,
240
- cacheRead: 0.3,
241
- cacheWrite: 3.75,
255
+ input: 5,
256
+ output: 25,
257
+ cacheRead: 0.5,
258
+ cacheWrite: 6.25,
242
259
  },
243
260
  contextWindow: 200000,
244
- maxTokens: 8192,
261
+ maxTokens: 64000,
245
262
  } satisfies Model<"bedrock-converse-stream">,
246
- "us.anthropic.claude-3-5-sonnet-20241022-v2:0": {
247
- id: "us.anthropic.claude-3-5-sonnet-20241022-v2:0",
248
- name: "Claude Sonnet 3.5 v2",
263
+ "global.anthropic.claude-sonnet-4-20250514-v1:0": {
264
+ id: "global.anthropic.claude-sonnet-4-20250514-v1:0",
265
+ name: "Claude Sonnet 4",
249
266
  api: "bedrock-converse-stream",
250
267
  provider: "amazon-bedrock",
251
268
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
252
- reasoning: false,
269
+ reasoning: true,
253
270
  input: ["text", "image"],
254
271
  cost: {
255
272
  input: 3,
@@ -258,15 +275,15 @@ export const MODELS = {
258
275
  cacheWrite: 3.75,
259
276
  },
260
277
  contextWindow: 200000,
261
- maxTokens: 8192,
278
+ maxTokens: 64000,
262
279
  } satisfies Model<"bedrock-converse-stream">,
263
- "us.anthropic.claude-3-7-sonnet-20250219-v1:0": {
264
- id: "us.anthropic.claude-3-7-sonnet-20250219-v1:0",
265
- name: "Claude Sonnet 3.7",
280
+ "global.anthropic.claude-sonnet-4-5-20250929-v1:0": {
281
+ id: "global.anthropic.claude-sonnet-4-5-20250929-v1:0",
282
+ name: "Claude Sonnet 4.5",
266
283
  api: "bedrock-converse-stream",
267
284
  provider: "amazon-bedrock",
268
285
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
269
- reasoning: false,
286
+ reasoning: true,
270
287
  input: ["text", "image"],
271
288
  cost: {
272
289
  input: 3,
@@ -275,181 +292,130 @@ export const MODELS = {
275
292
  cacheWrite: 3.75,
276
293
  },
277
294
  contextWindow: 200000,
278
- maxTokens: 8192,
295
+ maxTokens: 64000,
279
296
  } satisfies Model<"bedrock-converse-stream">,
280
- "us.anthropic.claude-3-haiku-20240307-v1:0": {
281
- id: "us.anthropic.claude-3-haiku-20240307-v1:0",
282
- name: "Claude Haiku 3",
297
+ "google.gemma-3-27b-it": {
298
+ id: "google.gemma-3-27b-it",
299
+ name: "Google Gemma 3 27B Instruct",
283
300
  api: "bedrock-converse-stream",
284
301
  provider: "amazon-bedrock",
285
302
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
286
303
  reasoning: false,
287
304
  input: ["text", "image"],
288
305
  cost: {
289
- input: 0.25,
290
- output: 1.25,
306
+ input: 0.12,
307
+ output: 0.2,
291
308
  cacheRead: 0,
292
309
  cacheWrite: 0,
293
310
  },
294
- contextWindow: 200000,
295
- maxTokens: 4096,
311
+ contextWindow: 202752,
312
+ maxTokens: 8192,
296
313
  } satisfies Model<"bedrock-converse-stream">,
297
- "us.anthropic.claude-3-opus-20240229-v1:0": {
298
- id: "us.anthropic.claude-3-opus-20240229-v1:0",
299
- name: "Claude Opus 3",
314
+ "google.gemma-3-4b-it": {
315
+ id: "google.gemma-3-4b-it",
316
+ name: "Gemma 3 4B IT",
300
317
  api: "bedrock-converse-stream",
301
318
  provider: "amazon-bedrock",
302
319
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
303
320
  reasoning: false,
304
321
  input: ["text", "image"],
305
322
  cost: {
306
- input: 15,
307
- output: 75,
323
+ input: 0.04,
324
+ output: 0.08,
308
325
  cacheRead: 0,
309
326
  cacheWrite: 0,
310
327
  },
311
- contextWindow: 200000,
328
+ contextWindow: 128000,
312
329
  maxTokens: 4096,
313
330
  } satisfies Model<"bedrock-converse-stream">,
314
- "us.anthropic.claude-3-sonnet-20240229-v1:0": {
315
- id: "us.anthropic.claude-3-sonnet-20240229-v1:0",
316
- name: "Claude Sonnet 3",
331
+ "meta.llama3-1-70b-instruct-v1:0": {
332
+ id: "meta.llama3-1-70b-instruct-v1:0",
333
+ name: "Llama 3.1 70B Instruct",
317
334
  api: "bedrock-converse-stream",
318
335
  provider: "amazon-bedrock",
319
336
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
320
337
  reasoning: false,
321
- input: ["text", "image"],
338
+ input: ["text"],
322
339
  cost: {
323
- input: 3,
324
- output: 15,
340
+ input: 0.72,
341
+ output: 0.72,
325
342
  cacheRead: 0,
326
343
  cacheWrite: 0,
327
344
  },
328
- contextWindow: 200000,
345
+ contextWindow: 128000,
329
346
  maxTokens: 4096,
330
347
  } satisfies Model<"bedrock-converse-stream">,
331
- "us.anthropic.claude-haiku-4-5-20251001-v1:0": {
332
- id: "us.anthropic.claude-haiku-4-5-20251001-v1:0",
333
- name: "Claude Haiku 4.5",
348
+ "meta.llama3-1-8b-instruct-v1:0": {
349
+ id: "meta.llama3-1-8b-instruct-v1:0",
350
+ name: "Llama 3.1 8B Instruct",
334
351
  api: "bedrock-converse-stream",
335
352
  provider: "amazon-bedrock",
336
353
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
337
- reasoning: true,
338
- input: ["text", "image"],
354
+ reasoning: false,
355
+ input: ["text"],
339
356
  cost: {
340
- input: 1,
341
- output: 5,
342
- cacheRead: 0.1,
343
- cacheWrite: 1.25,
357
+ input: 0.22,
358
+ output: 0.22,
359
+ cacheRead: 0,
360
+ cacheWrite: 0,
344
361
  },
345
- contextWindow: 200000,
346
- maxTokens: 64000,
347
- } satisfies Model<"bedrock-converse-stream">,
348
- "us.anthropic.claude-opus-4-1-20250805-v1:0": {
349
- id: "us.anthropic.claude-opus-4-1-20250805-v1:0",
350
- name: "Claude Opus 4.1",
351
- api: "bedrock-converse-stream",
352
- provider: "amazon-bedrock",
353
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
354
- reasoning: true,
355
- input: ["text", "image"],
356
- cost: {
357
- input: 15,
358
- output: 75,
359
- cacheRead: 1.5,
360
- cacheWrite: 18.75,
361
- },
362
- contextWindow: 200000,
363
- maxTokens: 32000,
364
- } satisfies Model<"bedrock-converse-stream">,
365
- "us.anthropic.claude-opus-4-20250514-v1:0": {
366
- id: "us.anthropic.claude-opus-4-20250514-v1:0",
367
- name: "Claude Opus 4",
368
- api: "bedrock-converse-stream",
369
- provider: "amazon-bedrock",
370
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
371
- reasoning: true,
372
- input: ["text", "image"],
373
- cost: {
374
- input: 15,
375
- output: 75,
376
- cacheRead: 1.5,
377
- cacheWrite: 18.75,
378
- },
379
- contextWindow: 200000,
380
- maxTokens: 32000,
381
- } satisfies Model<"bedrock-converse-stream">,
382
- "us.anthropic.claude-opus-4-5-20251101-v1:0": {
383
- id: "us.anthropic.claude-opus-4-5-20251101-v1:0",
384
- name: "Claude Opus 4.5",
385
- api: "bedrock-converse-stream",
386
- provider: "amazon-bedrock",
387
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
388
- reasoning: true,
389
- input: ["text", "image"],
390
- cost: {
391
- input: 5,
392
- output: 25,
393
- cacheRead: 0.5,
394
- cacheWrite: 6.25,
395
- },
396
- contextWindow: 200000,
397
- maxTokens: 64000,
362
+ contextWindow: 128000,
363
+ maxTokens: 4096,
398
364
  } satisfies Model<"bedrock-converse-stream">,
399
- "us.anthropic.claude-sonnet-4-20250514-v1:0": {
400
- id: "us.anthropic.claude-sonnet-4-20250514-v1:0",
401
- name: "Claude Sonnet 4",
365
+ "minimax.minimax-m2": {
366
+ id: "minimax.minimax-m2",
367
+ name: "MiniMax M2",
402
368
  api: "bedrock-converse-stream",
403
369
  provider: "amazon-bedrock",
404
370
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
405
371
  reasoning: true,
406
- input: ["text", "image"],
372
+ input: ["text"],
407
373
  cost: {
408
- input: 3,
409
- output: 15,
410
- cacheRead: 0.3,
411
- cacheWrite: 3.75,
374
+ input: 0.3,
375
+ output: 1.2,
376
+ cacheRead: 0,
377
+ cacheWrite: 0,
412
378
  },
413
- contextWindow: 200000,
414
- maxTokens: 64000,
379
+ contextWindow: 204608,
380
+ maxTokens: 128000,
415
381
  } satisfies Model<"bedrock-converse-stream">,
416
- "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
417
- id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
418
- name: "Claude Sonnet 4.5",
382
+ "mistral.ministral-3-14b-instruct": {
383
+ id: "mistral.ministral-3-14b-instruct",
384
+ name: "Ministral 14B 3.0",
419
385
  api: "bedrock-converse-stream",
420
386
  provider: "amazon-bedrock",
421
387
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
422
- reasoning: true,
423
- input: ["text", "image"],
388
+ reasoning: false,
389
+ input: ["text"],
424
390
  cost: {
425
- input: 3,
426
- output: 15,
427
- cacheRead: 0.3,
428
- cacheWrite: 3.75,
391
+ input: 0.2,
392
+ output: 0.2,
393
+ cacheRead: 0,
394
+ cacheWrite: 0,
429
395
  },
430
- contextWindow: 200000,
431
- maxTokens: 64000,
396
+ contextWindow: 128000,
397
+ maxTokens: 4096,
432
398
  } satisfies Model<"bedrock-converse-stream">,
433
- "us.cohere.command-r-plus-v1:0": {
434
- id: "us.cohere.command-r-plus-v1:0",
435
- name: "Command R+",
399
+ "mistral.ministral-3-8b-instruct": {
400
+ id: "mistral.ministral-3-8b-instruct",
401
+ name: "Ministral 3 8B",
436
402
  api: "bedrock-converse-stream",
437
403
  provider: "amazon-bedrock",
438
404
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
439
405
  reasoning: false,
440
406
  input: ["text"],
441
407
  cost: {
442
- input: 3,
443
- output: 15,
408
+ input: 0.15,
409
+ output: 0.15,
444
410
  cacheRead: 0,
445
411
  cacheWrite: 0,
446
412
  },
447
413
  contextWindow: 128000,
448
414
  maxTokens: 4096,
449
415
  } satisfies Model<"bedrock-converse-stream">,
450
- "us.cohere.command-r-v1:0": {
451
- id: "us.cohere.command-r-v1:0",
452
- name: "Command R",
416
+ "mistral.mistral-large-2402-v1:0": {
417
+ id: "mistral.mistral-large-2402-v1:0",
418
+ name: "Mistral Large (24.02)",
453
419
  api: "bedrock-converse-stream",
454
420
  provider: "amazon-bedrock",
455
421
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -464,451 +430,434 @@ export const MODELS = {
464
430
  contextWindow: 128000,
465
431
  maxTokens: 4096,
466
432
  } satisfies Model<"bedrock-converse-stream">,
467
- "us.deepseek.r1-v1:0": {
468
- id: "us.deepseek.r1-v1:0",
469
- name: "DeepSeek-R1",
433
+ "mistral.voxtral-mini-3b-2507": {
434
+ id: "mistral.voxtral-mini-3b-2507",
435
+ name: "Voxtral Mini 3B 2507",
470
436
  api: "bedrock-converse-stream",
471
437
  provider: "amazon-bedrock",
472
438
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
473
- reasoning: true,
439
+ reasoning: false,
474
440
  input: ["text"],
475
441
  cost: {
476
- input: 1.35,
477
- output: 5.4,
442
+ input: 0.04,
443
+ output: 0.04,
478
444
  cacheRead: 0,
479
445
  cacheWrite: 0,
480
446
  },
481
447
  contextWindow: 128000,
482
- maxTokens: 32768,
448
+ maxTokens: 4096,
483
449
  } satisfies Model<"bedrock-converse-stream">,
484
- "us.deepseek.v3-v1:0": {
485
- id: "us.deepseek.v3-v1:0",
486
- name: "DeepSeek-V3.1",
450
+ "mistral.voxtral-small-24b-2507": {
451
+ id: "mistral.voxtral-small-24b-2507",
452
+ name: "Voxtral Small 24B 2507",
487
453
  api: "bedrock-converse-stream",
488
454
  provider: "amazon-bedrock",
489
455
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
490
- reasoning: true,
456
+ reasoning: false,
491
457
  input: ["text"],
492
458
  cost: {
493
- input: 0.58,
494
- output: 1.68,
459
+ input: 0.15,
460
+ output: 0.35,
495
461
  cacheRead: 0,
496
462
  cacheWrite: 0,
497
463
  },
498
- contextWindow: 163840,
499
- maxTokens: 81920,
464
+ contextWindow: 32000,
465
+ maxTokens: 8192,
500
466
  } satisfies Model<"bedrock-converse-stream">,
501
- "us.global.anthropic.claude-opus-4-5-20251101-v1:0": {
502
- id: "us.global.anthropic.claude-opus-4-5-20251101-v1:0",
503
- name: "Claude Opus 4.5 (Global)",
467
+ "moonshot.kimi-k2-thinking": {
468
+ id: "moonshot.kimi-k2-thinking",
469
+ name: "Kimi K2 Thinking",
504
470
  api: "bedrock-converse-stream",
505
471
  provider: "amazon-bedrock",
506
472
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
507
473
  reasoning: true,
508
- input: ["text", "image"],
474
+ input: ["text"],
509
475
  cost: {
510
- input: 5,
511
- output: 25,
512
- cacheRead: 0.5,
513
- cacheWrite: 6.25,
476
+ input: 0.6,
477
+ output: 2.5,
478
+ cacheRead: 0,
479
+ cacheWrite: 0,
514
480
  },
515
- contextWindow: 200000,
516
- maxTokens: 64000,
481
+ contextWindow: 256000,
482
+ maxTokens: 256000,
517
483
  } satisfies Model<"bedrock-converse-stream">,
518
- "us.google.gemma-3-27b-it": {
519
- id: "us.google.gemma-3-27b-it",
520
- name: "Google Gemma 3 27B Instruct",
484
+ "nvidia.nemotron-nano-12b-v2": {
485
+ id: "nvidia.nemotron-nano-12b-v2",
486
+ name: "NVIDIA Nemotron Nano 12B v2 VL BF16",
521
487
  api: "bedrock-converse-stream",
522
488
  provider: "amazon-bedrock",
523
489
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
524
490
  reasoning: false,
525
491
  input: ["text", "image"],
526
492
  cost: {
527
- input: 0.12,
528
- output: 0.2,
493
+ input: 0.2,
494
+ output: 0.6,
529
495
  cacheRead: 0,
530
496
  cacheWrite: 0,
531
497
  },
532
- contextWindow: 202752,
533
- maxTokens: 8192,
498
+ contextWindow: 128000,
499
+ maxTokens: 4096,
534
500
  } satisfies Model<"bedrock-converse-stream">,
535
- "us.google.gemma-3-4b-it": {
536
- id: "us.google.gemma-3-4b-it",
537
- name: "Gemma 3 4B IT",
501
+ "nvidia.nemotron-nano-9b-v2": {
502
+ id: "nvidia.nemotron-nano-9b-v2",
503
+ name: "NVIDIA Nemotron Nano 9B v2",
538
504
  api: "bedrock-converse-stream",
539
505
  provider: "amazon-bedrock",
540
506
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
541
507
  reasoning: false,
542
- input: ["text", "image"],
508
+ input: ["text"],
543
509
  cost: {
544
- input: 0.04,
545
- output: 0.08,
510
+ input: 0.06,
511
+ output: 0.23,
546
512
  cacheRead: 0,
547
513
  cacheWrite: 0,
548
514
  },
549
515
  contextWindow: 128000,
550
516
  maxTokens: 4096,
551
517
  } satisfies Model<"bedrock-converse-stream">,
552
- "us.meta.llama3-1-70b-instruct-v1:0": {
553
- id: "us.meta.llama3-1-70b-instruct-v1:0",
554
- name: "Llama 3.1 70B Instruct",
518
+ "openai.gpt-oss-120b-1:0": {
519
+ id: "openai.gpt-oss-120b-1:0",
520
+ name: "gpt-oss-120b",
555
521
  api: "bedrock-converse-stream",
556
522
  provider: "amazon-bedrock",
557
523
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
558
524
  reasoning: false,
559
525
  input: ["text"],
560
526
  cost: {
561
- input: 0.72,
562
- output: 0.72,
527
+ input: 0.15,
528
+ output: 0.6,
563
529
  cacheRead: 0,
564
530
  cacheWrite: 0,
565
531
  },
566
532
  contextWindow: 128000,
567
533
  maxTokens: 4096,
568
534
  } satisfies Model<"bedrock-converse-stream">,
569
- "us.meta.llama3-1-8b-instruct-v1:0": {
570
- id: "us.meta.llama3-1-8b-instruct-v1:0",
571
- name: "Llama 3.1 8B Instruct",
535
+ "openai.gpt-oss-20b-1:0": {
536
+ id: "openai.gpt-oss-20b-1:0",
537
+ name: "gpt-oss-20b",
572
538
  api: "bedrock-converse-stream",
573
539
  provider: "amazon-bedrock",
574
540
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
575
541
  reasoning: false,
576
542
  input: ["text"],
577
543
  cost: {
578
- input: 0.22,
579
- output: 0.22,
544
+ input: 0.07,
545
+ output: 0.3,
580
546
  cacheRead: 0,
581
547
  cacheWrite: 0,
582
548
  },
583
549
  contextWindow: 128000,
584
550
  maxTokens: 4096,
585
551
  } satisfies Model<"bedrock-converse-stream">,
586
- "us.meta.llama3-2-11b-instruct-v1:0": {
587
- id: "us.meta.llama3-2-11b-instruct-v1:0",
588
- name: "Llama 3.2 11B Instruct",
552
+ "openai.gpt-oss-safeguard-120b": {
553
+ id: "openai.gpt-oss-safeguard-120b",
554
+ name: "GPT OSS Safeguard 120B",
589
555
  api: "bedrock-converse-stream",
590
556
  provider: "amazon-bedrock",
591
557
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
592
558
  reasoning: false,
593
- input: ["text", "image"],
559
+ input: ["text"],
594
560
  cost: {
595
- input: 0.16,
596
- output: 0.16,
561
+ input: 0.15,
562
+ output: 0.6,
597
563
  cacheRead: 0,
598
564
  cacheWrite: 0,
599
565
  },
600
566
  contextWindow: 128000,
601
567
  maxTokens: 4096,
602
568
  } satisfies Model<"bedrock-converse-stream">,
603
- "us.meta.llama3-2-1b-instruct-v1:0": {
604
- id: "us.meta.llama3-2-1b-instruct-v1:0",
605
- name: "Llama 3.2 1B Instruct",
569
+ "openai.gpt-oss-safeguard-20b": {
570
+ id: "openai.gpt-oss-safeguard-20b",
571
+ name: "GPT OSS Safeguard 20B",
606
572
  api: "bedrock-converse-stream",
607
573
  provider: "amazon-bedrock",
608
574
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
609
575
  reasoning: false,
610
576
  input: ["text"],
611
577
  cost: {
612
- input: 0.1,
613
- output: 0.1,
578
+ input: 0.07,
579
+ output: 0.2,
614
580
  cacheRead: 0,
615
581
  cacheWrite: 0,
616
582
  },
617
- contextWindow: 131000,
583
+ contextWindow: 128000,
618
584
  maxTokens: 4096,
619
585
  } satisfies Model<"bedrock-converse-stream">,
620
- "us.meta.llama3-2-3b-instruct-v1:0": {
621
- id: "us.meta.llama3-2-3b-instruct-v1:0",
622
- name: "Llama 3.2 3B Instruct",
586
+ "qwen.qwen3-235b-a22b-2507-v1:0": {
587
+ id: "qwen.qwen3-235b-a22b-2507-v1:0",
588
+ name: "Qwen3 235B A22B 2507",
623
589
  api: "bedrock-converse-stream",
624
590
  provider: "amazon-bedrock",
625
591
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
626
592
  reasoning: false,
627
593
  input: ["text"],
628
594
  cost: {
629
- input: 0.15,
630
- output: 0.15,
595
+ input: 0.22,
596
+ output: 0.88,
631
597
  cacheRead: 0,
632
598
  cacheWrite: 0,
633
599
  },
634
- contextWindow: 131000,
635
- maxTokens: 4096,
600
+ contextWindow: 262144,
601
+ maxTokens: 131072,
636
602
  } satisfies Model<"bedrock-converse-stream">,
637
- "us.meta.llama3-2-90b-instruct-v1:0": {
638
- id: "us.meta.llama3-2-90b-instruct-v1:0",
639
- name: "Llama 3.2 90B Instruct",
603
+ "qwen.qwen3-32b-v1:0": {
604
+ id: "qwen.qwen3-32b-v1:0",
605
+ name: "Qwen3 32B (dense)",
640
606
  api: "bedrock-converse-stream",
641
607
  provider: "amazon-bedrock",
642
608
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
643
- reasoning: false,
644
- input: ["text", "image"],
609
+ reasoning: true,
610
+ input: ["text"],
645
611
  cost: {
646
- input: 0.72,
647
- output: 0.72,
612
+ input: 0.15,
613
+ output: 0.6,
648
614
  cacheRead: 0,
649
615
  cacheWrite: 0,
650
616
  },
651
- contextWindow: 128000,
652
- maxTokens: 4096,
617
+ contextWindow: 16384,
618
+ maxTokens: 16384,
653
619
  } satisfies Model<"bedrock-converse-stream">,
654
- "us.meta.llama3-3-70b-instruct-v1:0": {
655
- id: "us.meta.llama3-3-70b-instruct-v1:0",
656
- name: "Llama 3.3 70B Instruct",
620
+ "qwen.qwen3-coder-30b-a3b-v1:0": {
621
+ id: "qwen.qwen3-coder-30b-a3b-v1:0",
622
+ name: "Qwen3 Coder 30B A3B Instruct",
657
623
  api: "bedrock-converse-stream",
658
624
  provider: "amazon-bedrock",
659
625
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
660
626
  reasoning: false,
661
627
  input: ["text"],
662
628
  cost: {
663
- input: 0.72,
664
- output: 0.72,
629
+ input: 0.15,
630
+ output: 0.6,
665
631
  cacheRead: 0,
666
632
  cacheWrite: 0,
667
633
  },
668
- contextWindow: 128000,
669
- maxTokens: 4096,
634
+ contextWindow: 262144,
635
+ maxTokens: 131072,
670
636
  } satisfies Model<"bedrock-converse-stream">,
671
- "us.meta.llama4-maverick-17b-instruct-v1:0": {
672
- id: "us.meta.llama4-maverick-17b-instruct-v1:0",
673
- name: "Llama 4 Maverick 17B Instruct",
637
+ "qwen.qwen3-coder-480b-a35b-v1:0": {
638
+ id: "qwen.qwen3-coder-480b-a35b-v1:0",
639
+ name: "Qwen3 Coder 480B A35B Instruct",
674
640
  api: "bedrock-converse-stream",
675
641
  provider: "amazon-bedrock",
676
642
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
677
643
  reasoning: false,
678
- input: ["text", "image"],
644
+ input: ["text"],
679
645
  cost: {
680
- input: 0.24,
681
- output: 0.97,
646
+ input: 0.22,
647
+ output: 1.8,
682
648
  cacheRead: 0,
683
649
  cacheWrite: 0,
684
650
  },
685
- contextWindow: 1000000,
686
- maxTokens: 16384,
651
+ contextWindow: 131072,
652
+ maxTokens: 65536,
687
653
  } satisfies Model<"bedrock-converse-stream">,
688
- "us.meta.llama4-scout-17b-instruct-v1:0": {
689
- id: "us.meta.llama4-scout-17b-instruct-v1:0",
690
- name: "Llama 4 Scout 17B Instruct",
654
+ "qwen.qwen3-next-80b-a3b": {
655
+ id: "qwen.qwen3-next-80b-a3b",
656
+ name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
691
657
  api: "bedrock-converse-stream",
692
658
  provider: "amazon-bedrock",
693
659
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
694
660
  reasoning: false,
695
- input: ["text", "image"],
661
+ input: ["text"],
696
662
  cost: {
697
- input: 0.17,
698
- output: 0.66,
663
+ input: 0.14,
664
+ output: 1.4,
699
665
  cacheRead: 0,
700
666
  cacheWrite: 0,
701
667
  },
702
- contextWindow: 3500000,
703
- maxTokens: 16384,
668
+ contextWindow: 262000,
669
+ maxTokens: 262000,
704
670
  } satisfies Model<"bedrock-converse-stream">,
705
- "us.minimax.minimax-m2": {
706
- id: "us.minimax.minimax-m2",
707
- name: "MiniMax M2",
708
- api: "bedrock-converse-stream",
709
- provider: "amazon-bedrock",
710
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
711
- reasoning: true,
712
- input: ["text"],
713
- cost: {
714
- input: 0.3,
715
- output: 1.2,
716
- cacheRead: 0,
717
- cacheWrite: 0,
718
- },
719
- contextWindow: 204608,
720
- maxTokens: 128000,
721
- } satisfies Model<"bedrock-converse-stream">,
722
- "us.mistral.ministral-3-14b-instruct": {
723
- id: "us.mistral.ministral-3-14b-instruct",
724
- name: "Ministral 14B 3.0",
671
+ "qwen.qwen3-vl-235b-a22b": {
672
+ id: "qwen.qwen3-vl-235b-a22b",
673
+ name: "Qwen/Qwen3-VL-235B-A22B-Instruct",
725
674
  api: "bedrock-converse-stream",
726
675
  provider: "amazon-bedrock",
727
676
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
728
677
  reasoning: false,
729
- input: ["text"],
678
+ input: ["text", "image"],
730
679
  cost: {
731
- input: 0.2,
732
- output: 0.2,
680
+ input: 0.3,
681
+ output: 1.5,
733
682
  cacheRead: 0,
734
683
  cacheWrite: 0,
735
684
  },
736
- contextWindow: 128000,
737
- maxTokens: 4096,
685
+ contextWindow: 262000,
686
+ maxTokens: 262000,
738
687
  } satisfies Model<"bedrock-converse-stream">,
739
- "us.mistral.ministral-3-8b-instruct": {
740
- id: "us.mistral.ministral-3-8b-instruct",
741
- name: "Ministral 3 8B",
688
+ "us.amazon.nova-lite-v1:0": {
689
+ id: "us.amazon.nova-lite-v1:0",
690
+ name: "Nova Lite",
742
691
  api: "bedrock-converse-stream",
743
692
  provider: "amazon-bedrock",
744
693
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
745
694
  reasoning: false,
746
- input: ["text"],
695
+ input: ["text", "image"],
747
696
  cost: {
748
- input: 0.15,
749
- output: 0.15,
750
- cacheRead: 0,
697
+ input: 0.06,
698
+ output: 0.24,
699
+ cacheRead: 0.015,
751
700
  cacheWrite: 0,
752
701
  },
753
- contextWindow: 128000,
754
- maxTokens: 4096,
702
+ contextWindow: 300000,
703
+ maxTokens: 8192,
755
704
  } satisfies Model<"bedrock-converse-stream">,
756
- "us.mistral.mistral-7b-instruct-v0:2": {
757
- id: "us.mistral.mistral-7b-instruct-v0:2",
758
- name: "Mistral-7B-Instruct-v0.3",
705
+ "us.amazon.nova-micro-v1:0": {
706
+ id: "us.amazon.nova-micro-v1:0",
707
+ name: "Nova Micro",
759
708
  api: "bedrock-converse-stream",
760
709
  provider: "amazon-bedrock",
761
710
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
762
711
  reasoning: false,
763
712
  input: ["text"],
764
713
  cost: {
765
- input: 0.11,
766
- output: 0.11,
767
- cacheRead: 0,
714
+ input: 0.035,
715
+ output: 0.14,
716
+ cacheRead: 0.00875,
768
717
  cacheWrite: 0,
769
718
  },
770
- contextWindow: 127000,
771
- maxTokens: 127000,
719
+ contextWindow: 128000,
720
+ maxTokens: 8192,
772
721
  } satisfies Model<"bedrock-converse-stream">,
773
- "us.mistral.mistral-large-2402-v1:0": {
774
- id: "us.mistral.mistral-large-2402-v1:0",
775
- name: "Mistral Large (24.02)",
722
+ "us.amazon.nova-premier-v1:0": {
723
+ id: "us.amazon.nova-premier-v1:0",
724
+ name: "Nova Premier",
776
725
  api: "bedrock-converse-stream",
777
726
  provider: "amazon-bedrock",
778
727
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
779
- reasoning: false,
780
- input: ["text"],
728
+ reasoning: true,
729
+ input: ["text", "image"],
781
730
  cost: {
782
- input: 0.5,
783
- output: 1.5,
731
+ input: 2.5,
732
+ output: 12.5,
784
733
  cacheRead: 0,
785
734
  cacheWrite: 0,
786
735
  },
787
- contextWindow: 128000,
788
- maxTokens: 4096,
736
+ contextWindow: 1000000,
737
+ maxTokens: 16384,
789
738
  } satisfies Model<"bedrock-converse-stream">,
790
- "us.mistral.voxtral-mini-3b-2507": {
791
- id: "us.mistral.voxtral-mini-3b-2507",
792
- name: "Voxtral Mini 3B 2507",
739
+ "us.amazon.nova-pro-v1:0": {
740
+ id: "us.amazon.nova-pro-v1:0",
741
+ name: "Nova Pro",
793
742
  api: "bedrock-converse-stream",
794
743
  provider: "amazon-bedrock",
795
744
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
796
745
  reasoning: false,
797
- input: ["text"],
746
+ input: ["text", "image"],
798
747
  cost: {
799
- input: 0.04,
800
- output: 0.04,
801
- cacheRead: 0,
748
+ input: 0.8,
749
+ output: 3.2,
750
+ cacheRead: 0.2,
802
751
  cacheWrite: 0,
803
752
  },
804
- contextWindow: 128000,
805
- maxTokens: 4096,
753
+ contextWindow: 300000,
754
+ maxTokens: 8192,
806
755
  } satisfies Model<"bedrock-converse-stream">,
807
- "us.mistral.voxtral-small-24b-2507": {
808
- id: "us.mistral.voxtral-small-24b-2507",
809
- name: "Voxtral Small 24B 2507",
756
+ "us.anthropic.claude-3-7-sonnet-20250219-v1:0": {
757
+ id: "us.anthropic.claude-3-7-sonnet-20250219-v1:0",
758
+ name: "Claude Sonnet 3.7",
810
759
  api: "bedrock-converse-stream",
811
760
  provider: "amazon-bedrock",
812
761
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
813
762
  reasoning: false,
814
- input: ["text"],
763
+ input: ["text", "image"],
815
764
  cost: {
816
- input: 0.15,
817
- output: 0.35,
818
- cacheRead: 0,
819
- cacheWrite: 0,
765
+ input: 3,
766
+ output: 15,
767
+ cacheRead: 0.3,
768
+ cacheWrite: 3.75,
820
769
  },
821
- contextWindow: 32000,
770
+ contextWindow: 200000,
822
771
  maxTokens: 8192,
823
772
  } satisfies Model<"bedrock-converse-stream">,
824
- "us.moonshot.kimi-k2-thinking": {
825
- id: "us.moonshot.kimi-k2-thinking",
826
- name: "Kimi K2 Thinking",
773
+ "us.anthropic.claude-opus-4-1-20250805-v1:0": {
774
+ id: "us.anthropic.claude-opus-4-1-20250805-v1:0",
775
+ name: "Claude Opus 4.1",
827
776
  api: "bedrock-converse-stream",
828
777
  provider: "amazon-bedrock",
829
778
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
830
779
  reasoning: true,
831
- input: ["text"],
780
+ input: ["text", "image"],
832
781
  cost: {
833
- input: 0.6,
834
- output: 2.5,
835
- cacheRead: 0,
836
- cacheWrite: 0,
782
+ input: 15,
783
+ output: 75,
784
+ cacheRead: 1.5,
785
+ cacheWrite: 18.75,
837
786
  },
838
- contextWindow: 256000,
839
- maxTokens: 256000,
787
+ contextWindow: 200000,
788
+ maxTokens: 32000,
840
789
  } satisfies Model<"bedrock-converse-stream">,
841
- "us.nvidia.nemotron-nano-12b-v2": {
842
- id: "us.nvidia.nemotron-nano-12b-v2",
843
- name: "NVIDIA Nemotron Nano 12B v2 VL BF16",
790
+ "us.anthropic.claude-opus-4-20250514-v1:0": {
791
+ id: "us.anthropic.claude-opus-4-20250514-v1:0",
792
+ name: "Claude Opus 4",
844
793
  api: "bedrock-converse-stream",
845
794
  provider: "amazon-bedrock",
846
795
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
847
- reasoning: false,
796
+ reasoning: true,
848
797
  input: ["text", "image"],
849
798
  cost: {
850
- input: 0.2,
851
- output: 0.6,
852
- cacheRead: 0,
853
- cacheWrite: 0,
799
+ input: 15,
800
+ output: 75,
801
+ cacheRead: 1.5,
802
+ cacheWrite: 18.75,
854
803
  },
855
- contextWindow: 128000,
856
- maxTokens: 4096,
804
+ contextWindow: 200000,
805
+ maxTokens: 32000,
857
806
  } satisfies Model<"bedrock-converse-stream">,
858
- "us.nvidia.nemotron-nano-9b-v2": {
859
- id: "us.nvidia.nemotron-nano-9b-v2",
860
- name: "NVIDIA Nemotron Nano 9B v2",
807
+ "us.deepseek.r1-v1:0": {
808
+ id: "us.deepseek.r1-v1:0",
809
+ name: "DeepSeek-R1",
861
810
  api: "bedrock-converse-stream",
862
811
  provider: "amazon-bedrock",
863
812
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
864
- reasoning: false,
813
+ reasoning: true,
865
814
  input: ["text"],
866
815
  cost: {
867
- input: 0.06,
868
- output: 0.23,
816
+ input: 1.35,
817
+ output: 5.4,
869
818
  cacheRead: 0,
870
819
  cacheWrite: 0,
871
820
  },
872
821
  contextWindow: 128000,
873
- maxTokens: 4096,
822
+ maxTokens: 32768,
874
823
  } satisfies Model<"bedrock-converse-stream">,
875
- "us.openai.gpt-oss-120b-1:0": {
876
- id: "us.openai.gpt-oss-120b-1:0",
877
- name: "gpt-oss-120b",
824
+ "us.meta.llama3-2-11b-instruct-v1:0": {
825
+ id: "us.meta.llama3-2-11b-instruct-v1:0",
826
+ name: "Llama 3.2 11B Instruct",
878
827
  api: "bedrock-converse-stream",
879
828
  provider: "amazon-bedrock",
880
829
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
881
830
  reasoning: false,
882
- input: ["text"],
831
+ input: ["text", "image"],
883
832
  cost: {
884
- input: 0.15,
885
- output: 0.6,
833
+ input: 0.16,
834
+ output: 0.16,
886
835
  cacheRead: 0,
887
836
  cacheWrite: 0,
888
837
  },
889
838
  contextWindow: 128000,
890
839
  maxTokens: 4096,
891
840
  } satisfies Model<"bedrock-converse-stream">,
892
- "us.openai.gpt-oss-20b-1:0": {
893
- id: "us.openai.gpt-oss-20b-1:0",
894
- name: "gpt-oss-20b",
841
+ "us.meta.llama3-2-1b-instruct-v1:0": {
842
+ id: "us.meta.llama3-2-1b-instruct-v1:0",
843
+ name: "Llama 3.2 1B Instruct",
895
844
  api: "bedrock-converse-stream",
896
845
  provider: "amazon-bedrock",
897
846
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
898
847
  reasoning: false,
899
848
  input: ["text"],
900
849
  cost: {
901
- input: 0.07,
902
- output: 0.3,
850
+ input: 0.1,
851
+ output: 0.1,
903
852
  cacheRead: 0,
904
853
  cacheWrite: 0,
905
854
  },
906
- contextWindow: 128000,
855
+ contextWindow: 131000,
907
856
  maxTokens: 4096,
908
857
  } satisfies Model<"bedrock-converse-stream">,
909
- "us.openai.gpt-oss-safeguard-120b": {
910
- id: "us.openai.gpt-oss-safeguard-120b",
911
- name: "GPT OSS Safeguard 120B",
858
+ "us.meta.llama3-2-3b-instruct-v1:0": {
859
+ id: "us.meta.llama3-2-3b-instruct-v1:0",
860
+ name: "Llama 3.2 3B Instruct",
912
861
  api: "bedrock-converse-stream",
913
862
  provider: "amazon-bedrock",
914
863
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -916,147 +865,96 @@ export const MODELS = {
916
865
  input: ["text"],
917
866
  cost: {
918
867
  input: 0.15,
919
- output: 0.6,
868
+ output: 0.15,
920
869
  cacheRead: 0,
921
870
  cacheWrite: 0,
922
871
  },
923
- contextWindow: 128000,
872
+ contextWindow: 131000,
924
873
  maxTokens: 4096,
925
874
  } satisfies Model<"bedrock-converse-stream">,
926
- "us.openai.gpt-oss-safeguard-20b": {
927
- id: "us.openai.gpt-oss-safeguard-20b",
928
- name: "GPT OSS Safeguard 20B",
875
+ "us.meta.llama3-2-90b-instruct-v1:0": {
876
+ id: "us.meta.llama3-2-90b-instruct-v1:0",
877
+ name: "Llama 3.2 90B Instruct",
929
878
  api: "bedrock-converse-stream",
930
879
  provider: "amazon-bedrock",
931
880
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
932
881
  reasoning: false,
933
- input: ["text"],
882
+ input: ["text", "image"],
934
883
  cost: {
935
- input: 0.07,
936
- output: 0.2,
884
+ input: 0.72,
885
+ output: 0.72,
937
886
  cacheRead: 0,
938
887
  cacheWrite: 0,
939
888
  },
940
889
  contextWindow: 128000,
941
890
  maxTokens: 4096,
942
891
  } satisfies Model<"bedrock-converse-stream">,
943
- "us.qwen.qwen3-235b-a22b-2507-v1:0": {
944
- id: "us.qwen.qwen3-235b-a22b-2507-v1:0",
945
- name: "Qwen3 235B A22B 2507",
892
+ "us.meta.llama3-3-70b-instruct-v1:0": {
893
+ id: "us.meta.llama3-3-70b-instruct-v1:0",
894
+ name: "Llama 3.3 70B Instruct",
946
895
  api: "bedrock-converse-stream",
947
896
  provider: "amazon-bedrock",
948
897
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
949
898
  reasoning: false,
950
899
  input: ["text"],
951
900
  cost: {
952
- input: 0.22,
953
- output: 0.88,
901
+ input: 0.72,
902
+ output: 0.72,
954
903
  cacheRead: 0,
955
904
  cacheWrite: 0,
956
905
  },
957
- contextWindow: 262144,
958
- maxTokens: 131072,
906
+ contextWindow: 128000,
907
+ maxTokens: 4096,
959
908
  } satisfies Model<"bedrock-converse-stream">,
960
- "us.qwen.qwen3-32b-v1:0": {
961
- id: "us.qwen.qwen3-32b-v1:0",
962
- name: "Qwen3 32B (dense)",
909
+ "us.meta.llama4-maverick-17b-instruct-v1:0": {
910
+ id: "us.meta.llama4-maverick-17b-instruct-v1:0",
911
+ name: "Llama 4 Maverick 17B Instruct",
963
912
  api: "bedrock-converse-stream",
964
913
  provider: "amazon-bedrock",
965
914
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
966
- reasoning: true,
967
- input: ["text"],
915
+ reasoning: false,
916
+ input: ["text", "image"],
968
917
  cost: {
969
- input: 0.15,
970
- output: 0.6,
918
+ input: 0.24,
919
+ output: 0.97,
971
920
  cacheRead: 0,
972
921
  cacheWrite: 0,
973
922
  },
974
- contextWindow: 16384,
923
+ contextWindow: 1000000,
975
924
  maxTokens: 16384,
976
925
  } satisfies Model<"bedrock-converse-stream">,
977
- "us.qwen.qwen3-coder-30b-a3b-v1:0": {
978
- id: "us.qwen.qwen3-coder-30b-a3b-v1:0",
979
- name: "Qwen3 Coder 30B A3B Instruct",
926
+ "us.meta.llama4-scout-17b-instruct-v1:0": {
927
+ id: "us.meta.llama4-scout-17b-instruct-v1:0",
928
+ name: "Llama 4 Scout 17B Instruct",
980
929
  api: "bedrock-converse-stream",
981
930
  provider: "amazon-bedrock",
982
931
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
983
932
  reasoning: false,
984
- input: ["text"],
933
+ input: ["text", "image"],
985
934
  cost: {
986
- input: 0.15,
987
- output: 0.6,
935
+ input: 0.17,
936
+ output: 0.66,
988
937
  cacheRead: 0,
989
938
  cacheWrite: 0,
990
939
  },
991
- contextWindow: 262144,
992
- maxTokens: 131072,
940
+ contextWindow: 3500000,
941
+ maxTokens: 16384,
993
942
  } satisfies Model<"bedrock-converse-stream">,
994
- "us.qwen.qwen3-coder-480b-a35b-v1:0": {
995
- id: "us.qwen.qwen3-coder-480b-a35b-v1:0",
996
- name: "Qwen3 Coder 480B A35B Instruct",
997
- api: "bedrock-converse-stream",
998
- provider: "amazon-bedrock",
999
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
943
+ },
944
+ "anthropic": {
945
+ "claude-3-5-haiku-20241022": {
946
+ id: "claude-3-5-haiku-20241022",
947
+ name: "Claude Haiku 3.5",
948
+ api: "anthropic-messages",
949
+ provider: "anthropic",
950
+ baseUrl: "https://api.anthropic.com",
1000
951
  reasoning: false,
1001
- input: ["text"],
952
+ input: ["text", "image"],
1002
953
  cost: {
1003
- input: 0.22,
1004
- output: 1.8,
1005
- cacheRead: 0,
1006
- cacheWrite: 0,
1007
- },
1008
- contextWindow: 131072,
1009
- maxTokens: 65536,
1010
- } satisfies Model<"bedrock-converse-stream">,
1011
- "us.qwen.qwen3-next-80b-a3b": {
1012
- id: "us.qwen.qwen3-next-80b-a3b",
1013
- name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
1014
- api: "bedrock-converse-stream",
1015
- provider: "amazon-bedrock",
1016
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1017
- reasoning: false,
1018
- input: ["text"],
1019
- cost: {
1020
- input: 0.14,
1021
- output: 1.4,
1022
- cacheRead: 0,
1023
- cacheWrite: 0,
1024
- },
1025
- contextWindow: 262000,
1026
- maxTokens: 262000,
1027
- } satisfies Model<"bedrock-converse-stream">,
1028
- "us.qwen.qwen3-vl-235b-a22b": {
1029
- id: "us.qwen.qwen3-vl-235b-a22b",
1030
- name: "Qwen/Qwen3-VL-235B-A22B-Instruct",
1031
- api: "bedrock-converse-stream",
1032
- provider: "amazon-bedrock",
1033
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1034
- reasoning: false,
1035
- input: ["text", "image"],
1036
- cost: {
1037
- input: 0.3,
1038
- output: 1.5,
1039
- cacheRead: 0,
1040
- cacheWrite: 0,
1041
- },
1042
- contextWindow: 262000,
1043
- maxTokens: 262000,
1044
- } satisfies Model<"bedrock-converse-stream">,
1045
- },
1046
- "anthropic": {
1047
- "claude-3-5-haiku-20241022": {
1048
- id: "claude-3-5-haiku-20241022",
1049
- name: "Claude Haiku 3.5",
1050
- api: "anthropic-messages",
1051
- provider: "anthropic",
1052
- baseUrl: "https://api.anthropic.com",
1053
- reasoning: false,
1054
- input: ["text", "image"],
1055
- cost: {
1056
- input: 0.8,
1057
- output: 4,
1058
- cacheRead: 0.08,
1059
- cacheWrite: 1,
954
+ input: 0.8,
955
+ output: 4,
956
+ cacheRead: 0.08,
957
+ cacheWrite: 1,
1060
958
  },
1061
959
  contextWindow: 200000,
1062
960
  maxTokens: 8192,
@@ -2011,6 +1909,24 @@ export const MODELS = {
2011
1909
  contextWindow: 128000,
2012
1910
  maxTokens: 64000,
2013
1911
  } satisfies Model<"openai-responses">,
1912
+ "gpt-5.2-codex": {
1913
+ id: "gpt-5.2-codex",
1914
+ name: "GPT-5.2-Codex",
1915
+ api: "openai-responses",
1916
+ provider: "github-copilot",
1917
+ baseUrl: "https://api.individual.githubcopilot.com",
1918
+ headers: {"User-Agent":"GitHubCopilotChat/0.35.0","Editor-Version":"vscode/1.107.0","Editor-Plugin-Version":"copilot-chat/0.35.0","Copilot-Integration-Id":"vscode-chat"},
1919
+ reasoning: true,
1920
+ input: ["text", "image"],
1921
+ cost: {
1922
+ input: 0,
1923
+ output: 0,
1924
+ cacheRead: 0,
1925
+ cacheWrite: 0,
1926
+ },
1927
+ contextWindow: 272000,
1928
+ maxTokens: 128000,
1929
+ } satisfies Model<"openai-responses">,
2014
1930
  "grok-code-fast-1": {
2015
1931
  id: "grok-code-fast-1",
2016
1932
  name: "Grok Code Fast 1",
@@ -3066,9 +2982,9 @@ export const MODELS = {
3066
2982
  "MiniMax-M2": {
3067
2983
  id: "MiniMax-M2",
3068
2984
  name: "MiniMax-M2",
3069
- api: "openai-completions",
2985
+ api: "anthropic-messages",
3070
2986
  provider: "minimax",
3071
- baseUrl: "https://api.minimax.chat/v1",
2987
+ baseUrl: "https://api.minimax.io/anthropic",
3072
2988
  reasoning: true,
3073
2989
  input: ["text"],
3074
2990
  cost: {
@@ -3079,13 +2995,13 @@ export const MODELS = {
3079
2995
  },
3080
2996
  contextWindow: 196608,
3081
2997
  maxTokens: 128000,
3082
- } satisfies Model<"openai-completions">,
2998
+ } satisfies Model<"anthropic-messages">,
3083
2999
  "MiniMax-M2.1": {
3084
3000
  id: "MiniMax-M2.1",
3085
3001
  name: "MiniMax-M2.1",
3086
- api: "openai-completions",
3002
+ api: "anthropic-messages",
3087
3003
  provider: "minimax",
3088
- baseUrl: "https://api.minimax.chat/v1",
3004
+ baseUrl: "https://api.minimax.io/anthropic",
3089
3005
  reasoning: true,
3090
3006
  input: ["text"],
3091
3007
  cost: {
@@ -3096,7 +3012,43 @@ export const MODELS = {
3096
3012
  },
3097
3013
  contextWindow: 204800,
3098
3014
  maxTokens: 131072,
3099
- } satisfies Model<"openai-completions">,
3015
+ } satisfies Model<"anthropic-messages">,
3016
+ },
3017
+ "minimax-cn": {
3018
+ "MiniMax-M2": {
3019
+ id: "MiniMax-M2",
3020
+ name: "MiniMax-M2",
3021
+ api: "anthropic-messages",
3022
+ provider: "minimax-cn",
3023
+ baseUrl: "https://api.minimaxi.com/anthropic",
3024
+ reasoning: true,
3025
+ input: ["text"],
3026
+ cost: {
3027
+ input: 0.3,
3028
+ output: 1.2,
3029
+ cacheRead: 0,
3030
+ cacheWrite: 0,
3031
+ },
3032
+ contextWindow: 196608,
3033
+ maxTokens: 128000,
3034
+ } satisfies Model<"anthropic-messages">,
3035
+ "MiniMax-M2.1": {
3036
+ id: "MiniMax-M2.1",
3037
+ name: "MiniMax-M2.1",
3038
+ api: "anthropic-messages",
3039
+ provider: "minimax-cn",
3040
+ baseUrl: "https://api.minimaxi.com/anthropic",
3041
+ reasoning: true,
3042
+ input: ["text"],
3043
+ cost: {
3044
+ input: 0.3,
3045
+ output: 1.2,
3046
+ cacheRead: 0,
3047
+ cacheWrite: 0,
3048
+ },
3049
+ contextWindow: 204800,
3050
+ maxTokens: 131072,
3051
+ } satisfies Model<"anthropic-messages">,
3100
3052
  },
3101
3053
  "mistral": {
3102
3054
  "codestral-latest": {
@@ -3934,6 +3886,23 @@ export const MODELS = {
3934
3886
  contextWindow: 128000,
3935
3887
  maxTokens: 16384,
3936
3888
  } satisfies Model<"openai-responses">,
3889
+ "gpt-5.2-codex": {
3890
+ id: "gpt-5.2-codex",
3891
+ name: "GPT-5.2 Codex",
3892
+ api: "openai-responses",
3893
+ provider: "openai",
3894
+ baseUrl: "https://api.openai.com/v1",
3895
+ reasoning: true,
3896
+ input: ["text", "image"],
3897
+ cost: {
3898
+ input: 1.75,
3899
+ output: 14,
3900
+ cacheRead: 0.175,
3901
+ cacheWrite: 0,
3902
+ },
3903
+ contextWindow: 400000,
3904
+ maxTokens: 128000,
3905
+ } satisfies Model<"openai-responses">,
3937
3906
  "gpt-5.2-pro": {
3938
3907
  id: "gpt-5.2-pro",
3939
3908
  name: "GPT-5.2 Pro",
@@ -4533,6 +4502,23 @@ export const MODELS = {
4533
4502
  contextWindow: 400000,
4534
4503
  maxTokens: 128000,
4535
4504
  } satisfies Model<"openai-responses">,
4505
+ "gpt-5.2-codex": {
4506
+ id: "gpt-5.2-codex",
4507
+ name: "GPT-5.2 Codex",
4508
+ api: "openai-responses",
4509
+ provider: "opencode",
4510
+ baseUrl: "https://opencode.ai/zen/v1",
4511
+ reasoning: true,
4512
+ input: ["text", "image"],
4513
+ cost: {
4514
+ input: 1.75,
4515
+ output: 14,
4516
+ cacheRead: 0.175,
4517
+ cacheWrite: 0,
4518
+ },
4519
+ contextWindow: 400000,
4520
+ maxTokens: 128000,
4521
+ } satisfies Model<"openai-responses">,
4536
4522
  "grok-code": {
4537
4523
  id: "grok-code",
4538
4524
  name: "Grok Code Fast 1",
@@ -4824,23 +4810,6 @@ export const MODELS = {
4824
4810
  contextWindow: 200000,
4825
4811
  maxTokens: 8192,
4826
4812
  } satisfies Model<"openai-completions">,
4827
- "anthropic/claude-3.5-haiku-20241022": {
4828
- id: "anthropic/claude-3.5-haiku-20241022",
4829
- name: "Anthropic: Claude 3.5 Haiku (2024-10-22)",
4830
- api: "openai-completions",
4831
- provider: "openrouter",
4832
- baseUrl: "https://openrouter.ai/api/v1",
4833
- reasoning: false,
4834
- input: ["text", "image"],
4835
- cost: {
4836
- input: 0.7999999999999999,
4837
- output: 4,
4838
- cacheRead: 0.08,
4839
- cacheWrite: 1,
4840
- },
4841
- contextWindow: 200000,
4842
- maxTokens: 8192,
4843
- } satisfies Model<"openai-completions">,
4844
4813
  "anthropic/claude-3.5-sonnet": {
4845
4814
  id: "anthropic/claude-3.5-sonnet",
4846
4815
  name: "Anthropic: Claude 3.5 Sonnet",
@@ -5470,23 +5439,6 @@ export const MODELS = {
5470
5439
  contextWindow: 1048576,
5471
5440
  maxTokens: 65536,
5472
5441
  } satisfies Model<"openai-completions">,
5473
- "google/gemini-2.5-flash-preview-09-2025": {
5474
- id: "google/gemini-2.5-flash-preview-09-2025",
5475
- name: "Google: Gemini 2.5 Flash Preview 09-2025",
5476
- api: "openai-completions",
5477
- provider: "openrouter",
5478
- baseUrl: "https://openrouter.ai/api/v1",
5479
- reasoning: true,
5480
- input: ["text", "image"],
5481
- cost: {
5482
- input: 0.3,
5483
- output: 2.5,
5484
- cacheRead: 0.075,
5485
- cacheWrite: 0.3833,
5486
- },
5487
- contextWindow: 1048576,
5488
- maxTokens: 65536,
5489
- } satisfies Model<"openai-completions">,
5490
5442
  "google/gemini-2.5-pro": {
5491
5443
  id: "google/gemini-2.5-pro",
5492
5444
  name: "Google: Gemini 2.5 Pro",
@@ -5742,23 +5694,6 @@ export const MODELS = {
5742
5694
  contextWindow: 16384,
5743
5695
  maxTokens: 16384,
5744
5696
  } satisfies Model<"openai-completions">,
5745
- "meta-llama/llama-3.2-3b-instruct": {
5746
- id: "meta-llama/llama-3.2-3b-instruct",
5747
- name: "Meta: Llama 3.2 3B Instruct",
5748
- api: "openai-completions",
5749
- provider: "openrouter",
5750
- baseUrl: "https://openrouter.ai/api/v1",
5751
- reasoning: false,
5752
- input: ["text"],
5753
- cost: {
5754
- input: 0.02,
5755
- output: 0.02,
5756
- cacheRead: 0,
5757
- cacheWrite: 0,
5758
- },
5759
- contextWindow: 131072,
5760
- maxTokens: 16384,
5761
- } satisfies Model<"openai-completions">,
5762
5697
  "meta-llama/llama-3.3-70b-instruct": {
5763
5698
  id: "meta-llama/llama-3.3-70b-instruct",
5764
5699
  name: "Meta: Llama 3.3 70B Instruct",
@@ -7221,6 +7156,23 @@ export const MODELS = {
7221
7156
  contextWindow: 128000,
7222
7157
  maxTokens: 16384,
7223
7158
  } satisfies Model<"openai-completions">,
7159
+ "openai/gpt-5.2-codex": {
7160
+ id: "openai/gpt-5.2-codex",
7161
+ name: "OpenAI: GPT-5.2-Codex",
7162
+ api: "openai-completions",
7163
+ provider: "openrouter",
7164
+ baseUrl: "https://openrouter.ai/api/v1",
7165
+ reasoning: true,
7166
+ input: ["text", "image"],
7167
+ cost: {
7168
+ input: 1.75,
7169
+ output: 14,
7170
+ cacheRead: 0.175,
7171
+ cacheWrite: 0,
7172
+ },
7173
+ contextWindow: 400000,
7174
+ maxTokens: 128000,
7175
+ } satisfies Model<"openai-completions">,
7224
7176
  "openai/gpt-5.2-pro": {
7225
7177
  id: "openai/gpt-5.2-pro",
7226
7178
  name: "OpenAI: GPT-5.2 Pro",
@@ -7825,13 +7777,13 @@ export const MODELS = {
7825
7777
  reasoning: true,
7826
7778
  input: ["text"],
7827
7779
  cost: {
7828
- input: 0.035,
7829
- output: 0.13799999999999998,
7780
+ input: 0.049999999999999996,
7781
+ output: 0.25,
7830
7782
  cacheRead: 0,
7831
7783
  cacheWrite: 0,
7832
7784
  },
7833
- contextWindow: 128000,
7834
- maxTokens: 20000,
7785
+ contextWindow: 32000,
7786
+ maxTokens: 8192,
7835
7787
  } satisfies Model<"openai-completions">,
7836
7788
  "qwen/qwen3-coder": {
7837
7789
  id: "qwen/qwen3-coder",
@@ -8394,6 +8346,23 @@ export const MODELS = {
8394
8346
  contextWindow: 256000,
8395
8347
  maxTokens: 10000,
8396
8348
  } satisfies Model<"openai-completions">,
8349
+ "xiaomi/mimo-v2-flash": {
8350
+ id: "xiaomi/mimo-v2-flash",
8351
+ name: "Xiaomi: MiMo-V2-Flash",
8352
+ api: "openai-completions",
8353
+ provider: "openrouter",
8354
+ baseUrl: "https://openrouter.ai/api/v1",
8355
+ reasoning: true,
8356
+ input: ["text"],
8357
+ cost: {
8358
+ input: 0.09999999999999999,
8359
+ output: 0.3,
8360
+ cacheRead: 0.02,
8361
+ cacheWrite: 0,
8362
+ },
8363
+ contextWindow: 262144,
8364
+ maxTokens: 32000,
8365
+ } satisfies Model<"openai-completions">,
8397
8366
  "xiaomi/mimo-v2-flash:free": {
8398
8367
  id: "xiaomi/mimo-v2-flash:free",
8399
8368
  name: "Xiaomi: MiMo-V2-Flash (free)",
@@ -8565,6 +8534,2116 @@ export const MODELS = {
8565
8534
  maxTokens: 65535,
8566
8535
  } satisfies Model<"openai-completions">,
8567
8536
  },
8537
+ "vercel-ai-gateway": {
8538
+ "alibaba/qwen-3-14b": {
8539
+ id: "alibaba/qwen-3-14b",
8540
+ name: "Qwen3-14B",
8541
+ api: "anthropic-messages",
8542
+ provider: "vercel-ai-gateway",
8543
+ baseUrl: "https://ai-gateway.vercel.sh",
8544
+ reasoning: true,
8545
+ input: ["text"],
8546
+ cost: {
8547
+ input: 0.06,
8548
+ output: 0.24,
8549
+ cacheRead: 0,
8550
+ cacheWrite: 0,
8551
+ },
8552
+ contextWindow: 40960,
8553
+ maxTokens: 16384,
8554
+ } satisfies Model<"anthropic-messages">,
8555
+ "alibaba/qwen-3-235b": {
8556
+ id: "alibaba/qwen-3-235b",
8557
+ name: "Qwen3 235B A22b Instruct 2507",
8558
+ api: "anthropic-messages",
8559
+ provider: "vercel-ai-gateway",
8560
+ baseUrl: "https://ai-gateway.vercel.sh",
8561
+ reasoning: false,
8562
+ input: ["text"],
8563
+ cost: {
8564
+ input: 0.071,
8565
+ output: 0.463,
8566
+ cacheRead: 0,
8567
+ cacheWrite: 0,
8568
+ },
8569
+ contextWindow: 40960,
8570
+ maxTokens: 16384,
8571
+ } satisfies Model<"anthropic-messages">,
8572
+ "alibaba/qwen-3-30b": {
8573
+ id: "alibaba/qwen-3-30b",
8574
+ name: "Qwen3-30B-A3B",
8575
+ api: "anthropic-messages",
8576
+ provider: "vercel-ai-gateway",
8577
+ baseUrl: "https://ai-gateway.vercel.sh",
8578
+ reasoning: true,
8579
+ input: ["text"],
8580
+ cost: {
8581
+ input: 0.08,
8582
+ output: 0.29,
8583
+ cacheRead: 0,
8584
+ cacheWrite: 0,
8585
+ },
8586
+ contextWindow: 40960,
8587
+ maxTokens: 16384,
8588
+ } satisfies Model<"anthropic-messages">,
8589
+ "alibaba/qwen-3-32b": {
8590
+ id: "alibaba/qwen-3-32b",
8591
+ name: "Qwen 3.32B",
8592
+ api: "anthropic-messages",
8593
+ provider: "vercel-ai-gateway",
8594
+ baseUrl: "https://ai-gateway.vercel.sh",
8595
+ reasoning: true,
8596
+ input: ["text"],
8597
+ cost: {
8598
+ input: 0.09999999999999999,
8599
+ output: 0.3,
8600
+ cacheRead: 0,
8601
+ cacheWrite: 0,
8602
+ },
8603
+ contextWindow: 40960,
8604
+ maxTokens: 16384,
8605
+ } satisfies Model<"anthropic-messages">,
8606
+ "alibaba/qwen3-235b-a22b-thinking": {
8607
+ id: "alibaba/qwen3-235b-a22b-thinking",
8608
+ name: "Qwen3 235B A22B Thinking 2507",
8609
+ api: "anthropic-messages",
8610
+ provider: "vercel-ai-gateway",
8611
+ baseUrl: "https://ai-gateway.vercel.sh",
8612
+ reasoning: true,
8613
+ input: ["text", "image"],
8614
+ cost: {
8615
+ input: 0.3,
8616
+ output: 2.9000000000000004,
8617
+ cacheRead: 0,
8618
+ cacheWrite: 0,
8619
+ },
8620
+ contextWindow: 262114,
8621
+ maxTokens: 262114,
8622
+ } satisfies Model<"anthropic-messages">,
8623
+ "alibaba/qwen3-coder": {
8624
+ id: "alibaba/qwen3-coder",
8625
+ name: "Qwen3 Coder 480B A35B Instruct",
8626
+ api: "anthropic-messages",
8627
+ provider: "vercel-ai-gateway",
8628
+ baseUrl: "https://ai-gateway.vercel.sh",
8629
+ reasoning: false,
8630
+ input: ["text"],
8631
+ cost: {
8632
+ input: 0.38,
8633
+ output: 1.53,
8634
+ cacheRead: 0,
8635
+ cacheWrite: 0,
8636
+ },
8637
+ contextWindow: 262144,
8638
+ maxTokens: 66536,
8639
+ } satisfies Model<"anthropic-messages">,
8640
+ "alibaba/qwen3-coder-30b-a3b": {
8641
+ id: "alibaba/qwen3-coder-30b-a3b",
8642
+ name: "Qwen 3 Coder 30B A3B Instruct",
8643
+ api: "anthropic-messages",
8644
+ provider: "vercel-ai-gateway",
8645
+ baseUrl: "https://ai-gateway.vercel.sh",
8646
+ reasoning: true,
8647
+ input: ["text"],
8648
+ cost: {
8649
+ input: 0.07,
8650
+ output: 0.27,
8651
+ cacheRead: 0,
8652
+ cacheWrite: 0,
8653
+ },
8654
+ contextWindow: 160000,
8655
+ maxTokens: 32768,
8656
+ } satisfies Model<"anthropic-messages">,
8657
+ "alibaba/qwen3-coder-plus": {
8658
+ id: "alibaba/qwen3-coder-plus",
8659
+ name: "Qwen3 Coder Plus",
8660
+ api: "anthropic-messages",
8661
+ provider: "vercel-ai-gateway",
8662
+ baseUrl: "https://ai-gateway.vercel.sh",
8663
+ reasoning: false,
8664
+ input: ["text"],
8665
+ cost: {
8666
+ input: 1,
8667
+ output: 5,
8668
+ cacheRead: 0,
8669
+ cacheWrite: 0,
8670
+ },
8671
+ contextWindow: 1000000,
8672
+ maxTokens: 65536,
8673
+ } satisfies Model<"anthropic-messages">,
8674
+ "alibaba/qwen3-max": {
8675
+ id: "alibaba/qwen3-max",
8676
+ name: "Qwen3 Max",
8677
+ api: "anthropic-messages",
8678
+ provider: "vercel-ai-gateway",
8679
+ baseUrl: "https://ai-gateway.vercel.sh",
8680
+ reasoning: false,
8681
+ input: ["text"],
8682
+ cost: {
8683
+ input: 1.2,
8684
+ output: 6,
8685
+ cacheRead: 0.24,
8686
+ cacheWrite: 0,
8687
+ },
8688
+ contextWindow: 262144,
8689
+ maxTokens: 32768,
8690
+ } satisfies Model<"anthropic-messages">,
8691
+ "alibaba/qwen3-max-preview": {
8692
+ id: "alibaba/qwen3-max-preview",
8693
+ name: "Qwen3 Max Preview",
8694
+ api: "anthropic-messages",
8695
+ provider: "vercel-ai-gateway",
8696
+ baseUrl: "https://ai-gateway.vercel.sh",
8697
+ reasoning: false,
8698
+ input: ["text"],
8699
+ cost: {
8700
+ input: 1.2,
8701
+ output: 6,
8702
+ cacheRead: 0.24,
8703
+ cacheWrite: 0,
8704
+ },
8705
+ contextWindow: 262144,
8706
+ maxTokens: 32768,
8707
+ } satisfies Model<"anthropic-messages">,
8708
+ "anthropic/claude-3-haiku": {
8709
+ id: "anthropic/claude-3-haiku",
8710
+ name: "Claude 3 Haiku",
8711
+ api: "anthropic-messages",
8712
+ provider: "vercel-ai-gateway",
8713
+ baseUrl: "https://ai-gateway.vercel.sh",
8714
+ reasoning: false,
8715
+ input: ["text", "image"],
8716
+ cost: {
8717
+ input: 0.25,
8718
+ output: 1.25,
8719
+ cacheRead: 0.03,
8720
+ cacheWrite: 0.3,
8721
+ },
8722
+ contextWindow: 200000,
8723
+ maxTokens: 4096,
8724
+ } satisfies Model<"anthropic-messages">,
8725
+ "anthropic/claude-3-opus": {
8726
+ id: "anthropic/claude-3-opus",
8727
+ name: "Claude 3 Opus",
8728
+ api: "anthropic-messages",
8729
+ provider: "vercel-ai-gateway",
8730
+ baseUrl: "https://ai-gateway.vercel.sh",
8731
+ reasoning: false,
8732
+ input: ["text", "image"],
8733
+ cost: {
8734
+ input: 15,
8735
+ output: 75,
8736
+ cacheRead: 0,
8737
+ cacheWrite: 0,
8738
+ },
8739
+ contextWindow: 200000,
8740
+ maxTokens: 8192,
8741
+ } satisfies Model<"anthropic-messages">,
8742
+ "anthropic/claude-3.5-haiku": {
8743
+ id: "anthropic/claude-3.5-haiku",
8744
+ name: "Claude 3.5 Haiku",
8745
+ api: "anthropic-messages",
8746
+ provider: "vercel-ai-gateway",
8747
+ baseUrl: "https://ai-gateway.vercel.sh",
8748
+ reasoning: false,
8749
+ input: ["text", "image"],
8750
+ cost: {
8751
+ input: 0.7999999999999999,
8752
+ output: 4,
8753
+ cacheRead: 0.08,
8754
+ cacheWrite: 1,
8755
+ },
8756
+ contextWindow: 200000,
8757
+ maxTokens: 8192,
8758
+ } satisfies Model<"anthropic-messages">,
8759
+ "anthropic/claude-3.5-sonnet": {
8760
+ id: "anthropic/claude-3.5-sonnet",
8761
+ name: "Claude 3.5 Sonnet",
8762
+ api: "anthropic-messages",
8763
+ provider: "vercel-ai-gateway",
8764
+ baseUrl: "https://ai-gateway.vercel.sh",
8765
+ reasoning: false,
8766
+ input: ["text", "image"],
8767
+ cost: {
8768
+ input: 3,
8769
+ output: 15,
8770
+ cacheRead: 0.3,
8771
+ cacheWrite: 3.75,
8772
+ },
8773
+ contextWindow: 200000,
8774
+ maxTokens: 8192,
8775
+ } satisfies Model<"anthropic-messages">,
8776
+ "anthropic/claude-3.5-sonnet-20240620": {
8777
+ id: "anthropic/claude-3.5-sonnet-20240620",
8778
+ name: "Claude 3.5 Sonnet (2024-06-20)",
8779
+ api: "anthropic-messages",
8780
+ provider: "vercel-ai-gateway",
8781
+ baseUrl: "https://ai-gateway.vercel.sh",
8782
+ reasoning: false,
8783
+ input: ["text", "image"],
8784
+ cost: {
8785
+ input: 3,
8786
+ output: 15,
8787
+ cacheRead: 0,
8788
+ cacheWrite: 0,
8789
+ },
8790
+ contextWindow: 200000,
8791
+ maxTokens: 8192,
8792
+ } satisfies Model<"anthropic-messages">,
8793
+ "anthropic/claude-3.7-sonnet": {
8794
+ id: "anthropic/claude-3.7-sonnet",
8795
+ name: "Claude 3.7 Sonnet",
8796
+ api: "anthropic-messages",
8797
+ provider: "vercel-ai-gateway",
8798
+ baseUrl: "https://ai-gateway.vercel.sh",
8799
+ reasoning: true,
8800
+ input: ["text", "image"],
8801
+ cost: {
8802
+ input: 3,
8803
+ output: 15,
8804
+ cacheRead: 0.3,
8805
+ cacheWrite: 3.75,
8806
+ },
8807
+ contextWindow: 200000,
8808
+ maxTokens: 64000,
8809
+ } satisfies Model<"anthropic-messages">,
8810
+ "anthropic/claude-haiku-4.5": {
8811
+ id: "anthropic/claude-haiku-4.5",
8812
+ name: "Claude Haiku 4.5",
8813
+ api: "anthropic-messages",
8814
+ provider: "vercel-ai-gateway",
8815
+ baseUrl: "https://ai-gateway.vercel.sh",
8816
+ reasoning: true,
8817
+ input: ["text", "image"],
8818
+ cost: {
8819
+ input: 1,
8820
+ output: 5,
8821
+ cacheRead: 0.09999999999999999,
8822
+ cacheWrite: 1.25,
8823
+ },
8824
+ contextWindow: 200000,
8825
+ maxTokens: 64000,
8826
+ } satisfies Model<"anthropic-messages">,
8827
+ "anthropic/claude-opus-4": {
8828
+ id: "anthropic/claude-opus-4",
8829
+ name: "Claude Opus 4",
8830
+ api: "anthropic-messages",
8831
+ provider: "vercel-ai-gateway",
8832
+ baseUrl: "https://ai-gateway.vercel.sh",
8833
+ reasoning: true,
8834
+ input: ["text", "image"],
8835
+ cost: {
8836
+ input: 15,
8837
+ output: 75,
8838
+ cacheRead: 1.5,
8839
+ cacheWrite: 18.75,
8840
+ },
8841
+ contextWindow: 200000,
8842
+ maxTokens: 32000,
8843
+ } satisfies Model<"anthropic-messages">,
8844
+ "anthropic/claude-opus-4.1": {
8845
+ id: "anthropic/claude-opus-4.1",
8846
+ name: "Claude Opus 4.1",
8847
+ api: "anthropic-messages",
8848
+ provider: "vercel-ai-gateway",
8849
+ baseUrl: "https://ai-gateway.vercel.sh",
8850
+ reasoning: true,
8851
+ input: ["text", "image"],
8852
+ cost: {
8853
+ input: 15,
8854
+ output: 75,
8855
+ cacheRead: 1.5,
8856
+ cacheWrite: 18.75,
8857
+ },
8858
+ contextWindow: 200000,
8859
+ maxTokens: 32000,
8860
+ } satisfies Model<"anthropic-messages">,
8861
+ "anthropic/claude-opus-4.5": {
8862
+ id: "anthropic/claude-opus-4.5",
8863
+ name: "Claude Opus 4.5",
8864
+ api: "anthropic-messages",
8865
+ provider: "vercel-ai-gateway",
8866
+ baseUrl: "https://ai-gateway.vercel.sh",
8867
+ reasoning: true,
8868
+ input: ["text", "image"],
8869
+ cost: {
8870
+ input: 5,
8871
+ output: 25,
8872
+ cacheRead: 0.5,
8873
+ cacheWrite: 6.25,
8874
+ },
8875
+ contextWindow: 200000,
8876
+ maxTokens: 64000,
8877
+ } satisfies Model<"anthropic-messages">,
8878
+ "anthropic/claude-sonnet-4": {
8879
+ id: "anthropic/claude-sonnet-4",
8880
+ name: "Claude Sonnet 4",
8881
+ api: "anthropic-messages",
8882
+ provider: "vercel-ai-gateway",
8883
+ baseUrl: "https://ai-gateway.vercel.sh",
8884
+ reasoning: true,
8885
+ input: ["text", "image"],
8886
+ cost: {
8887
+ input: 3,
8888
+ output: 15,
8889
+ cacheRead: 0.3,
8890
+ cacheWrite: 3.75,
8891
+ },
8892
+ contextWindow: 200000,
8893
+ maxTokens: 64000,
8894
+ } satisfies Model<"anthropic-messages">,
8895
+ "anthropic/claude-sonnet-4.5": {
8896
+ id: "anthropic/claude-sonnet-4.5",
8897
+ name: "Claude Sonnet 4.5",
8898
+ api: "anthropic-messages",
8899
+ provider: "vercel-ai-gateway",
8900
+ baseUrl: "https://ai-gateway.vercel.sh",
8901
+ reasoning: true,
8902
+ input: ["text", "image"],
8903
+ cost: {
8904
+ input: 3,
8905
+ output: 15,
8906
+ cacheRead: 0.3,
8907
+ cacheWrite: 3.75,
8908
+ },
8909
+ contextWindow: 200000,
8910
+ maxTokens: 64000,
8911
+ } satisfies Model<"anthropic-messages">,
8912
+ "bytedance/seed-1.6": {
8913
+ id: "bytedance/seed-1.6",
8914
+ name: "Seed 1.6",
8915
+ api: "anthropic-messages",
8916
+ provider: "vercel-ai-gateway",
8917
+ baseUrl: "https://ai-gateway.vercel.sh",
8918
+ reasoning: true,
8919
+ input: ["text"],
8920
+ cost: {
8921
+ input: 0.25,
8922
+ output: 2,
8923
+ cacheRead: 0.049999999999999996,
8924
+ cacheWrite: 0,
8925
+ },
8926
+ contextWindow: 256000,
8927
+ maxTokens: 32000,
8928
+ } satisfies Model<"anthropic-messages">,
8929
+ "cohere/command-a": {
8930
+ id: "cohere/command-a",
8931
+ name: "Command A",
8932
+ api: "anthropic-messages",
8933
+ provider: "vercel-ai-gateway",
8934
+ baseUrl: "https://ai-gateway.vercel.sh",
8935
+ reasoning: false,
8936
+ input: ["text"],
8937
+ cost: {
8938
+ input: 2.5,
8939
+ output: 10,
8940
+ cacheRead: 0,
8941
+ cacheWrite: 0,
8942
+ },
8943
+ contextWindow: 256000,
8944
+ maxTokens: 8000,
8945
+ } satisfies Model<"anthropic-messages">,
8946
+ "deepseek/deepseek-v3": {
8947
+ id: "deepseek/deepseek-v3",
8948
+ name: "DeepSeek V3 0324",
8949
+ api: "anthropic-messages",
8950
+ provider: "vercel-ai-gateway",
8951
+ baseUrl: "https://ai-gateway.vercel.sh",
8952
+ reasoning: false,
8953
+ input: ["text"],
8954
+ cost: {
8955
+ input: 0.77,
8956
+ output: 0.77,
8957
+ cacheRead: 0,
8958
+ cacheWrite: 0,
8959
+ },
8960
+ contextWindow: 163840,
8961
+ maxTokens: 16384,
8962
+ } satisfies Model<"anthropic-messages">,
8963
+ "deepseek/deepseek-v3.1": {
8964
+ id: "deepseek/deepseek-v3.1",
8965
+ name: "DeepSeek-V3.1",
8966
+ api: "anthropic-messages",
8967
+ provider: "vercel-ai-gateway",
8968
+ baseUrl: "https://ai-gateway.vercel.sh",
8969
+ reasoning: true,
8970
+ input: ["text"],
8971
+ cost: {
8972
+ input: 0.3,
8973
+ output: 1,
8974
+ cacheRead: 0,
8975
+ cacheWrite: 0,
8976
+ },
8977
+ contextWindow: 163840,
8978
+ maxTokens: 128000,
8979
+ } satisfies Model<"anthropic-messages">,
8980
+ "deepseek/deepseek-v3.1-terminus": {
8981
+ id: "deepseek/deepseek-v3.1-terminus",
8982
+ name: "DeepSeek V3.1 Terminus",
8983
+ api: "anthropic-messages",
8984
+ provider: "vercel-ai-gateway",
8985
+ baseUrl: "https://ai-gateway.vercel.sh",
8986
+ reasoning: true,
8987
+ input: ["text"],
8988
+ cost: {
8989
+ input: 0.27,
8990
+ output: 1,
8991
+ cacheRead: 0,
8992
+ cacheWrite: 0,
8993
+ },
8994
+ contextWindow: 131072,
8995
+ maxTokens: 65536,
8996
+ } satisfies Model<"anthropic-messages">,
8997
+ "deepseek/deepseek-v3.2-exp": {
8998
+ id: "deepseek/deepseek-v3.2-exp",
8999
+ name: "DeepSeek V3.2 Exp",
9000
+ api: "anthropic-messages",
9001
+ provider: "vercel-ai-gateway",
9002
+ baseUrl: "https://ai-gateway.vercel.sh",
9003
+ reasoning: true,
9004
+ input: ["text"],
9005
+ cost: {
9006
+ input: 0.27,
9007
+ output: 0.39999999999999997,
9008
+ cacheRead: 0,
9009
+ cacheWrite: 0,
9010
+ },
9011
+ contextWindow: 163840,
9012
+ maxTokens: 163840,
9013
+ } satisfies Model<"anthropic-messages">,
9014
+ "deepseek/deepseek-v3.2-thinking": {
9015
+ id: "deepseek/deepseek-v3.2-thinking",
9016
+ name: "DeepSeek V3.2 Thinking",
9017
+ api: "anthropic-messages",
9018
+ provider: "vercel-ai-gateway",
9019
+ baseUrl: "https://ai-gateway.vercel.sh",
9020
+ reasoning: true,
9021
+ input: ["text"],
9022
+ cost: {
9023
+ input: 0.28,
9024
+ output: 0.42,
9025
+ cacheRead: 0.028,
9026
+ cacheWrite: 0,
9027
+ },
9028
+ contextWindow: 128000,
9029
+ maxTokens: 64000,
9030
+ } satisfies Model<"anthropic-messages">,
9031
+ "google/gemini-2.0-flash": {
9032
+ id: "google/gemini-2.0-flash",
9033
+ name: "Gemini 2.0 Flash",
9034
+ api: "anthropic-messages",
9035
+ provider: "vercel-ai-gateway",
9036
+ baseUrl: "https://ai-gateway.vercel.sh",
9037
+ reasoning: false,
9038
+ input: ["text", "image"],
9039
+ cost: {
9040
+ input: 0.09999999999999999,
9041
+ output: 0.39999999999999997,
9042
+ cacheRead: 0.024999999999999998,
9043
+ cacheWrite: 0,
9044
+ },
9045
+ contextWindow: 1000000,
9046
+ maxTokens: 8192,
9047
+ } satisfies Model<"anthropic-messages">,
9048
+ "google/gemini-2.0-flash-lite": {
9049
+ id: "google/gemini-2.0-flash-lite",
9050
+ name: "Gemini 2.0 Flash Lite",
9051
+ api: "anthropic-messages",
9052
+ provider: "vercel-ai-gateway",
9053
+ baseUrl: "https://ai-gateway.vercel.sh",
9054
+ reasoning: false,
9055
+ input: ["text", "image"],
9056
+ cost: {
9057
+ input: 0.075,
9058
+ output: 0.3,
9059
+ cacheRead: 0,
9060
+ cacheWrite: 0,
9061
+ },
9062
+ contextWindow: 1048576,
9063
+ maxTokens: 8192,
9064
+ } satisfies Model<"anthropic-messages">,
9065
+ "google/gemini-2.5-flash": {
9066
+ id: "google/gemini-2.5-flash",
9067
+ name: "Gemini 2.5 Flash",
9068
+ api: "anthropic-messages",
9069
+ provider: "vercel-ai-gateway",
9070
+ baseUrl: "https://ai-gateway.vercel.sh",
9071
+ reasoning: true,
9072
+ input: ["text", "image"],
9073
+ cost: {
9074
+ input: 0.3,
9075
+ output: 2.5,
9076
+ cacheRead: 0.03,
9077
+ cacheWrite: 0,
9078
+ },
9079
+ contextWindow: 1000000,
9080
+ maxTokens: 64000,
9081
+ } satisfies Model<"anthropic-messages">,
9082
+ "google/gemini-2.5-flash-lite": {
9083
+ id: "google/gemini-2.5-flash-lite",
9084
+ name: "Gemini 2.5 Flash Lite",
9085
+ api: "anthropic-messages",
9086
+ provider: "vercel-ai-gateway",
9087
+ baseUrl: "https://ai-gateway.vercel.sh",
9088
+ reasoning: true,
9089
+ input: ["text", "image"],
9090
+ cost: {
9091
+ input: 0.09999999999999999,
9092
+ output: 0.39999999999999997,
9093
+ cacheRead: 0.01,
9094
+ cacheWrite: 0,
9095
+ },
9096
+ contextWindow: 1048576,
9097
+ maxTokens: 65536,
9098
+ } satisfies Model<"anthropic-messages">,
9099
+ "google/gemini-2.5-flash-lite-preview-09-2025": {
9100
+ id: "google/gemini-2.5-flash-lite-preview-09-2025",
9101
+ name: "Gemini 2.5 Flash Lite Preview 09-2025",
9102
+ api: "anthropic-messages",
9103
+ provider: "vercel-ai-gateway",
9104
+ baseUrl: "https://ai-gateway.vercel.sh",
9105
+ reasoning: true,
9106
+ input: ["text", "image"],
9107
+ cost: {
9108
+ input: 0.09999999999999999,
9109
+ output: 0.39999999999999997,
9110
+ cacheRead: 0.01,
9111
+ cacheWrite: 0,
9112
+ },
9113
+ contextWindow: 1048576,
9114
+ maxTokens: 65536,
9115
+ } satisfies Model<"anthropic-messages">,
9116
+ "google/gemini-2.5-flash-preview-09-2025": {
9117
+ id: "google/gemini-2.5-flash-preview-09-2025",
9118
+ name: "Gemini 2.5 Flash Preview 09-2025",
9119
+ api: "anthropic-messages",
9120
+ provider: "vercel-ai-gateway",
9121
+ baseUrl: "https://ai-gateway.vercel.sh",
9122
+ reasoning: true,
9123
+ input: ["text", "image"],
9124
+ cost: {
9125
+ input: 0.3,
9126
+ output: 2.5,
9127
+ cacheRead: 0.03,
9128
+ cacheWrite: 0,
9129
+ },
9130
+ contextWindow: 1000000,
9131
+ maxTokens: 65536,
9132
+ } satisfies Model<"anthropic-messages">,
9133
+ "google/gemini-2.5-pro": {
9134
+ id: "google/gemini-2.5-pro",
9135
+ name: "Gemini 2.5 Pro",
9136
+ api: "anthropic-messages",
9137
+ provider: "vercel-ai-gateway",
9138
+ baseUrl: "https://ai-gateway.vercel.sh",
9139
+ reasoning: true,
9140
+ input: ["text", "image"],
9141
+ cost: {
9142
+ input: 1.25,
9143
+ output: 10,
9144
+ cacheRead: 0.125,
9145
+ cacheWrite: 0,
9146
+ },
9147
+ contextWindow: 1048576,
9148
+ maxTokens: 65536,
9149
+ } satisfies Model<"anthropic-messages">,
9150
+ "google/gemini-3-flash": {
9151
+ id: "google/gemini-3-flash",
9152
+ name: "Gemini 3 Flash",
9153
+ api: "anthropic-messages",
9154
+ provider: "vercel-ai-gateway",
9155
+ baseUrl: "https://ai-gateway.vercel.sh",
9156
+ reasoning: true,
9157
+ input: ["text", "image"],
9158
+ cost: {
9159
+ input: 0.5,
9160
+ output: 3,
9161
+ cacheRead: 0.049999999999999996,
9162
+ cacheWrite: 0,
9163
+ },
9164
+ contextWindow: 1000000,
9165
+ maxTokens: 64000,
9166
+ } satisfies Model<"anthropic-messages">,
9167
+ "google/gemini-3-pro-preview": {
9168
+ id: "google/gemini-3-pro-preview",
9169
+ name: "Gemini 3 Pro Preview",
9170
+ api: "anthropic-messages",
9171
+ provider: "vercel-ai-gateway",
9172
+ baseUrl: "https://ai-gateway.vercel.sh",
9173
+ reasoning: true,
9174
+ input: ["text", "image"],
9175
+ cost: {
9176
+ input: 2,
9177
+ output: 12,
9178
+ cacheRead: 0.19999999999999998,
9179
+ cacheWrite: 0,
9180
+ },
9181
+ contextWindow: 1000000,
9182
+ maxTokens: 64000,
9183
+ } satisfies Model<"anthropic-messages">,
9184
+ "inception/mercury-coder-small": {
9185
+ id: "inception/mercury-coder-small",
9186
+ name: "Mercury Coder Small Beta",
9187
+ api: "anthropic-messages",
9188
+ provider: "vercel-ai-gateway",
9189
+ baseUrl: "https://ai-gateway.vercel.sh",
9190
+ reasoning: false,
9191
+ input: ["text"],
9192
+ cost: {
9193
+ input: 0.25,
9194
+ output: 1,
9195
+ cacheRead: 0,
9196
+ cacheWrite: 0,
9197
+ },
9198
+ contextWindow: 32000,
9199
+ maxTokens: 16384,
9200
+ } satisfies Model<"anthropic-messages">,
9201
+ "meituan/longcat-flash-chat": {
9202
+ id: "meituan/longcat-flash-chat",
9203
+ name: "LongCat Flash Chat",
9204
+ api: "anthropic-messages",
9205
+ provider: "vercel-ai-gateway",
9206
+ baseUrl: "https://ai-gateway.vercel.sh",
9207
+ reasoning: false,
9208
+ input: ["text"],
9209
+ cost: {
9210
+ input: 0,
9211
+ output: 0,
9212
+ cacheRead: 0,
9213
+ cacheWrite: 0,
9214
+ },
9215
+ contextWindow: 128000,
9216
+ maxTokens: 8192,
9217
+ } satisfies Model<"anthropic-messages">,
9218
+ "meituan/longcat-flash-thinking": {
9219
+ id: "meituan/longcat-flash-thinking",
9220
+ name: "LongCat Flash Thinking",
9221
+ api: "anthropic-messages",
9222
+ provider: "vercel-ai-gateway",
9223
+ baseUrl: "https://ai-gateway.vercel.sh",
9224
+ reasoning: true,
9225
+ input: ["text"],
9226
+ cost: {
9227
+ input: 0.15,
9228
+ output: 1.5,
9229
+ cacheRead: 0,
9230
+ cacheWrite: 0,
9231
+ },
9232
+ contextWindow: 128000,
9233
+ maxTokens: 8192,
9234
+ } satisfies Model<"anthropic-messages">,
9235
+ "meta/llama-3.1-70b": {
9236
+ id: "meta/llama-3.1-70b",
9237
+ name: "Llama 3.1 70B Instruct",
9238
+ api: "anthropic-messages",
9239
+ provider: "vercel-ai-gateway",
9240
+ baseUrl: "https://ai-gateway.vercel.sh",
9241
+ reasoning: false,
9242
+ input: ["text"],
9243
+ cost: {
9244
+ input: 0.39999999999999997,
9245
+ output: 0.39999999999999997,
9246
+ cacheRead: 0,
9247
+ cacheWrite: 0,
9248
+ },
9249
+ contextWindow: 131072,
9250
+ maxTokens: 16384,
9251
+ } satisfies Model<"anthropic-messages">,
9252
+ "meta/llama-3.1-8b": {
9253
+ id: "meta/llama-3.1-8b",
9254
+ name: "Llama 3.1 8B Instruct",
9255
+ api: "anthropic-messages",
9256
+ provider: "vercel-ai-gateway",
9257
+ baseUrl: "https://ai-gateway.vercel.sh",
9258
+ reasoning: false,
9259
+ input: ["text"],
9260
+ cost: {
9261
+ input: 0.03,
9262
+ output: 0.049999999999999996,
9263
+ cacheRead: 0,
9264
+ cacheWrite: 0,
9265
+ },
9266
+ contextWindow: 131072,
9267
+ maxTokens: 16384,
9268
+ } satisfies Model<"anthropic-messages">,
9269
+ "meta/llama-3.2-11b": {
9270
+ id: "meta/llama-3.2-11b",
9271
+ name: "Llama 3.2 11B Vision Instruct",
9272
+ api: "anthropic-messages",
9273
+ provider: "vercel-ai-gateway",
9274
+ baseUrl: "https://ai-gateway.vercel.sh",
9275
+ reasoning: false,
9276
+ input: ["text", "image"],
9277
+ cost: {
9278
+ input: 0.16,
9279
+ output: 0.16,
9280
+ cacheRead: 0,
9281
+ cacheWrite: 0,
9282
+ },
9283
+ contextWindow: 128000,
9284
+ maxTokens: 8192,
9285
+ } satisfies Model<"anthropic-messages">,
9286
+ "meta/llama-3.2-90b": {
9287
+ id: "meta/llama-3.2-90b",
9288
+ name: "Llama 3.2 90B Vision Instruct",
9289
+ api: "anthropic-messages",
9290
+ provider: "vercel-ai-gateway",
9291
+ baseUrl: "https://ai-gateway.vercel.sh",
9292
+ reasoning: false,
9293
+ input: ["text", "image"],
9294
+ cost: {
9295
+ input: 0.72,
9296
+ output: 0.72,
9297
+ cacheRead: 0,
9298
+ cacheWrite: 0,
9299
+ },
9300
+ contextWindow: 128000,
9301
+ maxTokens: 8192,
9302
+ } satisfies Model<"anthropic-messages">,
9303
+ "meta/llama-3.3-70b": {
9304
+ id: "meta/llama-3.3-70b",
9305
+ name: "Llama 3.3 70B Instruct",
9306
+ api: "anthropic-messages",
9307
+ provider: "vercel-ai-gateway",
9308
+ baseUrl: "https://ai-gateway.vercel.sh",
9309
+ reasoning: false,
9310
+ input: ["text"],
9311
+ cost: {
9312
+ input: 0.72,
9313
+ output: 0.72,
9314
+ cacheRead: 0,
9315
+ cacheWrite: 0,
9316
+ },
9317
+ contextWindow: 128000,
9318
+ maxTokens: 8192,
9319
+ } satisfies Model<"anthropic-messages">,
9320
+ "meta/llama-4-maverick": {
9321
+ id: "meta/llama-4-maverick",
9322
+ name: "Llama 4 Maverick 17B Instruct",
9323
+ api: "anthropic-messages",
9324
+ provider: "vercel-ai-gateway",
9325
+ baseUrl: "https://ai-gateway.vercel.sh",
9326
+ reasoning: false,
9327
+ input: ["text", "image"],
9328
+ cost: {
9329
+ input: 0.15,
9330
+ output: 0.6,
9331
+ cacheRead: 0,
9332
+ cacheWrite: 0,
9333
+ },
9334
+ contextWindow: 131072,
9335
+ maxTokens: 8192,
9336
+ } satisfies Model<"anthropic-messages">,
9337
+ "meta/llama-4-scout": {
9338
+ id: "meta/llama-4-scout",
9339
+ name: "Llama 4 Scout 17B Instruct",
9340
+ api: "anthropic-messages",
9341
+ provider: "vercel-ai-gateway",
9342
+ baseUrl: "https://ai-gateway.vercel.sh",
9343
+ reasoning: false,
9344
+ input: ["text", "image"],
9345
+ cost: {
9346
+ input: 0.08,
9347
+ output: 0.3,
9348
+ cacheRead: 0,
9349
+ cacheWrite: 0,
9350
+ },
9351
+ contextWindow: 131072,
9352
+ maxTokens: 8192,
9353
+ } satisfies Model<"anthropic-messages">,
9354
+ "minimax/minimax-m2": {
9355
+ id: "minimax/minimax-m2",
9356
+ name: "MiniMax M2",
9357
+ api: "anthropic-messages",
9358
+ provider: "vercel-ai-gateway",
9359
+ baseUrl: "https://ai-gateway.vercel.sh",
9360
+ reasoning: true,
9361
+ input: ["text"],
9362
+ cost: {
9363
+ input: 0.27,
9364
+ output: 1.15,
9365
+ cacheRead: 0,
9366
+ cacheWrite: 0,
9367
+ },
9368
+ contextWindow: 262114,
9369
+ maxTokens: 262114,
9370
+ } satisfies Model<"anthropic-messages">,
9371
+ "minimax/minimax-m2.1": {
9372
+ id: "minimax/minimax-m2.1",
9373
+ name: "MiniMax M2.1",
9374
+ api: "anthropic-messages",
9375
+ provider: "vercel-ai-gateway",
9376
+ baseUrl: "https://ai-gateway.vercel.sh",
9377
+ reasoning: true,
9378
+ input: ["text"],
9379
+ cost: {
9380
+ input: 0.28,
9381
+ output: 1.2,
9382
+ cacheRead: 0.14,
9383
+ cacheWrite: 0,
9384
+ },
9385
+ contextWindow: 196608,
9386
+ maxTokens: 196608,
9387
+ } satisfies Model<"anthropic-messages">,
9388
+ "minimax/minimax-m2.1-lightning": {
9389
+ id: "minimax/minimax-m2.1-lightning",
9390
+ name: "MiniMax M2.1 Lightning",
9391
+ api: "anthropic-messages",
9392
+ provider: "vercel-ai-gateway",
9393
+ baseUrl: "https://ai-gateway.vercel.sh",
9394
+ reasoning: true,
9395
+ input: ["text"],
9396
+ cost: {
9397
+ input: 0.3,
9398
+ output: 2.4,
9399
+ cacheRead: 0.03,
9400
+ cacheWrite: 0.375,
9401
+ },
9402
+ contextWindow: 204800,
9403
+ maxTokens: 131072,
9404
+ } satisfies Model<"anthropic-messages">,
9405
+ "mistral/codestral": {
9406
+ id: "mistral/codestral",
9407
+ name: "Mistral Codestral",
9408
+ api: "anthropic-messages",
9409
+ provider: "vercel-ai-gateway",
9410
+ baseUrl: "https://ai-gateway.vercel.sh",
9411
+ reasoning: false,
9412
+ input: ["text"],
9413
+ cost: {
9414
+ input: 0.3,
9415
+ output: 0.8999999999999999,
9416
+ cacheRead: 0,
9417
+ cacheWrite: 0,
9418
+ },
9419
+ contextWindow: 128000,
9420
+ maxTokens: 4000,
9421
+ } satisfies Model<"anthropic-messages">,
9422
+ "mistral/devstral-2": {
9423
+ id: "mistral/devstral-2",
9424
+ name: "Devstral 2",
9425
+ api: "anthropic-messages",
9426
+ provider: "vercel-ai-gateway",
9427
+ baseUrl: "https://ai-gateway.vercel.sh",
9428
+ reasoning: false,
9429
+ input: ["text"],
9430
+ cost: {
9431
+ input: 0,
9432
+ output: 0,
9433
+ cacheRead: 0,
9434
+ cacheWrite: 0,
9435
+ },
9436
+ contextWindow: 256000,
9437
+ maxTokens: 256000,
9438
+ } satisfies Model<"anthropic-messages">,
9439
+ "mistral/devstral-small": {
9440
+ id: "mistral/devstral-small",
9441
+ name: "Devstral Small 1.1",
9442
+ api: "anthropic-messages",
9443
+ provider: "vercel-ai-gateway",
9444
+ baseUrl: "https://ai-gateway.vercel.sh",
9445
+ reasoning: false,
9446
+ input: ["text"],
9447
+ cost: {
9448
+ input: 0.09999999999999999,
9449
+ output: 0.3,
9450
+ cacheRead: 0,
9451
+ cacheWrite: 0,
9452
+ },
9453
+ contextWindow: 128000,
9454
+ maxTokens: 64000,
9455
+ } satisfies Model<"anthropic-messages">,
9456
+ "mistral/devstral-small-2": {
9457
+ id: "mistral/devstral-small-2",
9458
+ name: "Devstral Small 2",
9459
+ api: "anthropic-messages",
9460
+ provider: "vercel-ai-gateway",
9461
+ baseUrl: "https://ai-gateway.vercel.sh",
9462
+ reasoning: false,
9463
+ input: ["text"],
9464
+ cost: {
9465
+ input: 0,
9466
+ output: 0,
9467
+ cacheRead: 0,
9468
+ cacheWrite: 0,
9469
+ },
9470
+ contextWindow: 256000,
9471
+ maxTokens: 256000,
9472
+ } satisfies Model<"anthropic-messages">,
9473
+ "mistral/ministral-3b": {
9474
+ id: "mistral/ministral-3b",
9475
+ name: "Ministral 3B",
9476
+ api: "anthropic-messages",
9477
+ provider: "vercel-ai-gateway",
9478
+ baseUrl: "https://ai-gateway.vercel.sh",
9479
+ reasoning: false,
9480
+ input: ["text"],
9481
+ cost: {
9482
+ input: 0.04,
9483
+ output: 0.04,
9484
+ cacheRead: 0,
9485
+ cacheWrite: 0,
9486
+ },
9487
+ contextWindow: 128000,
9488
+ maxTokens: 4000,
9489
+ } satisfies Model<"anthropic-messages">,
9490
+ "mistral/ministral-8b": {
9491
+ id: "mistral/ministral-8b",
9492
+ name: "Ministral 8B",
9493
+ api: "anthropic-messages",
9494
+ provider: "vercel-ai-gateway",
9495
+ baseUrl: "https://ai-gateway.vercel.sh",
9496
+ reasoning: false,
9497
+ input: ["text"],
9498
+ cost: {
9499
+ input: 0.09999999999999999,
9500
+ output: 0.09999999999999999,
9501
+ cacheRead: 0,
9502
+ cacheWrite: 0,
9503
+ },
9504
+ contextWindow: 128000,
9505
+ maxTokens: 4000,
9506
+ } satisfies Model<"anthropic-messages">,
9507
+ "mistral/mistral-medium": {
9508
+ id: "mistral/mistral-medium",
9509
+ name: "Mistral Medium 3.1",
9510
+ api: "anthropic-messages",
9511
+ provider: "vercel-ai-gateway",
9512
+ baseUrl: "https://ai-gateway.vercel.sh",
9513
+ reasoning: false,
9514
+ input: ["text", "image"],
9515
+ cost: {
9516
+ input: 0.39999999999999997,
9517
+ output: 2,
9518
+ cacheRead: 0,
9519
+ cacheWrite: 0,
9520
+ },
9521
+ contextWindow: 128000,
9522
+ maxTokens: 64000,
9523
+ } satisfies Model<"anthropic-messages">,
9524
+ "mistral/mistral-small": {
9525
+ id: "mistral/mistral-small",
9526
+ name: "Mistral Small",
9527
+ api: "anthropic-messages",
9528
+ provider: "vercel-ai-gateway",
9529
+ baseUrl: "https://ai-gateway.vercel.sh",
9530
+ reasoning: false,
9531
+ input: ["text", "image"],
9532
+ cost: {
9533
+ input: 0.09999999999999999,
9534
+ output: 0.3,
9535
+ cacheRead: 0,
9536
+ cacheWrite: 0,
9537
+ },
9538
+ contextWindow: 32000,
9539
+ maxTokens: 4000,
9540
+ } satisfies Model<"anthropic-messages">,
9541
+ "mistral/pixtral-12b": {
9542
+ id: "mistral/pixtral-12b",
9543
+ name: "Pixtral 12B 2409",
9544
+ api: "anthropic-messages",
9545
+ provider: "vercel-ai-gateway",
9546
+ baseUrl: "https://ai-gateway.vercel.sh",
9547
+ reasoning: false,
9548
+ input: ["text", "image"],
9549
+ cost: {
9550
+ input: 0.15,
9551
+ output: 0.15,
9552
+ cacheRead: 0,
9553
+ cacheWrite: 0,
9554
+ },
9555
+ contextWindow: 128000,
9556
+ maxTokens: 4000,
9557
+ } satisfies Model<"anthropic-messages">,
9558
+ "mistral/pixtral-large": {
9559
+ id: "mistral/pixtral-large",
9560
+ name: "Pixtral Large",
9561
+ api: "anthropic-messages",
9562
+ provider: "vercel-ai-gateway",
9563
+ baseUrl: "https://ai-gateway.vercel.sh",
9564
+ reasoning: false,
9565
+ input: ["text", "image"],
9566
+ cost: {
9567
+ input: 2,
9568
+ output: 6,
9569
+ cacheRead: 0,
9570
+ cacheWrite: 0,
9571
+ },
9572
+ contextWindow: 128000,
9573
+ maxTokens: 4000,
9574
+ } satisfies Model<"anthropic-messages">,
9575
+ "moonshotai/kimi-k2": {
9576
+ id: "moonshotai/kimi-k2",
9577
+ name: "Kimi K2",
9578
+ api: "anthropic-messages",
9579
+ provider: "vercel-ai-gateway",
9580
+ baseUrl: "https://ai-gateway.vercel.sh",
9581
+ reasoning: false,
9582
+ input: ["text"],
9583
+ cost: {
9584
+ input: 0.5,
9585
+ output: 2,
9586
+ cacheRead: 0,
9587
+ cacheWrite: 0,
9588
+ },
9589
+ contextWindow: 131072,
9590
+ maxTokens: 16384,
9591
+ } satisfies Model<"anthropic-messages">,
9592
+ "moonshotai/kimi-k2-thinking": {
9593
+ id: "moonshotai/kimi-k2-thinking",
9594
+ name: "Kimi K2 Thinking",
9595
+ api: "anthropic-messages",
9596
+ provider: "vercel-ai-gateway",
9597
+ baseUrl: "https://ai-gateway.vercel.sh",
9598
+ reasoning: true,
9599
+ input: ["text"],
9600
+ cost: {
9601
+ input: 0.47,
9602
+ output: 2,
9603
+ cacheRead: 0.14100000000000001,
9604
+ cacheWrite: 0,
9605
+ },
9606
+ contextWindow: 216144,
9607
+ maxTokens: 216144,
9608
+ } satisfies Model<"anthropic-messages">,
9609
+ "moonshotai/kimi-k2-thinking-turbo": {
9610
+ id: "moonshotai/kimi-k2-thinking-turbo",
9611
+ name: "Kimi K2 Thinking Turbo",
9612
+ api: "anthropic-messages",
9613
+ provider: "vercel-ai-gateway",
9614
+ baseUrl: "https://ai-gateway.vercel.sh",
9615
+ reasoning: true,
9616
+ input: ["text"],
9617
+ cost: {
9618
+ input: 1.15,
9619
+ output: 8,
9620
+ cacheRead: 0.15,
9621
+ cacheWrite: 0,
9622
+ },
9623
+ contextWindow: 262114,
9624
+ maxTokens: 262114,
9625
+ } satisfies Model<"anthropic-messages">,
9626
+ "moonshotai/kimi-k2-turbo": {
9627
+ id: "moonshotai/kimi-k2-turbo",
9628
+ name: "Kimi K2 Turbo",
9629
+ api: "anthropic-messages",
9630
+ provider: "vercel-ai-gateway",
9631
+ baseUrl: "https://ai-gateway.vercel.sh",
9632
+ reasoning: false,
9633
+ input: ["text"],
9634
+ cost: {
9635
+ input: 2.4,
9636
+ output: 10,
9637
+ cacheRead: 0,
9638
+ cacheWrite: 0,
9639
+ },
9640
+ contextWindow: 256000,
9641
+ maxTokens: 16384,
9642
+ } satisfies Model<"anthropic-messages">,
9643
+ "nvidia/nemotron-nano-12b-v2-vl": {
9644
+ id: "nvidia/nemotron-nano-12b-v2-vl",
9645
+ name: "Nvidia Nemotron Nano 12B V2 VL",
9646
+ api: "anthropic-messages",
9647
+ provider: "vercel-ai-gateway",
9648
+ baseUrl: "https://ai-gateway.vercel.sh",
9649
+ reasoning: true,
9650
+ input: ["text", "image"],
9651
+ cost: {
9652
+ input: 0.19999999999999998,
9653
+ output: 0.6,
9654
+ cacheRead: 0,
9655
+ cacheWrite: 0,
9656
+ },
9657
+ contextWindow: 131072,
9658
+ maxTokens: 131072,
9659
+ } satisfies Model<"anthropic-messages">,
9660
+ "nvidia/nemotron-nano-9b-v2": {
9661
+ id: "nvidia/nemotron-nano-9b-v2",
9662
+ name: "Nvidia Nemotron Nano 9B V2",
9663
+ api: "anthropic-messages",
9664
+ provider: "vercel-ai-gateway",
9665
+ baseUrl: "https://ai-gateway.vercel.sh",
9666
+ reasoning: true,
9667
+ input: ["text"],
9668
+ cost: {
9669
+ input: 0.04,
9670
+ output: 0.16,
9671
+ cacheRead: 0,
9672
+ cacheWrite: 0,
9673
+ },
9674
+ contextWindow: 131072,
9675
+ maxTokens: 131072,
9676
+ } satisfies Model<"anthropic-messages">,
9677
+ "openai/codex-mini": {
9678
+ id: "openai/codex-mini",
9679
+ name: "Codex Mini",
9680
+ api: "anthropic-messages",
9681
+ provider: "vercel-ai-gateway",
9682
+ baseUrl: "https://ai-gateway.vercel.sh",
9683
+ reasoning: true,
9684
+ input: ["text", "image"],
9685
+ cost: {
9686
+ input: 1.5,
9687
+ output: 6,
9688
+ cacheRead: 0.375,
9689
+ cacheWrite: 0,
9690
+ },
9691
+ contextWindow: 200000,
9692
+ maxTokens: 100000,
9693
+ } satisfies Model<"anthropic-messages">,
9694
+ "openai/gpt-4-turbo": {
9695
+ id: "openai/gpt-4-turbo",
9696
+ name: "GPT-4 Turbo",
9697
+ api: "anthropic-messages",
9698
+ provider: "vercel-ai-gateway",
9699
+ baseUrl: "https://ai-gateway.vercel.sh",
9700
+ reasoning: false,
9701
+ input: ["text", "image"],
9702
+ cost: {
9703
+ input: 10,
9704
+ output: 30,
9705
+ cacheRead: 0,
9706
+ cacheWrite: 0,
9707
+ },
9708
+ contextWindow: 128000,
9709
+ maxTokens: 4096,
9710
+ } satisfies Model<"anthropic-messages">,
9711
+ "openai/gpt-4.1": {
9712
+ id: "openai/gpt-4.1",
9713
+ name: "GPT-4.1",
9714
+ api: "anthropic-messages",
9715
+ provider: "vercel-ai-gateway",
9716
+ baseUrl: "https://ai-gateway.vercel.sh",
9717
+ reasoning: false,
9718
+ input: ["text", "image"],
9719
+ cost: {
9720
+ input: 2,
9721
+ output: 8,
9722
+ cacheRead: 0.5,
9723
+ cacheWrite: 0,
9724
+ },
9725
+ contextWindow: 1047576,
9726
+ maxTokens: 32768,
9727
+ } satisfies Model<"anthropic-messages">,
9728
+ "openai/gpt-4.1-mini": {
9729
+ id: "openai/gpt-4.1-mini",
9730
+ name: "GPT-4.1 mini",
9731
+ api: "anthropic-messages",
9732
+ provider: "vercel-ai-gateway",
9733
+ baseUrl: "https://ai-gateway.vercel.sh",
9734
+ reasoning: false,
9735
+ input: ["text", "image"],
9736
+ cost: {
9737
+ input: 0.39999999999999997,
9738
+ output: 1.5999999999999999,
9739
+ cacheRead: 0.09999999999999999,
9740
+ cacheWrite: 0,
9741
+ },
9742
+ contextWindow: 1047576,
9743
+ maxTokens: 32768,
9744
+ } satisfies Model<"anthropic-messages">,
9745
+ "openai/gpt-4.1-nano": {
9746
+ id: "openai/gpt-4.1-nano",
9747
+ name: "GPT-4.1 nano",
9748
+ api: "anthropic-messages",
9749
+ provider: "vercel-ai-gateway",
9750
+ baseUrl: "https://ai-gateway.vercel.sh",
9751
+ reasoning: false,
9752
+ input: ["text", "image"],
9753
+ cost: {
9754
+ input: 0.09999999999999999,
9755
+ output: 0.39999999999999997,
9756
+ cacheRead: 0.024999999999999998,
9757
+ cacheWrite: 0,
9758
+ },
9759
+ contextWindow: 1047576,
9760
+ maxTokens: 32768,
9761
+ } satisfies Model<"anthropic-messages">,
9762
+ "openai/gpt-4o": {
9763
+ id: "openai/gpt-4o",
9764
+ name: "GPT-4o",
9765
+ api: "anthropic-messages",
9766
+ provider: "vercel-ai-gateway",
9767
+ baseUrl: "https://ai-gateway.vercel.sh",
9768
+ reasoning: false,
9769
+ input: ["text", "image"],
9770
+ cost: {
9771
+ input: 2.5,
9772
+ output: 10,
9773
+ cacheRead: 1.25,
9774
+ cacheWrite: 0,
9775
+ },
9776
+ contextWindow: 128000,
9777
+ maxTokens: 16384,
9778
+ } satisfies Model<"anthropic-messages">,
9779
+ "openai/gpt-4o-mini": {
9780
+ id: "openai/gpt-4o-mini",
9781
+ name: "GPT-4o mini",
9782
+ api: "anthropic-messages",
9783
+ provider: "vercel-ai-gateway",
9784
+ baseUrl: "https://ai-gateway.vercel.sh",
9785
+ reasoning: false,
9786
+ input: ["text", "image"],
9787
+ cost: {
9788
+ input: 0.15,
9789
+ output: 0.6,
9790
+ cacheRead: 0.075,
9791
+ cacheWrite: 0,
9792
+ },
9793
+ contextWindow: 128000,
9794
+ maxTokens: 16384,
9795
+ } satisfies Model<"anthropic-messages">,
9796
+ "openai/gpt-5": {
9797
+ id: "openai/gpt-5",
9798
+ name: "GPT-5",
9799
+ api: "anthropic-messages",
9800
+ provider: "vercel-ai-gateway",
9801
+ baseUrl: "https://ai-gateway.vercel.sh",
9802
+ reasoning: true,
9803
+ input: ["text", "image"],
9804
+ cost: {
9805
+ input: 1.25,
9806
+ output: 10,
9807
+ cacheRead: 0.13,
9808
+ cacheWrite: 0,
9809
+ },
9810
+ contextWindow: 400000,
9811
+ maxTokens: 128000,
9812
+ } satisfies Model<"anthropic-messages">,
9813
+ "openai/gpt-5-chat": {
9814
+ id: "openai/gpt-5-chat",
9815
+ name: "GPT-5 Chat",
9816
+ api: "anthropic-messages",
9817
+ provider: "vercel-ai-gateway",
9818
+ baseUrl: "https://ai-gateway.vercel.sh",
9819
+ reasoning: true,
9820
+ input: ["text", "image"],
9821
+ cost: {
9822
+ input: 1.25,
9823
+ output: 10,
9824
+ cacheRead: 0.125,
9825
+ cacheWrite: 0,
9826
+ },
9827
+ contextWindow: 128000,
9828
+ maxTokens: 16384,
9829
+ } satisfies Model<"anthropic-messages">,
9830
+ "openai/gpt-5-codex": {
9831
+ id: "openai/gpt-5-codex",
9832
+ name: "GPT-5-Codex",
9833
+ api: "anthropic-messages",
9834
+ provider: "vercel-ai-gateway",
9835
+ baseUrl: "https://ai-gateway.vercel.sh",
9836
+ reasoning: true,
9837
+ input: ["text", "image"],
9838
+ cost: {
9839
+ input: 1.25,
9840
+ output: 10,
9841
+ cacheRead: 0.13,
9842
+ cacheWrite: 0,
9843
+ },
9844
+ contextWindow: 400000,
9845
+ maxTokens: 128000,
9846
+ } satisfies Model<"anthropic-messages">,
9847
+ "openai/gpt-5-mini": {
9848
+ id: "openai/gpt-5-mini",
9849
+ name: "GPT-5 mini",
9850
+ api: "anthropic-messages",
9851
+ provider: "vercel-ai-gateway",
9852
+ baseUrl: "https://ai-gateway.vercel.sh",
9853
+ reasoning: true,
9854
+ input: ["text", "image"],
9855
+ cost: {
9856
+ input: 0.25,
9857
+ output: 2,
9858
+ cacheRead: 0.03,
9859
+ cacheWrite: 0,
9860
+ },
9861
+ contextWindow: 400000,
9862
+ maxTokens: 128000,
9863
+ } satisfies Model<"anthropic-messages">,
9864
+ "openai/gpt-5-nano": {
9865
+ id: "openai/gpt-5-nano",
9866
+ name: "GPT-5 nano",
9867
+ api: "anthropic-messages",
9868
+ provider: "vercel-ai-gateway",
9869
+ baseUrl: "https://ai-gateway.vercel.sh",
9870
+ reasoning: true,
9871
+ input: ["text", "image"],
9872
+ cost: {
9873
+ input: 0.049999999999999996,
9874
+ output: 0.39999999999999997,
9875
+ cacheRead: 0.01,
9876
+ cacheWrite: 0,
9877
+ },
9878
+ contextWindow: 400000,
9879
+ maxTokens: 128000,
9880
+ } satisfies Model<"anthropic-messages">,
9881
+ "openai/gpt-5-pro": {
9882
+ id: "openai/gpt-5-pro",
9883
+ name: "GPT-5 pro",
9884
+ api: "anthropic-messages",
9885
+ provider: "vercel-ai-gateway",
9886
+ baseUrl: "https://ai-gateway.vercel.sh",
9887
+ reasoning: true,
9888
+ input: ["text", "image"],
9889
+ cost: {
9890
+ input: 15,
9891
+ output: 120,
9892
+ cacheRead: 0,
9893
+ cacheWrite: 0,
9894
+ },
9895
+ contextWindow: 400000,
9896
+ maxTokens: 272000,
9897
+ } satisfies Model<"anthropic-messages">,
9898
+ "openai/gpt-5.1-codex": {
9899
+ id: "openai/gpt-5.1-codex",
9900
+ name: "GPT-5.1-Codex",
9901
+ api: "anthropic-messages",
9902
+ provider: "vercel-ai-gateway",
9903
+ baseUrl: "https://ai-gateway.vercel.sh",
9904
+ reasoning: true,
9905
+ input: ["text", "image"],
9906
+ cost: {
9907
+ input: 1.25,
9908
+ output: 10,
9909
+ cacheRead: 0.125,
9910
+ cacheWrite: 0,
9911
+ },
9912
+ contextWindow: 400000,
9913
+ maxTokens: 128000,
9914
+ } satisfies Model<"anthropic-messages">,
9915
+ "openai/gpt-5.1-codex-max": {
9916
+ id: "openai/gpt-5.1-codex-max",
9917
+ name: "GPT 5.1 Codex Max",
9918
+ api: "anthropic-messages",
9919
+ provider: "vercel-ai-gateway",
9920
+ baseUrl: "https://ai-gateway.vercel.sh",
9921
+ reasoning: true,
9922
+ input: ["text", "image"],
9923
+ cost: {
9924
+ input: 1.25,
9925
+ output: 10,
9926
+ cacheRead: 0.125,
9927
+ cacheWrite: 0,
9928
+ },
9929
+ contextWindow: 400000,
9930
+ maxTokens: 128000,
9931
+ } satisfies Model<"anthropic-messages">,
9932
+ "openai/gpt-5.1-codex-mini": {
9933
+ id: "openai/gpt-5.1-codex-mini",
9934
+ name: "GPT-5.1 Codex mini",
9935
+ api: "anthropic-messages",
9936
+ provider: "vercel-ai-gateway",
9937
+ baseUrl: "https://ai-gateway.vercel.sh",
9938
+ reasoning: true,
9939
+ input: ["text", "image"],
9940
+ cost: {
9941
+ input: 0.25,
9942
+ output: 2,
9943
+ cacheRead: 0.024999999999999998,
9944
+ cacheWrite: 0,
9945
+ },
9946
+ contextWindow: 400000,
9947
+ maxTokens: 128000,
9948
+ } satisfies Model<"anthropic-messages">,
9949
+ "openai/gpt-5.1-instant": {
9950
+ id: "openai/gpt-5.1-instant",
9951
+ name: "GPT-5.1 Instant",
9952
+ api: "anthropic-messages",
9953
+ provider: "vercel-ai-gateway",
9954
+ baseUrl: "https://ai-gateway.vercel.sh",
9955
+ reasoning: true,
9956
+ input: ["text", "image"],
9957
+ cost: {
9958
+ input: 1.25,
9959
+ output: 10,
9960
+ cacheRead: 0.125,
9961
+ cacheWrite: 0,
9962
+ },
9963
+ contextWindow: 128000,
9964
+ maxTokens: 16384,
9965
+ } satisfies Model<"anthropic-messages">,
9966
+ "openai/gpt-5.1-thinking": {
9967
+ id: "openai/gpt-5.1-thinking",
9968
+ name: "GPT 5.1 Thinking",
9969
+ api: "anthropic-messages",
9970
+ provider: "vercel-ai-gateway",
9971
+ baseUrl: "https://ai-gateway.vercel.sh",
9972
+ reasoning: true,
9973
+ input: ["text", "image"],
9974
+ cost: {
9975
+ input: 1.25,
9976
+ output: 10,
9977
+ cacheRead: 0.125,
9978
+ cacheWrite: 0,
9979
+ },
9980
+ contextWindow: 400000,
9981
+ maxTokens: 128000,
9982
+ } satisfies Model<"anthropic-messages">,
9983
+ "openai/gpt-5.2": {
9984
+ id: "openai/gpt-5.2",
9985
+ name: "GPT-5.2",
9986
+ api: "anthropic-messages",
9987
+ provider: "vercel-ai-gateway",
9988
+ baseUrl: "https://ai-gateway.vercel.sh",
9989
+ reasoning: true,
9990
+ input: ["text", "image"],
9991
+ cost: {
9992
+ input: 1.75,
9993
+ output: 14,
9994
+ cacheRead: 0.175,
9995
+ cacheWrite: 0,
9996
+ },
9997
+ contextWindow: 400000,
9998
+ maxTokens: 128000,
9999
+ } satisfies Model<"anthropic-messages">,
10000
+ "openai/gpt-5.2-chat": {
10001
+ id: "openai/gpt-5.2-chat",
10002
+ name: "GPT-5.2 Chat",
10003
+ api: "anthropic-messages",
10004
+ provider: "vercel-ai-gateway",
10005
+ baseUrl: "https://ai-gateway.vercel.sh",
10006
+ reasoning: true,
10007
+ input: ["text", "image"],
10008
+ cost: {
10009
+ input: 1.75,
10010
+ output: 14,
10011
+ cacheRead: 0.175,
10012
+ cacheWrite: 0,
10013
+ },
10014
+ contextWindow: 128000,
10015
+ maxTokens: 16384,
10016
+ } satisfies Model<"anthropic-messages">,
10017
+ "openai/gpt-5.2-codex": {
10018
+ id: "openai/gpt-5.2-codex",
10019
+ name: "GPT-5.2-Codex",
10020
+ api: "anthropic-messages",
10021
+ provider: "vercel-ai-gateway",
10022
+ baseUrl: "https://ai-gateway.vercel.sh",
10023
+ reasoning: true,
10024
+ input: ["text", "image"],
10025
+ cost: {
10026
+ input: 1.75,
10027
+ output: 14,
10028
+ cacheRead: 0.175,
10029
+ cacheWrite: 0,
10030
+ },
10031
+ contextWindow: 400000,
10032
+ maxTokens: 128000,
10033
+ } satisfies Model<"anthropic-messages">,
10034
+ "openai/gpt-5.2-pro": {
10035
+ id: "openai/gpt-5.2-pro",
10036
+ name: "GPT 5.2 ",
10037
+ api: "anthropic-messages",
10038
+ provider: "vercel-ai-gateway",
10039
+ baseUrl: "https://ai-gateway.vercel.sh",
10040
+ reasoning: true,
10041
+ input: ["text", "image"],
10042
+ cost: {
10043
+ input: 21,
10044
+ output: 168,
10045
+ cacheRead: 0,
10046
+ cacheWrite: 0,
10047
+ },
10048
+ contextWindow: 400000,
10049
+ maxTokens: 128000,
10050
+ } satisfies Model<"anthropic-messages">,
10051
+ "openai/gpt-oss-120b": {
10052
+ id: "openai/gpt-oss-120b",
10053
+ name: "gpt-oss-120b",
10054
+ api: "anthropic-messages",
10055
+ provider: "vercel-ai-gateway",
10056
+ baseUrl: "https://ai-gateway.vercel.sh",
10057
+ reasoning: true,
10058
+ input: ["text"],
10059
+ cost: {
10060
+ input: 0.09999999999999999,
10061
+ output: 0.5,
10062
+ cacheRead: 0,
10063
+ cacheWrite: 0,
10064
+ },
10065
+ contextWindow: 131072,
10066
+ maxTokens: 131072,
10067
+ } satisfies Model<"anthropic-messages">,
10068
+ "openai/gpt-oss-20b": {
10069
+ id: "openai/gpt-oss-20b",
10070
+ name: "gpt-oss-20b",
10071
+ api: "anthropic-messages",
10072
+ provider: "vercel-ai-gateway",
10073
+ baseUrl: "https://ai-gateway.vercel.sh",
10074
+ reasoning: true,
10075
+ input: ["text"],
10076
+ cost: {
10077
+ input: 0.07,
10078
+ output: 0.3,
10079
+ cacheRead: 0,
10080
+ cacheWrite: 0,
10081
+ },
10082
+ contextWindow: 128000,
10083
+ maxTokens: 8192,
10084
+ } satisfies Model<"anthropic-messages">,
10085
+ "openai/gpt-oss-safeguard-20b": {
10086
+ id: "openai/gpt-oss-safeguard-20b",
10087
+ name: "gpt-oss-safeguard-20b",
10088
+ api: "anthropic-messages",
10089
+ provider: "vercel-ai-gateway",
10090
+ baseUrl: "https://ai-gateway.vercel.sh",
10091
+ reasoning: true,
10092
+ input: ["text"],
10093
+ cost: {
10094
+ input: 0.075,
10095
+ output: 0.3,
10096
+ cacheRead: 0.037,
10097
+ cacheWrite: 0,
10098
+ },
10099
+ contextWindow: 131072,
10100
+ maxTokens: 65536,
10101
+ } satisfies Model<"anthropic-messages">,
10102
+ "openai/o1": {
10103
+ id: "openai/o1",
10104
+ name: "o1",
10105
+ api: "anthropic-messages",
10106
+ provider: "vercel-ai-gateway",
10107
+ baseUrl: "https://ai-gateway.vercel.sh",
10108
+ reasoning: true,
10109
+ input: ["text", "image"],
10110
+ cost: {
10111
+ input: 15,
10112
+ output: 60,
10113
+ cacheRead: 7.5,
10114
+ cacheWrite: 0,
10115
+ },
10116
+ contextWindow: 200000,
10117
+ maxTokens: 100000,
10118
+ } satisfies Model<"anthropic-messages">,
10119
+ "openai/o3": {
10120
+ id: "openai/o3",
10121
+ name: "o3",
10122
+ api: "anthropic-messages",
10123
+ provider: "vercel-ai-gateway",
10124
+ baseUrl: "https://ai-gateway.vercel.sh",
10125
+ reasoning: true,
10126
+ input: ["text", "image"],
10127
+ cost: {
10128
+ input: 2,
10129
+ output: 8,
10130
+ cacheRead: 0.5,
10131
+ cacheWrite: 0,
10132
+ },
10133
+ contextWindow: 200000,
10134
+ maxTokens: 100000,
10135
+ } satisfies Model<"anthropic-messages">,
10136
+ "openai/o3-deep-research": {
10137
+ id: "openai/o3-deep-research",
10138
+ name: "o3-deep-research",
10139
+ api: "anthropic-messages",
10140
+ provider: "vercel-ai-gateway",
10141
+ baseUrl: "https://ai-gateway.vercel.sh",
10142
+ reasoning: true,
10143
+ input: ["text", "image"],
10144
+ cost: {
10145
+ input: 10,
10146
+ output: 40,
10147
+ cacheRead: 2.5,
10148
+ cacheWrite: 0,
10149
+ },
10150
+ contextWindow: 200000,
10151
+ maxTokens: 100000,
10152
+ } satisfies Model<"anthropic-messages">,
10153
+ "openai/o3-mini": {
10154
+ id: "openai/o3-mini",
10155
+ name: "o3-mini",
10156
+ api: "anthropic-messages",
10157
+ provider: "vercel-ai-gateway",
10158
+ baseUrl: "https://ai-gateway.vercel.sh",
10159
+ reasoning: true,
10160
+ input: ["text"],
10161
+ cost: {
10162
+ input: 1.1,
10163
+ output: 4.4,
10164
+ cacheRead: 0.55,
10165
+ cacheWrite: 0,
10166
+ },
10167
+ contextWindow: 200000,
10168
+ maxTokens: 100000,
10169
+ } satisfies Model<"anthropic-messages">,
10170
+ "openai/o3-pro": {
10171
+ id: "openai/o3-pro",
10172
+ name: "o3 Pro",
10173
+ api: "anthropic-messages",
10174
+ provider: "vercel-ai-gateway",
10175
+ baseUrl: "https://ai-gateway.vercel.sh",
10176
+ reasoning: true,
10177
+ input: ["text", "image"],
10178
+ cost: {
10179
+ input: 20,
10180
+ output: 80,
10181
+ cacheRead: 0,
10182
+ cacheWrite: 0,
10183
+ },
10184
+ contextWindow: 200000,
10185
+ maxTokens: 100000,
10186
+ } satisfies Model<"anthropic-messages">,
10187
+ "openai/o4-mini": {
10188
+ id: "openai/o4-mini",
10189
+ name: "o4-mini",
10190
+ api: "anthropic-messages",
10191
+ provider: "vercel-ai-gateway",
10192
+ baseUrl: "https://ai-gateway.vercel.sh",
10193
+ reasoning: true,
10194
+ input: ["text", "image"],
10195
+ cost: {
10196
+ input: 1.1,
10197
+ output: 4.4,
10198
+ cacheRead: 0.275,
10199
+ cacheWrite: 0,
10200
+ },
10201
+ contextWindow: 200000,
10202
+ maxTokens: 100000,
10203
+ } satisfies Model<"anthropic-messages">,
10204
+ "perplexity/sonar": {
10205
+ id: "perplexity/sonar",
10206
+ name: "Sonar",
10207
+ api: "anthropic-messages",
10208
+ provider: "vercel-ai-gateway",
10209
+ baseUrl: "https://ai-gateway.vercel.sh",
10210
+ reasoning: false,
10211
+ input: ["text", "image"],
10212
+ cost: {
10213
+ input: 1,
10214
+ output: 1,
10215
+ cacheRead: 0,
10216
+ cacheWrite: 0,
10217
+ },
10218
+ contextWindow: 127000,
10219
+ maxTokens: 8000,
10220
+ } satisfies Model<"anthropic-messages">,
10221
+ "perplexity/sonar-pro": {
10222
+ id: "perplexity/sonar-pro",
10223
+ name: "Sonar Pro",
10224
+ api: "anthropic-messages",
10225
+ provider: "vercel-ai-gateway",
10226
+ baseUrl: "https://ai-gateway.vercel.sh",
10227
+ reasoning: false,
10228
+ input: ["text", "image"],
10229
+ cost: {
10230
+ input: 3,
10231
+ output: 15,
10232
+ cacheRead: 0,
10233
+ cacheWrite: 0,
10234
+ },
10235
+ contextWindow: 200000,
10236
+ maxTokens: 8000,
10237
+ } satisfies Model<"anthropic-messages">,
10238
+ "prime-intellect/intellect-3": {
10239
+ id: "prime-intellect/intellect-3",
10240
+ name: "INTELLECT 3",
10241
+ api: "anthropic-messages",
10242
+ provider: "vercel-ai-gateway",
10243
+ baseUrl: "https://ai-gateway.vercel.sh",
10244
+ reasoning: true,
10245
+ input: ["text"],
10246
+ cost: {
10247
+ input: 0.19999999999999998,
10248
+ output: 1.1,
10249
+ cacheRead: 0,
10250
+ cacheWrite: 0,
10251
+ },
10252
+ contextWindow: 131072,
10253
+ maxTokens: 131072,
10254
+ } satisfies Model<"anthropic-messages">,
10255
+ "stealth/sonoma-dusk-alpha": {
10256
+ id: "stealth/sonoma-dusk-alpha",
10257
+ name: "Sonoma Dusk Alpha",
10258
+ api: "anthropic-messages",
10259
+ provider: "vercel-ai-gateway",
10260
+ baseUrl: "https://ai-gateway.vercel.sh",
10261
+ reasoning: false,
10262
+ input: ["text", "image"],
10263
+ cost: {
10264
+ input: 0.19999999999999998,
10265
+ output: 0.5,
10266
+ cacheRead: 0.049999999999999996,
10267
+ cacheWrite: 0,
10268
+ },
10269
+ contextWindow: 2000000,
10270
+ maxTokens: 131072,
10271
+ } satisfies Model<"anthropic-messages">,
10272
+ "stealth/sonoma-sky-alpha": {
10273
+ id: "stealth/sonoma-sky-alpha",
10274
+ name: "Sonoma Sky Alpha",
10275
+ api: "anthropic-messages",
10276
+ provider: "vercel-ai-gateway",
10277
+ baseUrl: "https://ai-gateway.vercel.sh",
10278
+ reasoning: false,
10279
+ input: ["text", "image"],
10280
+ cost: {
10281
+ input: 0.19999999999999998,
10282
+ output: 0.5,
10283
+ cacheRead: 0.049999999999999996,
10284
+ cacheWrite: 0,
10285
+ },
10286
+ contextWindow: 2000000,
10287
+ maxTokens: 131072,
10288
+ } satisfies Model<"anthropic-messages">,
10289
+ "vercel/v0-1.0-md": {
10290
+ id: "vercel/v0-1.0-md",
10291
+ name: "v0-1.0-md",
10292
+ api: "anthropic-messages",
10293
+ provider: "vercel-ai-gateway",
10294
+ baseUrl: "https://ai-gateway.vercel.sh",
10295
+ reasoning: false,
10296
+ input: ["text", "image"],
10297
+ cost: {
10298
+ input: 3,
10299
+ output: 15,
10300
+ cacheRead: 0,
10301
+ cacheWrite: 0,
10302
+ },
10303
+ contextWindow: 128000,
10304
+ maxTokens: 32000,
10305
+ } satisfies Model<"anthropic-messages">,
10306
+ "vercel/v0-1.5-md": {
10307
+ id: "vercel/v0-1.5-md",
10308
+ name: "v0-1.5-md",
10309
+ api: "anthropic-messages",
10310
+ provider: "vercel-ai-gateway",
10311
+ baseUrl: "https://ai-gateway.vercel.sh",
10312
+ reasoning: false,
10313
+ input: ["text", "image"],
10314
+ cost: {
10315
+ input: 3,
10316
+ output: 15,
10317
+ cacheRead: 0,
10318
+ cacheWrite: 0,
10319
+ },
10320
+ contextWindow: 128000,
10321
+ maxTokens: 32768,
10322
+ } satisfies Model<"anthropic-messages">,
10323
+ "xai/grok-2-vision": {
10324
+ id: "xai/grok-2-vision",
10325
+ name: "Grok 2 Vision",
10326
+ api: "anthropic-messages",
10327
+ provider: "vercel-ai-gateway",
10328
+ baseUrl: "https://ai-gateway.vercel.sh",
10329
+ reasoning: false,
10330
+ input: ["text", "image"],
10331
+ cost: {
10332
+ input: 2,
10333
+ output: 10,
10334
+ cacheRead: 0,
10335
+ cacheWrite: 0,
10336
+ },
10337
+ contextWindow: 32768,
10338
+ maxTokens: 32768,
10339
+ } satisfies Model<"anthropic-messages">,
10340
+ "xai/grok-3": {
10341
+ id: "xai/grok-3",
10342
+ name: "Grok 3 Beta",
10343
+ api: "anthropic-messages",
10344
+ provider: "vercel-ai-gateway",
10345
+ baseUrl: "https://ai-gateway.vercel.sh",
10346
+ reasoning: false,
10347
+ input: ["text"],
10348
+ cost: {
10349
+ input: 3,
10350
+ output: 15,
10351
+ cacheRead: 0,
10352
+ cacheWrite: 0,
10353
+ },
10354
+ contextWindow: 131072,
10355
+ maxTokens: 131072,
10356
+ } satisfies Model<"anthropic-messages">,
10357
+ "xai/grok-3-fast": {
10358
+ id: "xai/grok-3-fast",
10359
+ name: "Grok 3 Fast Beta",
10360
+ api: "anthropic-messages",
10361
+ provider: "vercel-ai-gateway",
10362
+ baseUrl: "https://ai-gateway.vercel.sh",
10363
+ reasoning: false,
10364
+ input: ["text"],
10365
+ cost: {
10366
+ input: 5,
10367
+ output: 25,
10368
+ cacheRead: 0,
10369
+ cacheWrite: 0,
10370
+ },
10371
+ contextWindow: 131072,
10372
+ maxTokens: 131072,
10373
+ } satisfies Model<"anthropic-messages">,
10374
+ "xai/grok-3-mini": {
10375
+ id: "xai/grok-3-mini",
10376
+ name: "Grok 3 Mini Beta",
10377
+ api: "anthropic-messages",
10378
+ provider: "vercel-ai-gateway",
10379
+ baseUrl: "https://ai-gateway.vercel.sh",
10380
+ reasoning: false,
10381
+ input: ["text"],
10382
+ cost: {
10383
+ input: 0.3,
10384
+ output: 0.5,
10385
+ cacheRead: 0,
10386
+ cacheWrite: 0,
10387
+ },
10388
+ contextWindow: 131072,
10389
+ maxTokens: 131072,
10390
+ } satisfies Model<"anthropic-messages">,
10391
+ "xai/grok-3-mini-fast": {
10392
+ id: "xai/grok-3-mini-fast",
10393
+ name: "Grok 3 Mini Fast Beta",
10394
+ api: "anthropic-messages",
10395
+ provider: "vercel-ai-gateway",
10396
+ baseUrl: "https://ai-gateway.vercel.sh",
10397
+ reasoning: false,
10398
+ input: ["text"],
10399
+ cost: {
10400
+ input: 0.6,
10401
+ output: 4,
10402
+ cacheRead: 0,
10403
+ cacheWrite: 0,
10404
+ },
10405
+ contextWindow: 131072,
10406
+ maxTokens: 131072,
10407
+ } satisfies Model<"anthropic-messages">,
10408
+ "xai/grok-4": {
10409
+ id: "xai/grok-4",
10410
+ name: "Grok 4",
10411
+ api: "anthropic-messages",
10412
+ provider: "vercel-ai-gateway",
10413
+ baseUrl: "https://ai-gateway.vercel.sh",
10414
+ reasoning: true,
10415
+ input: ["text", "image"],
10416
+ cost: {
10417
+ input: 3,
10418
+ output: 15,
10419
+ cacheRead: 0,
10420
+ cacheWrite: 0,
10421
+ },
10422
+ contextWindow: 256000,
10423
+ maxTokens: 256000,
10424
+ } satisfies Model<"anthropic-messages">,
10425
+ "xai/grok-4-fast-non-reasoning": {
10426
+ id: "xai/grok-4-fast-non-reasoning",
10427
+ name: "Grok 4 Fast Non-Reasoning",
10428
+ api: "anthropic-messages",
10429
+ provider: "vercel-ai-gateway",
10430
+ baseUrl: "https://ai-gateway.vercel.sh",
10431
+ reasoning: false,
10432
+ input: ["text"],
10433
+ cost: {
10434
+ input: 0.19999999999999998,
10435
+ output: 0.5,
10436
+ cacheRead: 0.049999999999999996,
10437
+ cacheWrite: 0,
10438
+ },
10439
+ contextWindow: 2000000,
10440
+ maxTokens: 256000,
10441
+ } satisfies Model<"anthropic-messages">,
10442
+ "xai/grok-4-fast-reasoning": {
10443
+ id: "xai/grok-4-fast-reasoning",
10444
+ name: "Grok 4 Fast Reasoning",
10445
+ api: "anthropic-messages",
10446
+ provider: "vercel-ai-gateway",
10447
+ baseUrl: "https://ai-gateway.vercel.sh",
10448
+ reasoning: true,
10449
+ input: ["text"],
10450
+ cost: {
10451
+ input: 0.19999999999999998,
10452
+ output: 0.5,
10453
+ cacheRead: 0.049999999999999996,
10454
+ cacheWrite: 0,
10455
+ },
10456
+ contextWindow: 2000000,
10457
+ maxTokens: 256000,
10458
+ } satisfies Model<"anthropic-messages">,
10459
+ "xai/grok-4.1-fast-non-reasoning": {
10460
+ id: "xai/grok-4.1-fast-non-reasoning",
10461
+ name: "Grok 4.1 Fast Non-Reasoning",
10462
+ api: "anthropic-messages",
10463
+ provider: "vercel-ai-gateway",
10464
+ baseUrl: "https://ai-gateway.vercel.sh",
10465
+ reasoning: false,
10466
+ input: ["text"],
10467
+ cost: {
10468
+ input: 0.19999999999999998,
10469
+ output: 0.5,
10470
+ cacheRead: 0.049999999999999996,
10471
+ cacheWrite: 0,
10472
+ },
10473
+ contextWindow: 2000000,
10474
+ maxTokens: 30000,
10475
+ } satisfies Model<"anthropic-messages">,
10476
+ "xai/grok-4.1-fast-reasoning": {
10477
+ id: "xai/grok-4.1-fast-reasoning",
10478
+ name: "Grok 4.1 Fast Reasoning",
10479
+ api: "anthropic-messages",
10480
+ provider: "vercel-ai-gateway",
10481
+ baseUrl: "https://ai-gateway.vercel.sh",
10482
+ reasoning: true,
10483
+ input: ["text"],
10484
+ cost: {
10485
+ input: 0.19999999999999998,
10486
+ output: 0.5,
10487
+ cacheRead: 0.049999999999999996,
10488
+ cacheWrite: 0,
10489
+ },
10490
+ contextWindow: 2000000,
10491
+ maxTokens: 30000,
10492
+ } satisfies Model<"anthropic-messages">,
10493
+ "xai/grok-code-fast-1": {
10494
+ id: "xai/grok-code-fast-1",
10495
+ name: "Grok Code Fast 1",
10496
+ api: "anthropic-messages",
10497
+ provider: "vercel-ai-gateway",
10498
+ baseUrl: "https://ai-gateway.vercel.sh",
10499
+ reasoning: true,
10500
+ input: ["text"],
10501
+ cost: {
10502
+ input: 0.19999999999999998,
10503
+ output: 1.5,
10504
+ cacheRead: 0.02,
10505
+ cacheWrite: 0,
10506
+ },
10507
+ contextWindow: 256000,
10508
+ maxTokens: 256000,
10509
+ } satisfies Model<"anthropic-messages">,
10510
+ "xiaomi/mimo-v2-flash": {
10511
+ id: "xiaomi/mimo-v2-flash",
10512
+ name: "MiMo V2 Flash",
10513
+ api: "anthropic-messages",
10514
+ provider: "vercel-ai-gateway",
10515
+ baseUrl: "https://ai-gateway.vercel.sh",
10516
+ reasoning: true,
10517
+ input: ["text"],
10518
+ cost: {
10519
+ input: 0.098,
10520
+ output: 0.293,
10521
+ cacheRead: 0,
10522
+ cacheWrite: 0,
10523
+ },
10524
+ contextWindow: 262144,
10525
+ maxTokens: 32000,
10526
+ } satisfies Model<"anthropic-messages">,
10527
+ "zai/glm-4.5": {
10528
+ id: "zai/glm-4.5",
10529
+ name: "GLM-4.5",
10530
+ api: "anthropic-messages",
10531
+ provider: "vercel-ai-gateway",
10532
+ baseUrl: "https://ai-gateway.vercel.sh",
10533
+ reasoning: true,
10534
+ input: ["text"],
10535
+ cost: {
10536
+ input: 0.6,
10537
+ output: 2.2,
10538
+ cacheRead: 0,
10539
+ cacheWrite: 0,
10540
+ },
10541
+ contextWindow: 131072,
10542
+ maxTokens: 131072,
10543
+ } satisfies Model<"anthropic-messages">,
10544
+ "zai/glm-4.5-air": {
10545
+ id: "zai/glm-4.5-air",
10546
+ name: "GLM 4.5 Air",
10547
+ api: "anthropic-messages",
10548
+ provider: "vercel-ai-gateway",
10549
+ baseUrl: "https://ai-gateway.vercel.sh",
10550
+ reasoning: true,
10551
+ input: ["text"],
10552
+ cost: {
10553
+ input: 0.19999999999999998,
10554
+ output: 1.1,
10555
+ cacheRead: 0,
10556
+ cacheWrite: 0,
10557
+ },
10558
+ contextWindow: 128000,
10559
+ maxTokens: 96000,
10560
+ } satisfies Model<"anthropic-messages">,
10561
+ "zai/glm-4.5v": {
10562
+ id: "zai/glm-4.5v",
10563
+ name: "GLM 4.5V",
10564
+ api: "anthropic-messages",
10565
+ provider: "vercel-ai-gateway",
10566
+ baseUrl: "https://ai-gateway.vercel.sh",
10567
+ reasoning: true,
10568
+ input: ["text", "image"],
10569
+ cost: {
10570
+ input: 0.6,
10571
+ output: 1.7999999999999998,
10572
+ cacheRead: 0,
10573
+ cacheWrite: 0,
10574
+ },
10575
+ contextWindow: 65536,
10576
+ maxTokens: 66000,
10577
+ } satisfies Model<"anthropic-messages">,
10578
+ "zai/glm-4.6": {
10579
+ id: "zai/glm-4.6",
10580
+ name: "GLM 4.6",
10581
+ api: "anthropic-messages",
10582
+ provider: "vercel-ai-gateway",
10583
+ baseUrl: "https://ai-gateway.vercel.sh",
10584
+ reasoning: true,
10585
+ input: ["text"],
10586
+ cost: {
10587
+ input: 0.44999999999999996,
10588
+ output: 1.7999999999999998,
10589
+ cacheRead: 0.11,
10590
+ cacheWrite: 0,
10591
+ },
10592
+ contextWindow: 200000,
10593
+ maxTokens: 96000,
10594
+ } satisfies Model<"anthropic-messages">,
10595
+ "zai/glm-4.6v": {
10596
+ id: "zai/glm-4.6v",
10597
+ name: "GLM-4.6V",
10598
+ api: "anthropic-messages",
10599
+ provider: "vercel-ai-gateway",
10600
+ baseUrl: "https://ai-gateway.vercel.sh",
10601
+ reasoning: true,
10602
+ input: ["text", "image"],
10603
+ cost: {
10604
+ input: 0.3,
10605
+ output: 0.8999999999999999,
10606
+ cacheRead: 0.049999999999999996,
10607
+ cacheWrite: 0,
10608
+ },
10609
+ contextWindow: 128000,
10610
+ maxTokens: 24000,
10611
+ } satisfies Model<"anthropic-messages">,
10612
+ "zai/glm-4.6v-flash": {
10613
+ id: "zai/glm-4.6v-flash",
10614
+ name: "GLM-4.6V-Flash",
10615
+ api: "anthropic-messages",
10616
+ provider: "vercel-ai-gateway",
10617
+ baseUrl: "https://ai-gateway.vercel.sh",
10618
+ reasoning: true,
10619
+ input: ["text", "image"],
10620
+ cost: {
10621
+ input: 0,
10622
+ output: 0,
10623
+ cacheRead: 0,
10624
+ cacheWrite: 0,
10625
+ },
10626
+ contextWindow: 128000,
10627
+ maxTokens: 24000,
10628
+ } satisfies Model<"anthropic-messages">,
10629
+ "zai/glm-4.7": {
10630
+ id: "zai/glm-4.7",
10631
+ name: "GLM 4.7",
10632
+ api: "anthropic-messages",
10633
+ provider: "vercel-ai-gateway",
10634
+ baseUrl: "https://ai-gateway.vercel.sh",
10635
+ reasoning: true,
10636
+ input: ["text"],
10637
+ cost: {
10638
+ input: 0.43,
10639
+ output: 1.75,
10640
+ cacheRead: 0.08,
10641
+ cacheWrite: 0,
10642
+ },
10643
+ contextWindow: 202752,
10644
+ maxTokens: 120000,
10645
+ } satisfies Model<"anthropic-messages">,
10646
+ },
8568
10647
  "xai": {
8569
10648
  "grok-2": {
8570
10649
  id: "grok-2",
@@ -8948,7 +11027,7 @@ export const MODELS = {
8948
11027
  api: "openai-completions",
8949
11028
  provider: "zai",
8950
11029
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
8951
- compat: {"supportsDeveloperRole":false},
11030
+ compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
8952
11031
  reasoning: true,
8953
11032
  input: ["text"],
8954
11033
  cost: {
@@ -8966,7 +11045,7 @@ export const MODELS = {
8966
11045
  api: "openai-completions",
8967
11046
  provider: "zai",
8968
11047
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
8969
- compat: {"supportsDeveloperRole":false},
11048
+ compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
8970
11049
  reasoning: true,
8971
11050
  input: ["text"],
8972
11051
  cost: {
@@ -8984,7 +11063,7 @@ export const MODELS = {
8984
11063
  api: "openai-completions",
8985
11064
  provider: "zai",
8986
11065
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
8987
- compat: {"supportsDeveloperRole":false},
11066
+ compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
8988
11067
  reasoning: true,
8989
11068
  input: ["text"],
8990
11069
  cost: {
@@ -9002,7 +11081,7 @@ export const MODELS = {
9002
11081
  api: "openai-completions",
9003
11082
  provider: "zai",
9004
11083
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
9005
- compat: {"supportsDeveloperRole":false},
11084
+ compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
9006
11085
  reasoning: true,
9007
11086
  input: ["text", "image"],
9008
11087
  cost: {
@@ -9020,7 +11099,7 @@ export const MODELS = {
9020
11099
  api: "openai-completions",
9021
11100
  provider: "zai",
9022
11101
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
9023
- compat: {"supportsDeveloperRole":false},
11102
+ compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
9024
11103
  reasoning: true,
9025
11104
  input: ["text"],
9026
11105
  cost: {
@@ -9038,7 +11117,7 @@ export const MODELS = {
9038
11117
  api: "openai-completions",
9039
11118
  provider: "zai",
9040
11119
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
9041
- compat: {"supportsDeveloperRole":false},
11120
+ compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
9042
11121
  reasoning: true,
9043
11122
  input: ["text", "image"],
9044
11123
  cost: {
@@ -9056,7 +11135,7 @@ export const MODELS = {
9056
11135
  api: "openai-completions",
9057
11136
  provider: "zai",
9058
11137
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
9059
- compat: {"supportsDeveloperRole":false},
11138
+ compat: {"supportsDeveloperRole":false,"thinkingFormat":"zai"},
9060
11139
  reasoning: true,
9061
11140
  input: ["text"],
9062
11141
  cost: {