@fleetagent/pi-ai 0.2.4 → 0.2.5

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.
@@ -165,8 +165,8 @@ export const MODELS = {
165
165
  thinkingLevelMap: { "xhigh": "max" },
166
166
  input: ["text", "image"],
167
167
  cost: {
168
- input: 5,
169
- output: 25,
168
+ input: 5.5,
169
+ output: 27.5,
170
170
  cacheRead: 0.5,
171
171
  cacheWrite: 6.25,
172
172
  },
@@ -251,13 +251,13 @@ export const MODELS = {
251
251
  reasoning: true,
252
252
  input: ["text", "image"],
253
253
  cost: {
254
- input: 3,
255
- output: 15,
256
- cacheRead: 0.3,
257
- cacheWrite: 3.75,
254
+ input: 3.3,
255
+ output: 16.5,
256
+ cacheRead: 0.33,
257
+ cacheWrite: 4.125,
258
258
  },
259
259
  contextWindow: 1000000,
260
- maxTokens: 64000,
260
+ maxTokens: 128000,
261
261
  },
262
262
  "anthropic.claude-sonnet-5": {
263
263
  id: "anthropic.claude-sonnet-5",
@@ -327,6 +327,23 @@ export const MODELS = {
327
327
  contextWindow: 300000,
328
328
  maxTokens: 10000,
329
329
  },
330
+ "apac.anthropic.claude-sonnet-4-20250514-v1:0": {
331
+ id: "apac.anthropic.claude-sonnet-4-20250514-v1:0",
332
+ name: "Claude Sonnet 4 (APAC)",
333
+ api: "bedrock-converse-stream",
334
+ provider: "amazon-bedrock",
335
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
336
+ reasoning: true,
337
+ input: ["text", "image"],
338
+ cost: {
339
+ input: 3,
340
+ output: 15,
341
+ cacheRead: 0.3,
342
+ cacheWrite: 3.75,
343
+ },
344
+ contextWindow: 200000,
345
+ maxTokens: 64000,
346
+ },
330
347
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
331
348
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
332
349
  name: "Claude Haiku 4.5 (AU)",
@@ -336,10 +353,10 @@ export const MODELS = {
336
353
  reasoning: true,
337
354
  input: ["text", "image"],
338
355
  cost: {
339
- input: 1,
340
- output: 5,
341
- cacheRead: 0.1,
342
- cacheWrite: 1.25,
356
+ input: 1.1,
357
+ output: 5.5,
358
+ cacheRead: 0.11,
359
+ cacheWrite: 1.375,
343
360
  },
344
361
  contextWindow: 200000,
345
362
  maxTokens: 64000,
@@ -354,8 +371,8 @@ export const MODELS = {
354
371
  thinkingLevelMap: { "xhigh": "max" },
355
372
  input: ["text", "image"],
356
373
  cost: {
357
- input: 16.5,
358
- output: 82.5,
374
+ input: 5.5,
375
+ output: 27.5,
359
376
  cacheRead: 0.5,
360
377
  cacheWrite: 6.25,
361
378
  },
@@ -389,10 +406,10 @@ export const MODELS = {
389
406
  reasoning: true,
390
407
  input: ["text", "image"],
391
408
  cost: {
392
- input: 5,
393
- output: 25,
394
- cacheRead: 0.5,
395
- cacheWrite: 6.25,
409
+ input: 5.5,
410
+ output: 27.5,
411
+ cacheRead: 0.55,
412
+ cacheWrite: 6.875,
396
413
  },
397
414
  contextWindow: 1000000,
398
415
  maxTokens: 128000,
@@ -406,10 +423,10 @@ export const MODELS = {
406
423
  reasoning: true,
407
424
  input: ["text", "image"],
408
425
  cost: {
409
- input: 5,
410
- output: 25,
411
- cacheRead: 0.5,
412
- cacheWrite: 6.25,
426
+ input: 5.5,
427
+ output: 27.5,
428
+ cacheRead: 0.55,
429
+ cacheWrite: 6.875,
413
430
  },
414
431
  contextWindow: 1000000,
415
432
  maxTokens: 128000,
@@ -423,10 +440,10 @@ export const MODELS = {
423
440
  reasoning: true,
424
441
  input: ["text", "image"],
425
442
  cost: {
426
- input: 3,
427
- output: 15,
428
- cacheRead: 0.3,
429
- cacheWrite: 3.75,
443
+ input: 3.3,
444
+ output: 16.5,
445
+ cacheRead: 0.33,
446
+ cacheWrite: 4.125,
430
447
  },
431
448
  contextWindow: 200000,
432
449
  maxTokens: 64000,
@@ -457,10 +474,10 @@ export const MODELS = {
457
474
  reasoning: true,
458
475
  input: ["text", "image"],
459
476
  cost: {
460
- input: 2,
461
- output: 10,
462
- cacheRead: 0.2,
463
- cacheWrite: 2.5,
477
+ input: 2.2,
478
+ output: 11,
479
+ cacheRead: 0.22,
480
+ cacheWrite: 2.75,
464
481
  },
465
482
  contextWindow: 1000000,
466
483
  maxTokens: 128000,
@@ -705,6 +722,23 @@ export const MODELS = {
705
722
  contextWindow: 1000000,
706
723
  maxTokens: 128000,
707
724
  },
725
+ "eu.anthropic.claude-sonnet-4-20250514-v1:0": {
726
+ id: "eu.anthropic.claude-sonnet-4-20250514-v1:0",
727
+ name: "Claude Sonnet 4 (EU)",
728
+ api: "bedrock-converse-stream",
729
+ provider: "amazon-bedrock",
730
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
731
+ reasoning: true,
732
+ input: ["text", "image"],
733
+ cost: {
734
+ input: 3,
735
+ output: 15,
736
+ cacheRead: 0.3,
737
+ cacheWrite: 3.75,
738
+ },
739
+ contextWindow: 200000,
740
+ maxTokens: 64000,
741
+ },
708
742
  "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": {
709
743
  id: "eu.anthropic.claude-sonnet-4-5-20250929-v1:0",
710
744
  name: "Claude Sonnet 4.5 (EU)",
@@ -737,7 +771,7 @@ export const MODELS = {
737
771
  cacheWrite: 4.125,
738
772
  },
739
773
  contextWindow: 1000000,
740
- maxTokens: 64000,
774
+ maxTokens: 128000,
741
775
  },
742
776
  "eu.anthropic.claude-sonnet-5": {
743
777
  id: "eu.anthropic.claude-sonnet-5",
@@ -928,6 +962,23 @@ export const MODELS = {
928
962
  contextWindow: 1000000,
929
963
  maxTokens: 128000,
930
964
  },
965
+ "global.anthropic.claude-sonnet-4-20250514-v1:0": {
966
+ id: "global.anthropic.claude-sonnet-4-20250514-v1:0",
967
+ name: "Claude Sonnet 4 (Global)",
968
+ api: "bedrock-converse-stream",
969
+ provider: "amazon-bedrock",
970
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
971
+ reasoning: true,
972
+ input: ["text", "image"],
973
+ cost: {
974
+ input: 3,
975
+ output: 15,
976
+ cacheRead: 0.3,
977
+ cacheWrite: 3.75,
978
+ },
979
+ contextWindow: 200000,
980
+ maxTokens: 64000,
981
+ },
931
982
  "global.anthropic.claude-sonnet-4-5-20250929-v1:0": {
932
983
  id: "global.anthropic.claude-sonnet-4-5-20250929-v1:0",
933
984
  name: "Claude Sonnet 4.5 (Global)",
@@ -960,7 +1011,7 @@ export const MODELS = {
960
1011
  cacheWrite: 3.75,
961
1012
  },
962
1013
  contextWindow: 1000000,
963
- maxTokens: 64000,
1014
+ maxTokens: 128000,
964
1015
  },
965
1016
  "global.anthropic.claude-sonnet-5": {
966
1017
  id: "global.anthropic.claude-sonnet-5",
@@ -1181,10 +1232,10 @@ export const MODELS = {
1181
1232
  reasoning: true,
1182
1233
  input: ["text", "image"],
1183
1234
  cost: {
1184
- input: 1,
1185
- output: 5,
1186
- cacheRead: 0.1,
1187
- cacheWrite: 1.25,
1235
+ input: 1.1,
1236
+ output: 5.5,
1237
+ cacheRead: 0.11,
1238
+ cacheWrite: 1.375,
1188
1239
  },
1189
1240
  contextWindow: 200000,
1190
1241
  maxTokens: 64000,
@@ -1199,10 +1250,10 @@ export const MODELS = {
1199
1250
  thinkingLevelMap: { "xhigh": "xhigh" },
1200
1251
  input: ["text", "image"],
1201
1252
  cost: {
1202
- input: 5,
1203
- output: 25,
1204
- cacheRead: 0.5,
1205
- cacheWrite: 6.25,
1253
+ input: 5.5,
1254
+ output: 27.5,
1255
+ cacheRead: 0.55,
1256
+ cacheWrite: 6.875,
1206
1257
  },
1207
1258
  contextWindow: 1000000,
1208
1259
  maxTokens: 128000,
@@ -1216,10 +1267,10 @@ export const MODELS = {
1216
1267
  reasoning: true,
1217
1268
  input: ["text", "image"],
1218
1269
  cost: {
1219
- input: 5,
1220
- output: 25,
1221
- cacheRead: 0.5,
1222
- cacheWrite: 6.25,
1270
+ input: 5.5,
1271
+ output: 27.5,
1272
+ cacheRead: 0.55,
1273
+ cacheWrite: 6.875,
1223
1274
  },
1224
1275
  contextWindow: 1000000,
1225
1276
  maxTokens: 128000,
@@ -1233,10 +1284,10 @@ export const MODELS = {
1233
1284
  reasoning: true,
1234
1285
  input: ["text", "image"],
1235
1286
  cost: {
1236
- input: 5,
1237
- output: 25,
1238
- cacheRead: 0.5,
1239
- cacheWrite: 6.25,
1287
+ input: 5.5,
1288
+ output: 27.5,
1289
+ cacheRead: 0.55,
1290
+ cacheWrite: 6.875,
1240
1291
  },
1241
1292
  contextWindow: 1000000,
1242
1293
  maxTokens: 128000,
@@ -1250,10 +1301,10 @@ export const MODELS = {
1250
1301
  reasoning: true,
1251
1302
  input: ["text", "image"],
1252
1303
  cost: {
1253
- input: 3,
1254
- output: 15,
1255
- cacheRead: 0.3,
1256
- cacheWrite: 3.75,
1304
+ input: 3.3,
1305
+ output: 16.5,
1306
+ cacheRead: 0.33,
1307
+ cacheWrite: 4.125,
1257
1308
  },
1258
1309
  contextWindow: 200000,
1259
1310
  maxTokens: 64000,
@@ -1267,13 +1318,13 @@ export const MODELS = {
1267
1318
  reasoning: true,
1268
1319
  input: ["text", "image"],
1269
1320
  cost: {
1270
- input: 3,
1271
- output: 15,
1272
- cacheRead: 0.3,
1273
- cacheWrite: 3.75,
1321
+ input: 3.3,
1322
+ output: 16.5,
1323
+ cacheRead: 0.33,
1324
+ cacheWrite: 4.125,
1274
1325
  },
1275
1326
  contextWindow: 1000000,
1276
- maxTokens: 64000,
1327
+ maxTokens: 128000,
1277
1328
  },
1278
1329
  "jp.anthropic.claude-sonnet-5": {
1279
1330
  id: "jp.anthropic.claude-sonnet-5",
@@ -1284,10 +1335,10 @@ export const MODELS = {
1284
1335
  reasoning: true,
1285
1336
  input: ["text", "image"],
1286
1337
  cost: {
1287
- input: 2,
1288
- output: 10,
1289
- cacheRead: 0.2,
1290
- cacheWrite: 2.5,
1338
+ input: 2.2,
1339
+ output: 11,
1340
+ cacheRead: 0.22,
1341
+ cacheWrite: 2.75,
1291
1342
  },
1292
1343
  contextWindow: 1000000,
1293
1344
  maxTokens: 128000,
@@ -2140,10 +2191,10 @@ export const MODELS = {
2140
2191
  reasoning: true,
2141
2192
  input: ["text", "image"],
2142
2193
  cost: {
2143
- input: 10,
2144
- output: 50,
2145
- cacheRead: 1,
2146
- cacheWrite: 12.5,
2194
+ input: 11,
2195
+ output: 55,
2196
+ cacheRead: 1.1,
2197
+ cacheWrite: 13.75,
2147
2198
  },
2148
2199
  contextWindow: 1000000,
2149
2200
  maxTokens: 128000,
@@ -2174,10 +2225,10 @@ export const MODELS = {
2174
2225
  reasoning: true,
2175
2226
  input: ["text", "image"],
2176
2227
  cost: {
2177
- input: 1,
2178
- output: 5,
2179
- cacheRead: 0.1,
2180
- cacheWrite: 1.25,
2228
+ input: 1.1,
2229
+ output: 5.5,
2230
+ cacheRead: 0.11,
2231
+ cacheWrite: 1.375,
2181
2232
  },
2182
2233
  contextWindow: 200000,
2183
2234
  maxTokens: 64000,
@@ -2208,10 +2259,10 @@ export const MODELS = {
2208
2259
  reasoning: true,
2209
2260
  input: ["text", "image"],
2210
2261
  cost: {
2211
- input: 5,
2212
- output: 25,
2213
- cacheRead: 0.5,
2214
- cacheWrite: 6.25,
2262
+ input: 5.5,
2263
+ output: 27.5,
2264
+ cacheRead: 0.55,
2265
+ cacheWrite: 6.875,
2215
2266
  },
2216
2267
  contextWindow: 200000,
2217
2268
  maxTokens: 64000,
@@ -2226,8 +2277,8 @@ export const MODELS = {
2226
2277
  thinkingLevelMap: { "xhigh": "max" },
2227
2278
  input: ["text", "image"],
2228
2279
  cost: {
2229
- input: 5,
2230
- output: 25,
2280
+ input: 5.5,
2281
+ output: 27.5,
2231
2282
  cacheRead: 0.5,
2232
2283
  cacheWrite: 6.25,
2233
2284
  },
@@ -2244,10 +2295,10 @@ export const MODELS = {
2244
2295
  thinkingLevelMap: { "xhigh": "xhigh" },
2245
2296
  input: ["text", "image"],
2246
2297
  cost: {
2247
- input: 5,
2248
- output: 25,
2249
- cacheRead: 0.5,
2250
- cacheWrite: 6.25,
2298
+ input: 5.5,
2299
+ output: 27.5,
2300
+ cacheRead: 0.55,
2301
+ cacheWrite: 6.875,
2251
2302
  },
2252
2303
  contextWindow: 1000000,
2253
2304
  maxTokens: 128000,
@@ -2261,10 +2312,10 @@ export const MODELS = {
2261
2312
  reasoning: true,
2262
2313
  input: ["text", "image"],
2263
2314
  cost: {
2264
- input: 5,
2265
- output: 25,
2266
- cacheRead: 0.5,
2267
- cacheWrite: 6.25,
2315
+ input: 5.5,
2316
+ output: 27.5,
2317
+ cacheRead: 0.55,
2318
+ cacheWrite: 6.875,
2268
2319
  },
2269
2320
  contextWindow: 1000000,
2270
2321
  maxTokens: 128000,
@@ -2278,17 +2329,17 @@ export const MODELS = {
2278
2329
  reasoning: true,
2279
2330
  input: ["text", "image"],
2280
2331
  cost: {
2281
- input: 5,
2282
- output: 25,
2283
- cacheRead: 0.5,
2284
- cacheWrite: 6.25,
2332
+ input: 5.5,
2333
+ output: 27.5,
2334
+ cacheRead: 0.55,
2335
+ cacheWrite: 6.875,
2285
2336
  },
2286
2337
  contextWindow: 1000000,
2287
2338
  maxTokens: 128000,
2288
2339
  },
2289
- "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
2290
- id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
2291
- name: "Claude Sonnet 4.5 (US)",
2340
+ "us.anthropic.claude-sonnet-4-20250514-v1:0": {
2341
+ id: "us.anthropic.claude-sonnet-4-20250514-v1:0",
2342
+ name: "Claude Sonnet 4 (US)",
2292
2343
  api: "bedrock-converse-stream",
2293
2344
  provider: "amazon-bedrock",
2294
2345
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -2303,6 +2354,23 @@ export const MODELS = {
2303
2354
  contextWindow: 200000,
2304
2355
  maxTokens: 64000,
2305
2356
  },
2357
+ "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
2358
+ id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
2359
+ name: "Claude Sonnet 4.5 (US)",
2360
+ api: "bedrock-converse-stream",
2361
+ provider: "amazon-bedrock",
2362
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2363
+ reasoning: true,
2364
+ input: ["text", "image"],
2365
+ cost: {
2366
+ input: 3.3,
2367
+ output: 16.5,
2368
+ cacheRead: 0.33,
2369
+ cacheWrite: 4.125,
2370
+ },
2371
+ contextWindow: 200000,
2372
+ maxTokens: 64000,
2373
+ },
2306
2374
  "us.anthropic.claude-sonnet-4-6": {
2307
2375
  id: "us.anthropic.claude-sonnet-4-6",
2308
2376
  name: "Claude Sonnet 4.6 (US)",
@@ -2312,13 +2380,13 @@ export const MODELS = {
2312
2380
  reasoning: true,
2313
2381
  input: ["text", "image"],
2314
2382
  cost: {
2315
- input: 3,
2316
- output: 15,
2317
- cacheRead: 0.3,
2318
- cacheWrite: 3.75,
2383
+ input: 3.3,
2384
+ output: 16.5,
2385
+ cacheRead: 0.33,
2386
+ cacheWrite: 4.125,
2319
2387
  },
2320
2388
  contextWindow: 1000000,
2321
- maxTokens: 64000,
2389
+ maxTokens: 128000,
2322
2390
  },
2323
2391
  "us.anthropic.claude-sonnet-5": {
2324
2392
  id: "us.anthropic.claude-sonnet-5",
@@ -2329,10 +2397,10 @@ export const MODELS = {
2329
2397
  reasoning: true,
2330
2398
  input: ["text", "image"],
2331
2399
  cost: {
2332
- input: 2,
2333
- output: 10,
2334
- cacheRead: 0.2,
2335
- cacheWrite: 2.5,
2400
+ input: 2.2,
2401
+ output: 11,
2402
+ cacheRead: 0.22,
2403
+ cacheWrite: 2.75,
2336
2404
  },
2337
2405
  contextWindow: 1000000,
2338
2406
  maxTokens: 128000,
@@ -4695,9 +4763,9 @@ export const MODELS = {
4695
4763
  contextWindow: 1000000,
4696
4764
  maxTokens: 384000,
4697
4765
  },
4698
- "accounts/fireworks/models/deepseek-v4-pro-0813": {
4699
- id: "accounts/fireworks/models/deepseek-v4-pro-0813",
4700
- name: "DeepSeek V4 Pro 0813",
4766
+ "accounts/fireworks/models/deepseek-v4-pro": {
4767
+ id: "accounts/fireworks/models/deepseek-v4-pro",
4768
+ name: "DeepSeek V4 Pro",
4701
4769
  api: "anthropic-messages",
4702
4770
  provider: "fireworks",
4703
4771
  baseUrl: "https://api.fireworks.ai/inference",
@@ -4705,8 +4773,26 @@ export const MODELS = {
4705
4773
  reasoning: true,
4706
4774
  input: ["text"],
4707
4775
  cost: {
4708
- input: 1.32,
4709
- output: 3.96,
4776
+ input: 1.2,
4777
+ output: 1.2,
4778
+ cacheRead: 0.6,
4779
+ cacheWrite: 0,
4780
+ },
4781
+ contextWindow: 1000000,
4782
+ maxTokens: 384000,
4783
+ },
4784
+ "accounts/fireworks/models/deepseek-v4-pro-0813": {
4785
+ id: "accounts/fireworks/models/deepseek-v4-pro-0813",
4786
+ name: "DeepSeek V4 Pro 0813",
4787
+ api: "anthropic-messages",
4788
+ provider: "fireworks",
4789
+ baseUrl: "https://api.fireworks.ai/inference",
4790
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4791
+ reasoning: true,
4792
+ input: ["text"],
4793
+ cost: {
4794
+ input: 1.32,
4795
+ output: 3.96,
4710
4796
  cacheRead: 0.044,
4711
4797
  cacheWrite: 0,
4712
4798
  },
@@ -4875,41 +4961,41 @@ export const MODELS = {
4875
4961
  contextWindow: 1048576,
4876
4962
  maxTokens: 131072,
4877
4963
  },
4878
- "accounts/fireworks/models/minimax-m3": {
4879
- id: "accounts/fireworks/models/minimax-m3",
4880
- name: "MiniMax-M3",
4964
+ "accounts/fireworks/models/minimax-m2p7": {
4965
+ id: "accounts/fireworks/models/minimax-m2p7",
4966
+ name: "MiniMax-M2.7",
4881
4967
  api: "anthropic-messages",
4882
4968
  provider: "fireworks",
4883
4969
  baseUrl: "https://api.fireworks.ai/inference",
4884
4970
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4885
4971
  reasoning: true,
4886
- input: ["text", "image"],
4972
+ input: ["text"],
4887
4973
  cost: {
4888
- input: 0.3,
4974
+ input: 1.2,
4889
4975
  output: 1.2,
4890
- cacheRead: 0.06,
4976
+ cacheRead: 0.6,
4891
4977
  cacheWrite: 0,
4892
4978
  },
4893
- contextWindow: 512000,
4894
- maxTokens: 512000,
4979
+ contextWindow: 196608,
4980
+ maxTokens: 131072,
4895
4981
  },
4896
- "accounts/fireworks/models/mistral-large-3-fp8": {
4897
- id: "accounts/fireworks/models/mistral-large-3-fp8",
4898
- name: "Mistral Large 3 675B Instruct 2512",
4982
+ "accounts/fireworks/models/minimax-m3": {
4983
+ id: "accounts/fireworks/models/minimax-m3",
4984
+ name: "MiniMax-M3",
4899
4985
  api: "anthropic-messages",
4900
4986
  provider: "fireworks",
4901
4987
  baseUrl: "https://api.fireworks.ai/inference",
4902
4988
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4903
- reasoning: false,
4904
- input: ["text", "image"],
4989
+ reasoning: true,
4990
+ input: ["text"],
4905
4991
  cost: {
4906
- input: 0,
4907
- output: 0,
4908
- cacheRead: 0,
4992
+ input: 0.3,
4993
+ output: 1.2,
4994
+ cacheRead: 0.06,
4909
4995
  cacheWrite: 0,
4910
4996
  },
4911
- contextWindow: 262144,
4912
- maxTokens: 262144,
4997
+ contextWindow: 512000,
4998
+ maxTokens: 512000,
4913
4999
  },
4914
5000
  "accounts/fireworks/models/muse-glimmer-30b": {
4915
5001
  id: "accounts/fireworks/models/muse-glimmer-30b",
@@ -4941,7 +5027,7 @@ export const MODELS = {
4941
5027
  cost: {
4942
5028
  input: 0.6,
4943
5029
  output: 2.4,
4944
- cacheRead: 0.119,
5030
+ cacheRead: 0.12,
4945
5031
  cacheWrite: 0,
4946
5032
  },
4947
5033
  contextWindow: 262144,
@@ -5009,7 +5095,7 @@ export const MODELS = {
5009
5095
  baseUrl: "https://api.fireworks.ai/inference",
5010
5096
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5011
5097
  reasoning: true,
5012
- input: ["text"],
5098
+ input: ["text", "image"],
5013
5099
  cost: {
5014
5100
  input: 2,
5015
5101
  output: 6,
@@ -5019,6 +5105,42 @@ export const MODELS = {
5019
5105
  contextWindow: 262144,
5020
5106
  maxTokens: 131072,
5021
5107
  },
5108
+ "accounts/fireworks/routers/deepseek-flash-latest": {
5109
+ id: "accounts/fireworks/routers/deepseek-flash-latest",
5110
+ name: "DeepSeek Flash Latest",
5111
+ api: "anthropic-messages",
5112
+ provider: "fireworks",
5113
+ baseUrl: "https://api.fireworks.ai/inference",
5114
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5115
+ reasoning: true,
5116
+ input: ["text", "image"],
5117
+ cost: {
5118
+ input: 0.22,
5119
+ output: 0.66,
5120
+ cacheRead: 0.007,
5121
+ cacheWrite: 0,
5122
+ },
5123
+ contextWindow: 1000000,
5124
+ maxTokens: 384000,
5125
+ },
5126
+ "accounts/fireworks/routers/deepseek-pro-latest": {
5127
+ id: "accounts/fireworks/routers/deepseek-pro-latest",
5128
+ name: "DeepSeek Pro Latest",
5129
+ api: "anthropic-messages",
5130
+ provider: "fireworks",
5131
+ baseUrl: "https://api.fireworks.ai/inference",
5132
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5133
+ reasoning: true,
5134
+ input: ["text"],
5135
+ cost: {
5136
+ input: 1.32,
5137
+ output: 3.96,
5138
+ cacheRead: 0.044,
5139
+ cacheWrite: 0,
5140
+ },
5141
+ contextWindow: 1000000,
5142
+ maxTokens: 384000,
5143
+ },
5022
5144
  "accounts/fireworks/routers/glm-5p2-fast": {
5023
5145
  id: "accounts/fireworks/routers/glm-5p2-fast",
5024
5146
  name: "GLM 5.2 Fast",
@@ -5055,6 +5177,78 @@ export const MODELS = {
5055
5177
  contextWindow: 1048572,
5056
5178
  maxTokens: 262144,
5057
5179
  },
5180
+ "accounts/fireworks/routers/glm-fast-latest": {
5181
+ id: "accounts/fireworks/routers/glm-fast-latest",
5182
+ name: "GLM 5.3 Fast (Latest)",
5183
+ api: "anthropic-messages",
5184
+ provider: "fireworks",
5185
+ baseUrl: "https://api.fireworks.ai/inference",
5186
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5187
+ reasoning: true,
5188
+ input: ["text"],
5189
+ cost: {
5190
+ input: 2.1,
5191
+ output: 6.6,
5192
+ cacheRead: 0.39,
5193
+ cacheWrite: 0,
5194
+ },
5195
+ contextWindow: 1048572,
5196
+ maxTokens: 262144,
5197
+ },
5198
+ "accounts/fireworks/routers/glm-flash-latest": {
5199
+ id: "accounts/fireworks/routers/glm-flash-latest",
5200
+ name: "GLM Flash Latest (GLM 5.3 Flash)",
5201
+ api: "anthropic-messages",
5202
+ provider: "fireworks",
5203
+ baseUrl: "https://api.fireworks.ai/inference",
5204
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5205
+ reasoning: true,
5206
+ input: ["text", "image"],
5207
+ cost: {
5208
+ input: 0.15,
5209
+ output: 0.5,
5210
+ cacheRead: 0.03,
5211
+ cacheWrite: 0,
5212
+ },
5213
+ contextWindow: 1048573,
5214
+ maxTokens: 131072,
5215
+ },
5216
+ "accounts/fireworks/routers/glm-latest": {
5217
+ id: "accounts/fireworks/routers/glm-latest",
5218
+ name: "GLM Latest",
5219
+ api: "anthropic-messages",
5220
+ provider: "fireworks",
5221
+ baseUrl: "https://api.fireworks.ai/inference",
5222
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5223
+ reasoning: true,
5224
+ input: ["text"],
5225
+ cost: {
5226
+ input: 1.4,
5227
+ output: 4.4,
5228
+ cacheRead: 0.26,
5229
+ cacheWrite: 0,
5230
+ },
5231
+ contextWindow: 1048573,
5232
+ maxTokens: 262144,
5233
+ },
5234
+ "accounts/fireworks/routers/kimi-fast-latest": {
5235
+ id: "accounts/fireworks/routers/kimi-fast-latest",
5236
+ name: "Kimi Fast Latest",
5237
+ api: "anthropic-messages",
5238
+ provider: "fireworks",
5239
+ baseUrl: "https://api.fireworks.ai/inference",
5240
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5241
+ reasoning: true,
5242
+ input: ["text", "image"],
5243
+ cost: {
5244
+ input: 4.5,
5245
+ output: 22.5,
5246
+ cacheRead: 0.45,
5247
+ cacheWrite: 0,
5248
+ },
5249
+ contextWindow: 1048576,
5250
+ maxTokens: 131072,
5251
+ },
5058
5252
  "accounts/fireworks/routers/kimi-k2p6-turbo": {
5059
5253
  id: "accounts/fireworks/routers/kimi-k2p6-turbo",
5060
5254
  name: "Kimi K2.6 Turbo",
@@ -5091,6 +5285,60 @@ export const MODELS = {
5091
5285
  contextWindow: 1048576,
5092
5286
  maxTokens: 131072,
5093
5287
  },
5288
+ "accounts/fireworks/routers/kimi-latest": {
5289
+ id: "accounts/fireworks/routers/kimi-latest",
5290
+ name: "Kimi Latest",
5291
+ api: "anthropic-messages",
5292
+ provider: "fireworks",
5293
+ baseUrl: "https://api.fireworks.ai/inference",
5294
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5295
+ reasoning: true,
5296
+ input: ["text", "image"],
5297
+ cost: {
5298
+ input: 3,
5299
+ output: 15,
5300
+ cacheRead: 0.3,
5301
+ cacheWrite: 0,
5302
+ },
5303
+ contextWindow: 1048576,
5304
+ maxTokens: 131072,
5305
+ },
5306
+ "accounts/fireworks/routers/minimax-latest": {
5307
+ id: "accounts/fireworks/routers/minimax-latest",
5308
+ name: "MiniMax Latest",
5309
+ api: "anthropic-messages",
5310
+ provider: "fireworks",
5311
+ baseUrl: "https://api.fireworks.ai/inference",
5312
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5313
+ reasoning: true,
5314
+ input: ["text"],
5315
+ cost: {
5316
+ input: 0.3,
5317
+ output: 1.2,
5318
+ cacheRead: 0.06,
5319
+ cacheWrite: 0,
5320
+ },
5321
+ contextWindow: 512000,
5322
+ maxTokens: 512000,
5323
+ },
5324
+ "accounts/fireworks/routers/qwen-max-latest": {
5325
+ id: "accounts/fireworks/routers/qwen-max-latest",
5326
+ name: "Qwen Max Latest (Qwen3.8 Max)",
5327
+ api: "anthropic-messages",
5328
+ provider: "fireworks",
5329
+ baseUrl: "https://api.fireworks.ai/inference",
5330
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5331
+ reasoning: true,
5332
+ input: ["text", "image"],
5333
+ cost: {
5334
+ input: 2,
5335
+ output: 6,
5336
+ cacheRead: 0.25,
5337
+ cacheWrite: 0,
5338
+ },
5339
+ contextWindow: 262144,
5340
+ maxTokens: 131072,
5341
+ },
5094
5342
  },
5095
5343
  "github-copilot": {
5096
5344
  "claude-fable-5": {
@@ -7757,7 +8005,7 @@ export const MODELS = {
7757
8005
  },
7758
8006
  "kimi-for-coding": {
7759
8007
  id: "kimi-for-coding",
7760
- name: "Kimi K2.7 Code",
8008
+ name: "kimi-for-coding",
7761
8009
  api: "anthropic-messages",
7762
8010
  provider: "kimi-coding",
7763
8011
  baseUrl: "https://api.kimi.com/coding",
@@ -7770,7 +8018,7 @@ export const MODELS = {
7770
8018
  cacheRead: 0,
7771
8019
  cacheWrite: 0,
7772
8020
  },
7773
- contextWindow: 262144,
8021
+ contextWindow: 1048576,
7774
8022
  maxTokens: 32768,
7775
8023
  },
7776
8024
  "kimi-for-coding-highspeed": {
@@ -10698,6 +10946,23 @@ export const MODELS = {
10698
10946
  contextWindow: 262144,
10699
10947
  maxTokens: 65536,
10700
10948
  },
10949
+ "union-alpha": {
10950
+ id: "union-alpha",
10951
+ name: "Union Alpha Free",
10952
+ api: "anthropic-messages",
10953
+ provider: "opencode",
10954
+ baseUrl: "https://opencode.ai/zen",
10955
+ reasoning: true,
10956
+ input: ["text", "image"],
10957
+ cost: {
10958
+ input: 0,
10959
+ output: 0,
10960
+ cacheRead: 0,
10961
+ cacheWrite: 0,
10962
+ },
10963
+ contextWindow: 262144,
10964
+ maxTokens: 131072,
10965
+ },
10701
10966
  },
10702
10967
  "opencode-go": {
10703
10968
  "deepseek-v4-flash": {
@@ -11171,6 +11436,23 @@ export const MODELS = {
11171
11436
  contextWindow: 1000000,
11172
11437
  maxTokens: 131072,
11173
11438
  },
11439
+ "union-alpha": {
11440
+ id: "union-alpha",
11441
+ name: "Union Alpha Free",
11442
+ api: "anthropic-messages",
11443
+ provider: "opencode-go",
11444
+ baseUrl: "https://opencode.ai/zen/go",
11445
+ reasoning: true,
11446
+ input: ["text", "image"],
11447
+ cost: {
11448
+ input: 0,
11449
+ output: 0,
11450
+ cacheRead: 0,
11451
+ cacheWrite: 0,
11452
+ },
11453
+ contextWindow: 262144,
11454
+ maxTokens: 131072,
11455
+ },
11174
11456
  },
11175
11457
  "openrouter": {
11176
11458
  "aion-labs/aion-2.0": {
@@ -11621,7 +11903,7 @@ export const MODELS = {
11621
11903
  },
11622
11904
  "anthropic/claude-opus-5": {
11623
11905
  id: "anthropic/claude-opus-5",
11624
- name: "Claude Opus 5",
11906
+ name: "Anthropic: Claude Opus 5",
11625
11907
  api: "openai-completions",
11626
11908
  provider: "openrouter",
11627
11909
  baseUrl: "https://openrouter.ai/api/v1",
@@ -11638,7 +11920,7 @@ export const MODELS = {
11638
11920
  },
11639
11921
  "anthropic/claude-opus-5:batch": {
11640
11922
  id: "anthropic/claude-opus-5:batch",
11641
- name: "Claude Opus 5 (batch)",
11923
+ name: "Anthropic: Claude Opus 5 (batch)",
11642
11924
  api: "openai-completions",
11643
11925
  provider: "openrouter",
11644
11926
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12106,9 +12388,9 @@ export const MODELS = {
12106
12388
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12107
12389
  input: ["text"],
12108
12390
  cost: {
12109
- input: 0.06678,
12110
- output: 0.13356,
12111
- cacheRead: 0.013356,
12391
+ input: 0.07,
12392
+ output: 0.14,
12393
+ cacheRead: 0.014,
12112
12394
  cacheWrite: 0,
12113
12395
  },
12114
12396
  contextWindow: 1048576,
@@ -12125,9 +12407,9 @@ export const MODELS = {
12125
12407
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12126
12408
  input: ["text"],
12127
12409
  cost: {
12128
- input: 0.04,
12129
- output: 0.08,
12130
- cacheRead: 0.008,
12410
+ input: 0.06,
12411
+ output: 0.12,
12412
+ cacheRead: 0.012,
12131
12413
  cacheWrite: 0,
12132
12414
  },
12133
12415
  contextWindow: 1310720,
@@ -12201,13 +12483,13 @@ export const MODELS = {
12201
12483
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12202
12484
  input: ["text"],
12203
12485
  cost: {
12204
- input: 0.788916,
12205
- output: 1.577832,
12206
- cacheRead: 0.065743,
12486
+ input: 1.5999999999999999,
12487
+ output: 3.1999999999999997,
12488
+ cacheRead: 0.135,
12207
12489
  cacheWrite: 0,
12208
12490
  },
12209
12491
  contextWindow: 1048576,
12210
- maxTokens: 384000,
12492
+ maxTokens: 393216,
12211
12493
  },
12212
12494
  "deepseek/deepseek-v4-pro-0813": {
12213
12495
  id: "deepseek/deepseek-v4-pro-0813",
@@ -12220,9 +12502,9 @@ export const MODELS = {
12220
12502
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12221
12503
  input: ["text"],
12222
12504
  cost: {
12223
- input: 0.57816,
12224
- output: 1.73448,
12225
- cacheRead: 0.018396000000000003,
12505
+ input: 0.57948,
12506
+ output: 1.73844,
12507
+ cacheRead: 0.018438,
12226
12508
  cacheWrite: 0,
12227
12509
  },
12228
12510
  contextWindow: 1048576,
@@ -12385,23 +12667,6 @@ export const MODELS = {
12385
12667
  contextWindow: 1048576,
12386
12668
  maxTokens: 65536,
12387
12669
  },
12388
- "google/gemini-2.5-pro-preview-05-06": {
12389
- id: "google/gemini-2.5-pro-preview-05-06",
12390
- name: "Google: Gemini 2.5 Pro Preview 05-06",
12391
- api: "openai-completions",
12392
- provider: "openrouter",
12393
- baseUrl: "https://openrouter.ai/api/v1",
12394
- reasoning: true,
12395
- input: ["text", "image"],
12396
- cost: {
12397
- input: 1.25,
12398
- output: 10,
12399
- cacheRead: 0.125,
12400
- cacheWrite: 0.375,
12401
- },
12402
- contextWindow: 1048576,
12403
- maxTokens: 65535,
12404
- },
12405
12670
  "google/gemini-2.5-pro:batch": {
12406
12671
  id: "google/gemini-2.5-pro:batch",
12407
12672
  name: "Google: Gemini 2.5 Pro (batch)",
@@ -12785,13 +13050,13 @@ export const MODELS = {
12785
13050
  reasoning: true,
12786
13051
  input: ["text", "image"],
12787
13052
  cost: {
12788
- input: 0.041999999999999996,
12789
- output: 0.22,
12790
- cacheRead: 0,
13053
+ input: 0.09,
13054
+ output: 0.3,
13055
+ cacheRead: 0.049999999999999996,
12791
13056
  cacheWrite: 0,
12792
13057
  },
12793
13058
  contextWindow: 262144,
12794
- maxTokens: 32768,
13059
+ maxTokens: 235929,
12795
13060
  },
12796
13061
  "google/gemma-4-26b-a4b-it:free": {
12797
13062
  id: "google/gemma-4-26b-a4b-it:free",
@@ -12827,26 +13092,9 @@ export const MODELS = {
12827
13092
  contextWindow: 262144,
12828
13093
  maxTokens: 16384,
12829
13094
  },
12830
- "google/gemma-4-31b-it:batch": {
12831
- id: "google/gemma-4-31b-it:batch",
12832
- name: "Google: Gemma 4 31B (batch)",
12833
- api: "openai-completions",
12834
- provider: "openrouter",
12835
- baseUrl: "https://openrouter.ai/api/v1",
12836
- reasoning: true,
12837
- input: ["text", "image"],
12838
- cost: {
12839
- input: 0.39,
12840
- output: 0.9700000000000001,
12841
- cacheRead: 0,
12842
- cacheWrite: 0,
12843
- },
12844
- contextWindow: 262144,
12845
- maxTokens: 235929,
12846
- },
12847
- "google/gemma-4-31b-it:free": {
12848
- id: "google/gemma-4-31b-it:free",
12849
- name: "Google: Gemma 4 31B (free)",
13095
+ "google/gemma-4-31b-it:free": {
13096
+ id: "google/gemma-4-31b-it:free",
13097
+ name: "Google: Gemma 4 31B (free)",
12850
13098
  api: "openai-completions",
12851
13099
  provider: "openrouter",
12852
13100
  baseUrl: "https://openrouter.ai/api/v1",
@@ -13093,13 +13341,13 @@ export const MODELS = {
13093
13341
  reasoning: false,
13094
13342
  input: ["text"],
13095
13343
  cost: {
13096
- input: 0.72,
13097
- output: 0.72,
13344
+ input: 0.39999999999999997,
13345
+ output: 0.39999999999999997,
13098
13346
  cacheRead: 0,
13099
13347
  cacheWrite: 0,
13100
13348
  },
13101
13349
  contextWindow: 131072,
13102
- maxTokens: 8192,
13350
+ maxTokens: 16384,
13103
13351
  },
13104
13352
  "meta-llama/llama-3.1-8b-instruct": {
13105
13353
  id: "meta-llama/llama-3.1-8b-instruct",
@@ -13144,13 +13392,13 @@ export const MODELS = {
13144
13392
  reasoning: false,
13145
13393
  input: ["text", "image"],
13146
13394
  cost: {
13147
- input: 0.19999999999999998,
13148
- output: 0.696,
13395
+ input: 0.1875,
13396
+ output: 0.6525,
13149
13397
  cacheRead: 0,
13150
13398
  cacheWrite: 0,
13151
13399
  },
13152
13400
  contextWindow: 1048576,
13153
- maxTokens: 115200,
13401
+ maxTokens: 16384,
13154
13402
  },
13155
13403
  "meta-llama/llama-4-scout": {
13156
13404
  id: "meta-llama/llama-4-scout",
@@ -13297,7 +13545,7 @@ export const MODELS = {
13297
13545
  reasoning: true,
13298
13546
  input: ["text"],
13299
13547
  cost: {
13300
- input: 0.55,
13548
+ input: 0.39999999999999997,
13301
13549
  output: 2.2,
13302
13550
  cacheRead: 0,
13303
13551
  cacheWrite: 0,
@@ -13756,8 +14004,8 @@ export const MODELS = {
13756
14004
  reasoning: false,
13757
14005
  input: ["text", "image"],
13758
14006
  cost: {
13759
- input: 0.075,
13760
- output: 0.19999999999999998,
14007
+ input: 0.09375,
14008
+ output: 0.25,
13761
14009
  cacheRead: 0,
13762
14010
  cacheWrite: 0,
13763
14011
  },
@@ -13813,7 +14061,7 @@ export const MODELS = {
13813
14061
  cacheWrite: 0,
13814
14062
  },
13815
14063
  contextWindow: 131072,
13816
- maxTokens: 100352,
14064
+ maxTokens: 98304,
13817
14065
  },
13818
14066
  "moonshotai/kimi-k2-0905": {
13819
14067
  id: "moonshotai/kimi-k2-0905",
@@ -13830,7 +14078,7 @@ export const MODELS = {
13830
14078
  cacheWrite: 0,
13831
14079
  },
13832
14080
  contextWindow: 262144,
13833
- maxTokens: 100352,
14081
+ maxTokens: 98304,
13834
14082
  },
13835
14083
  "moonshotai/kimi-k2-thinking": {
13836
14084
  id: "moonshotai/kimi-k2-thinking",
@@ -13843,11 +14091,11 @@ export const MODELS = {
13843
14091
  cost: {
13844
14092
  input: 0.6,
13845
14093
  output: 2.5,
13846
- cacheRead: 0,
14094
+ cacheRead: 0.15,
13847
14095
  cacheWrite: 0,
13848
14096
  },
13849
14097
  contextWindow: 262144,
13850
- maxTokens: 235929,
14098
+ maxTokens: 98304,
13851
14099
  },
13852
14100
  "moonshotai/kimi-k2.5": {
13853
14101
  id: "moonshotai/kimi-k2.5",
@@ -13893,9 +14141,9 @@ export const MODELS = {
13893
14141
  reasoning: true,
13894
14142
  input: ["text", "image"],
13895
14143
  cost: {
13896
- input: 0.71,
13897
- output: 3.5,
13898
- cacheRead: 0.15,
14144
+ input: 0.7062,
14145
+ output: 3.21,
14146
+ cacheRead: 0.18,
13899
14147
  cacheWrite: 0,
13900
14148
  },
13901
14149
  contextWindow: 262144,
@@ -13910,9 +14158,9 @@ export const MODELS = {
13910
14158
  reasoning: true,
13911
14159
  input: ["text", "image"],
13912
14160
  cost: {
13913
- input: 2.30272875,
13914
- output: 11.550195,
13915
- cacheRead: 0.26316900000000004,
14161
+ input: 3,
14162
+ output: 15,
14163
+ cacheRead: 0.3,
13916
14164
  cacheWrite: 0,
13917
14165
  },
13918
14166
  contextWindow: 1048576,
@@ -13978,9 +14226,9 @@ export const MODELS = {
13978
14226
  reasoning: true,
13979
14227
  input: ["text"],
13980
14228
  cost: {
13981
- input: 0.049999999999999996,
13982
- output: 0.19999999999999998,
13983
- cacheRead: 0.03,
14229
+ input: 0.06,
14230
+ output: 0.24,
14231
+ cacheRead: 0,
13984
14232
  cacheWrite: 0,
13985
14233
  },
13986
14234
  contextWindow: 262144,
@@ -14012,13 +14260,13 @@ export const MODELS = {
14012
14260
  reasoning: true,
14013
14261
  input: ["text"],
14014
14262
  cost: {
14015
- input: 0.08499999999999999,
14016
- output: 0.39999999999999997,
14263
+ input: 0.08,
14264
+ output: 0.44999999999999996,
14017
14265
  cacheRead: 0,
14018
14266
  cacheWrite: 0,
14019
14267
  },
14020
14268
  contextWindow: 262144,
14021
- maxTokens: 16384,
14269
+ maxTokens: 235929,
14022
14270
  },
14023
14271
  "nvidia/nemotron-3-super-120b-a12b:free": {
14024
14272
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -14207,23 +14455,6 @@ export const MODELS = {
14207
14455
  contextWindow: 128000,
14208
14456
  maxTokens: 4096,
14209
14457
  },
14210
- "openai/gpt-4-turbo-preview": {
14211
- id: "openai/gpt-4-turbo-preview",
14212
- name: "OpenAI: GPT-4 Turbo Preview",
14213
- api: "openai-completions",
14214
- provider: "openrouter",
14215
- baseUrl: "https://openrouter.ai/api/v1",
14216
- reasoning: false,
14217
- input: ["text"],
14218
- cost: {
14219
- input: 10,
14220
- output: 30,
14221
- cacheRead: 0,
14222
- cacheWrite: 0,
14223
- },
14224
- contextWindow: 128000,
14225
- maxTokens: 4096,
14226
- },
14227
14458
  "openai/gpt-4-turbo:batch": {
14228
14459
  id: "openai/gpt-4-turbo:batch",
14229
14460
  name: "OpenAI: GPT-4 Turbo (batch)",
@@ -15815,13 +16046,13 @@ export const MODELS = {
15815
16046
  reasoning: true,
15816
16047
  input: ["text"],
15817
16048
  cost: {
15818
- input: 0.22749999999999998,
15819
- output: 0.9099999999999999,
16049
+ input: 0.12,
16050
+ output: 0.24,
15820
16051
  cacheRead: 0,
15821
16052
  cacheWrite: 0,
15822
16053
  },
15823
16054
  contextWindow: 131072,
15824
- maxTokens: 8192,
16055
+ maxTokens: 16384,
15825
16056
  },
15826
16057
  "qwen/qwen3-235b-a22b": {
15827
16058
  id: "qwen/qwen3-235b-a22b",
@@ -15900,13 +16131,13 @@ export const MODELS = {
15900
16131
  reasoning: false,
15901
16132
  input: ["text"],
15902
16133
  cost: {
15903
- input: 0.09,
15904
- output: 0.3,
16134
+ input: 0.04815,
16135
+ output: 0.19305,
15905
16136
  cacheRead: 0,
15906
16137
  cacheWrite: 0,
15907
16138
  },
15908
16139
  contextWindow: 262144,
15909
- maxTokens: 235929,
16140
+ maxTokens: 32000,
15910
16141
  },
15911
16142
  "qwen/qwen3-30b-a3b-thinking-2507": {
15912
16143
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -16110,7 +16341,7 @@ export const MODELS = {
16110
16341
  cacheWrite: 0,
16111
16342
  },
16112
16343
  contextWindow: 262144,
16113
- maxTokens: 235929,
16344
+ maxTokens: 32768,
16114
16345
  },
16115
16346
  "qwen/qwen3-vl-235b-a22b-instruct": {
16116
16347
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -16155,13 +16386,13 @@ export const MODELS = {
16155
16386
  reasoning: false,
16156
16387
  input: ["text", "image"],
16157
16388
  cost: {
16158
- input: 0.15,
16159
- output: 0.6,
16389
+ input: 0.13,
16390
+ output: 0.52,
16160
16391
  cacheRead: 0,
16161
16392
  cacheWrite: 0,
16162
16393
  },
16163
16394
  contextWindow: 262144,
16164
- maxTokens: 16384,
16395
+ maxTokens: 32768,
16165
16396
  },
16166
16397
  "qwen/qwen3-vl-30b-a3b-thinking": {
16167
16398
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -16274,13 +16505,13 @@ export const MODELS = {
16274
16505
  reasoning: true,
16275
16506
  input: ["text", "image"],
16276
16507
  cost: {
16277
- input: 0.3125,
16278
- output: 1.25,
16279
- cacheRead: 0.15625,
16508
+ input: 0.1625,
16509
+ output: 1.3,
16510
+ cacheRead: 0,
16280
16511
  cacheWrite: 0,
16281
16512
  },
16282
16513
  contextWindow: 262144,
16283
- maxTokens: 16384,
16514
+ maxTokens: 65536,
16284
16515
  },
16285
16516
  "qwen/qwen3.5-397b-a17b": {
16286
16517
  id: "qwen/qwen3.5-397b-a17b",
@@ -16724,6 +16955,23 @@ export const MODELS = {
16724
16955
  contextWindow: 131072,
16725
16956
  maxTokens: 16384,
16726
16957
  },
16958
+ "stealth/union-alpha": {
16959
+ id: "stealth/union-alpha",
16960
+ name: "Union Alpha",
16961
+ api: "openai-completions",
16962
+ provider: "openrouter",
16963
+ baseUrl: "https://openrouter.ai/api/v1",
16964
+ reasoning: false,
16965
+ input: ["text", "image"],
16966
+ cost: {
16967
+ input: 0,
16968
+ output: 0,
16969
+ cacheRead: 0,
16970
+ cacheWrite: 0,
16971
+ },
16972
+ contextWindow: 262144,
16973
+ maxTokens: 131072,
16974
+ },
16727
16975
  "stepfun/step-3.5-flash": {
16728
16976
  id: "stepfun/step-3.5-flash",
16729
16977
  name: "StepFun: Step 3.5 Flash",
@@ -16809,23 +17057,6 @@ export const MODELS = {
16809
17057
  contextWindow: 1048576,
16810
17058
  maxTokens: 64000,
16811
17059
  },
16812
- "thedrummer/unslopnemo-12b": {
16813
- id: "thedrummer/unslopnemo-12b",
16814
- name: "TheDrummer: UnslopNemo 12B",
16815
- api: "openai-completions",
16816
- provider: "openrouter",
16817
- baseUrl: "https://openrouter.ai/api/v1",
16818
- reasoning: false,
16819
- input: ["text"],
16820
- cost: {
16821
- input: 0.39999999999999997,
16822
- output: 0.39999999999999997,
16823
- cacheRead: 0,
16824
- cacheWrite: 0,
16825
- },
16826
- contextWindow: 1024000,
16827
- maxTokens: 819200,
16828
- },
16829
17060
  "thinkingmachines/inkling": {
16830
17061
  id: "thinkingmachines/inkling",
16831
17062
  name: "Thinking Machines: Inkling",
@@ -16860,23 +17091,6 @@ export const MODELS = {
16860
17091
  contextWindow: 1048576,
16861
17092
  maxTokens: 262144,
16862
17093
  },
16863
- "thinkingmachines/inkling-small:batch": {
16864
- id: "thinkingmachines/inkling-small:batch",
16865
- name: "Thinking Machines: Inkling Small (batch)",
16866
- api: "openai-completions",
16867
- provider: "openrouter",
16868
- baseUrl: "https://openrouter.ai/api/v1",
16869
- reasoning: true,
16870
- input: ["text", "image"],
16871
- cost: {
16872
- input: 0.5,
16873
- output: 1.2,
16874
- cacheRead: 0.09999999999999999,
16875
- cacheWrite: 0,
16876
- },
16877
- contextWindow: 524288,
16878
- maxTokens: 471859,
16879
- },
16880
17094
  "thinkingmachines/inkling-small:free": {
16881
17095
  id: "thinkingmachines/inkling-small:free",
16882
17096
  name: "Thinking Machines: Inkling Small (free)",
@@ -17277,13 +17491,13 @@ export const MODELS = {
17277
17491
  reasoning: true,
17278
17492
  input: ["text"],
17279
17493
  cost: {
17280
- input: 0.6,
17281
- output: 2,
17282
- cacheRead: 0.15,
17494
+ input: 1.4,
17495
+ output: 4.4,
17496
+ cacheRead: 0.14,
17283
17497
  cacheWrite: 0,
17284
17498
  },
17285
17499
  contextWindow: 1048576,
17286
- maxTokens: 182476,
17500
+ maxTokens: 131072,
17287
17501
  },
17288
17502
  "z-ai/glm-5.2:batch": {
17289
17503
  id: "z-ai/glm-5.2:batch",
@@ -17317,7 +17531,7 @@ export const MODELS = {
17317
17531
  cacheWrite: 0,
17318
17532
  },
17319
17533
  contextWindow: 1310720,
17320
- maxTokens: 943718,
17534
+ maxTokens: 943717,
17321
17535
  },
17322
17536
  "z-ai/glm-5.3-flash": {
17323
17537
  id: "z-ai/glm-5.3-flash",
@@ -17328,9 +17542,9 @@ export const MODELS = {
17328
17542
  reasoning: true,
17329
17543
  input: ["text", "image"],
17330
17544
  cost: {
17331
- input: 0.075,
17332
- output: 0.25,
17333
- cacheRead: 0.015,
17545
+ input: 0.09,
17546
+ output: 0.3,
17547
+ cacheRead: 0.018,
17334
17548
  cacheWrite: 0,
17335
17549
  },
17336
17550
  contextWindow: 1310720,
@@ -17455,6 +17669,40 @@ export const MODELS = {
17455
17669
  contextWindow: 1000000,
17456
17670
  maxTokens: 128000,
17457
17671
  },
17672
+ "~deepseek/deepseek-flash-latest": {
17673
+ id: "~deepseek/deepseek-flash-latest",
17674
+ name: "DeepSeek: DeepSeek Flash Latest",
17675
+ api: "openai-completions",
17676
+ provider: "openrouter",
17677
+ baseUrl: "https://openrouter.ai/api/v1",
17678
+ reasoning: true,
17679
+ input: ["text", "image"],
17680
+ cost: {
17681
+ input: 0.15,
17682
+ output: 0.6,
17683
+ cacheRead: 0.015,
17684
+ cacheWrite: 0,
17685
+ },
17686
+ contextWindow: 1048576,
17687
+ maxTokens: 393216,
17688
+ },
17689
+ "~deepseek/deepseek-pro-latest": {
17690
+ id: "~deepseek/deepseek-pro-latest",
17691
+ name: "DeepSeek: DeepSeek Pro Latest",
17692
+ api: "openai-completions",
17693
+ provider: "openrouter",
17694
+ baseUrl: "https://openrouter.ai/api/v1",
17695
+ reasoning: true,
17696
+ input: ["text"],
17697
+ cost: {
17698
+ input: 0.57948,
17699
+ output: 1.73844,
17700
+ cacheRead: 0.018438,
17701
+ cacheWrite: 0,
17702
+ },
17703
+ contextWindow: 1048576,
17704
+ maxTokens: 393216,
17705
+ },
17458
17706
  "~deepseek/deepseek-v4-flash-latest": {
17459
17707
  id: "~deepseek/deepseek-v4-flash-latest",
17460
17708
  name: "DeepSeek: DeepSeek V4 Flash Latest",
@@ -17467,8 +17715,8 @@ export const MODELS = {
17467
17715
  input: ["text"],
17468
17716
  cost: {
17469
17717
  input: 0.03,
17470
- output: 0.07,
17471
- cacheRead: 0.003,
17718
+ output: 0.13,
17719
+ cacheRead: 0.01,
17472
17720
  cacheWrite: 0,
17473
17721
  },
17474
17722
  contextWindow: 1310720,
@@ -17517,9 +17765,9 @@ export const MODELS = {
17517
17765
  reasoning: true,
17518
17766
  input: ["text", "image"],
17519
17767
  cost: {
17520
- input: 2.125,
17521
- output: 11.9,
17522
- cacheRead: 0.2465,
17768
+ input: 2.0999999999999996,
17769
+ output: 10.950000000000001,
17770
+ cacheRead: 0.22999999999999998,
17523
17771
  cacheWrite: 0,
17524
17772
  },
17525
17773
  contextWindow: 1048576,
@@ -17653,13 +17901,13 @@ export const MODELS = {
17653
17901
  reasoning: true,
17654
17902
  input: ["text"],
17655
17903
  cost: {
17656
- input: 0.8727,
17657
- output: 3.36,
17658
- cacheRead: 0.16390000000000002,
17904
+ input: 0.8775,
17905
+ output: 2.9699999999999998,
17906
+ cacheRead: 0.1755,
17659
17907
  cacheWrite: 0,
17660
17908
  },
17661
17909
  contextWindow: 1310720,
17662
- maxTokens: 943718,
17910
+ maxTokens: 235929,
17663
17911
  },
17664
17912
  },
17665
17913
  "together": {
@@ -17832,6 +18080,25 @@ export const MODELS = {
17832
18080
  contextWindow: 1048576,
17833
18081
  maxTokens: 384000,
17834
18082
  },
18083
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
18084
+ id: "deepseek-ai/DeepSeek-V4.1-Flash",
18085
+ name: "DeepSeek V4.1 Flash",
18086
+ api: "openai-completions",
18087
+ provider: "together",
18088
+ baseUrl: "https://api.together.ai/v1",
18089
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
18090
+ reasoning: true,
18091
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
18092
+ input: ["text", "image"],
18093
+ cost: {
18094
+ input: 0.3,
18095
+ output: 1.2,
18096
+ cacheRead: 0.006,
18097
+ cacheWrite: 0,
18098
+ },
18099
+ contextWindow: 1048576,
18100
+ maxTokens: 384000,
18101
+ },
17835
18102
  "google/gemma-4-31B-it": {
17836
18103
  id: "google/gemma-4-31B-it",
17837
18104
  name: "Gemma 4 31B Instruct",
@@ -18140,7 +18407,7 @@ export const MODELS = {
18140
18407
  cost: {
18141
18408
  input: 1.3,
18142
18409
  output: 7.8,
18143
- cacheRead: 0.26,
18410
+ cacheRead: 0.13,
18144
18411
  cacheWrite: 1.625,
18145
18412
  },
18146
18413
  contextWindow: 240000,
@@ -18296,8 +18563,8 @@ export const MODELS = {
18296
18563
  cacheRead: 0,
18297
18564
  cacheWrite: 0,
18298
18565
  },
18299
- contextWindow: 131072,
18300
- maxTokens: 32768,
18566
+ contextWindow: 262114,
18567
+ maxTokens: 262114,
18301
18568
  },
18302
18569
  "alibaba/qwen3-next-80b-a3b-thinking": {
18303
18570
  id: "alibaba/qwen3-next-80b-a3b-thinking",
@@ -18313,8 +18580,8 @@ export const MODELS = {
18313
18580
  cacheRead: 0,
18314
18581
  cacheWrite: 0,
18315
18582
  },
18316
- contextWindow: 131072,
18317
- maxTokens: 32768,
18583
+ contextWindow: 262144,
18584
+ maxTokens: 262144,
18318
18585
  },
18319
18586
  "alibaba/qwen3-vl-235b-a22b-instruct": {
18320
18587
  id: "alibaba/qwen3-vl-235b-a22b-instruct",
@@ -18378,7 +18645,7 @@ export const MODELS = {
18378
18645
  cost: {
18379
18646
  input: 0.09999999999999999,
18380
18647
  output: 0.39999999999999997,
18381
- cacheRead: 0.001,
18648
+ cacheRead: 0.01,
18382
18649
  cacheWrite: 0.125,
18383
18650
  },
18384
18651
  contextWindow: 1000000,
@@ -18394,7 +18661,7 @@ export const MODELS = {
18394
18661
  input: ["text", "image"],
18395
18662
  cost: {
18396
18663
  input: 0.39999999999999997,
18397
- output: 2.4,
18664
+ output: 2.5,
18398
18665
  cacheRead: 0.04,
18399
18666
  cacheWrite: 0.5,
18400
18667
  },
@@ -18429,7 +18696,7 @@ export const MODELS = {
18429
18696
  cost: {
18430
18697
  input: 0.5,
18431
18698
  output: 3,
18432
- cacheRead: 0.09999999999999999,
18699
+ cacheRead: 0.049999999999999996,
18433
18700
  cacheWrite: 0.625,
18434
18701
  },
18435
18702
  contextWindow: 1000000,
@@ -18529,7 +18796,7 @@ export const MODELS = {
18529
18796
  reasoning: true,
18530
18797
  input: ["text", "image"],
18531
18798
  cost: {
18532
- input: 0.16,
18799
+ input: 0.15,
18533
18800
  output: 0.47,
18534
18801
  cacheRead: 0.016,
18535
18802
  cacheWrite: 0.19999999999999998,
@@ -18549,9 +18816,9 @@ export const MODELS = {
18549
18816
  input: 2,
18550
18817
  output: 6,
18551
18818
  cacheRead: 0.25,
18552
- cacheWrite: 2.5,
18819
+ cacheWrite: 0,
18553
18820
  },
18554
- contextWindow: 1000000,
18821
+ contextWindow: 262144,
18555
18822
  maxTokens: 128000,
18556
18823
  },
18557
18824
  "alibaba/qwen3.8-max-0902": {
@@ -18969,6 +19236,23 @@ export const MODELS = {
18969
19236
  contextWindow: 256000,
18970
19237
  maxTokens: 64000,
18971
19238
  },
19239
+ "bytedance/seed-2.1-turbo": {
19240
+ id: "bytedance/seed-2.1-turbo",
19241
+ name: "Seed 2.1 Turbo",
19242
+ api: "anthropic-messages",
19243
+ provider: "vercel-ai-gateway",
19244
+ baseUrl: "https://ai-gateway.vercel.sh",
19245
+ reasoning: true,
19246
+ input: ["text", "image"],
19247
+ cost: {
19248
+ input: 0.5,
19249
+ output: 2.5,
19250
+ cacheRead: 0.09999999999999999,
19251
+ cacheWrite: 0,
19252
+ },
19253
+ contextWindow: 262144,
19254
+ maxTokens: 262144,
19255
+ },
18972
19256
  "cohere/command-a": {
18973
19257
  id: "cohere/command-a",
18974
19258
  name: "Command A",
@@ -19454,9 +19738,9 @@ export const MODELS = {
19454
19738
  reasoning: true,
19455
19739
  input: ["text"],
19456
19740
  cost: {
19457
- input: 0.06,
19458
- output: 0.18,
19459
- cacheRead: 0.012,
19741
+ input: 0.020999999999999998,
19742
+ output: 0.063,
19743
+ cacheRead: 0.004200000000000001,
19460
19744
  cacheWrite: 0,
19461
19745
  },
19462
19746
  contextWindow: 256000,
@@ -19581,74 +19865,6 @@ export const MODELS = {
19581
19865
  contextWindow: 1000000,
19582
19866
  maxTokens: 32000,
19583
19867
  },
19584
- "kwaipilot/kat-coder-air-v2.5": {
19585
- id: "kwaipilot/kat-coder-air-v2.5",
19586
- name: "Kat Coder Air V2.5",
19587
- api: "anthropic-messages",
19588
- provider: "vercel-ai-gateway",
19589
- baseUrl: "https://ai-gateway.vercel.sh",
19590
- reasoning: true,
19591
- input: ["text", "image"],
19592
- cost: {
19593
- input: 0.15,
19594
- output: 0.6,
19595
- cacheRead: 0.03,
19596
- cacheWrite: 0,
19597
- },
19598
- contextWindow: 256000,
19599
- maxTokens: 80000,
19600
- },
19601
- "kwaipilot/kat-coder-pro-v1": {
19602
- id: "kwaipilot/kat-coder-pro-v1",
19603
- name: "KAT-Coder-Pro V1",
19604
- api: "anthropic-messages",
19605
- provider: "vercel-ai-gateway",
19606
- baseUrl: "https://ai-gateway.vercel.sh",
19607
- reasoning: false,
19608
- input: ["text"],
19609
- cost: {
19610
- input: 0.3,
19611
- output: 1.2,
19612
- cacheRead: 0.06,
19613
- cacheWrite: 0,
19614
- },
19615
- contextWindow: 256000,
19616
- maxTokens: 32000,
19617
- },
19618
- "kwaipilot/kat-coder-pro-v2": {
19619
- id: "kwaipilot/kat-coder-pro-v2",
19620
- name: "Kat Coder Pro V2",
19621
- api: "anthropic-messages",
19622
- provider: "vercel-ai-gateway",
19623
- baseUrl: "https://ai-gateway.vercel.sh",
19624
- reasoning: true,
19625
- input: ["text"],
19626
- cost: {
19627
- input: 0.3,
19628
- output: 1.2,
19629
- cacheRead: 0.06,
19630
- cacheWrite: 0,
19631
- },
19632
- contextWindow: 256000,
19633
- maxTokens: 256000,
19634
- },
19635
- "kwaipilot/kat-coder-pro-v2.5": {
19636
- id: "kwaipilot/kat-coder-pro-v2.5",
19637
- name: "Kat Coder Pro V2.5",
19638
- api: "anthropic-messages",
19639
- provider: "vercel-ai-gateway",
19640
- baseUrl: "https://ai-gateway.vercel.sh",
19641
- reasoning: true,
19642
- input: ["text", "image"],
19643
- cost: {
19644
- input: 0.74,
19645
- output: 2.96,
19646
- cacheRead: 0.15,
19647
- cacheWrite: 0,
19648
- },
19649
- contextWindow: 256000,
19650
- maxTokens: 80000,
19651
- },
19652
19868
  "meta/llama-3.1-70b": {
19653
19869
  id: "meta/llama-3.1-70b",
19654
19870
  name: "Llama 3.1 70B Instruct",
@@ -19983,46 +20199,12 @@ export const MODELS = {
19983
20199
  cost: {
19984
20200
  input: 0.3,
19985
20201
  output: 0.8999999999999999,
19986
- cacheRead: 0,
20202
+ cacheRead: 0.03,
19987
20203
  cacheWrite: 0,
19988
20204
  },
19989
20205
  contextWindow: 128000,
19990
20206
  maxTokens: 4000,
19991
20207
  },
19992
- "mistral/devstral-2": {
19993
- id: "mistral/devstral-2",
19994
- name: "Devstral 2",
19995
- api: "anthropic-messages",
19996
- provider: "vercel-ai-gateway",
19997
- baseUrl: "https://ai-gateway.vercel.sh",
19998
- reasoning: false,
19999
- input: ["text"],
20000
- cost: {
20001
- input: 0.39999999999999997,
20002
- output: 2,
20003
- cacheRead: 0,
20004
- cacheWrite: 0,
20005
- },
20006
- contextWindow: 256000,
20007
- maxTokens: 256000,
20008
- },
20009
- "mistral/devstral-small-2": {
20010
- id: "mistral/devstral-small-2",
20011
- name: "Devstral Small 2",
20012
- api: "anthropic-messages",
20013
- provider: "vercel-ai-gateway",
20014
- baseUrl: "https://ai-gateway.vercel.sh",
20015
- reasoning: false,
20016
- input: ["text", "image"],
20017
- cost: {
20018
- input: 0.09999999999999999,
20019
- output: 0.3,
20020
- cacheRead: 0,
20021
- cacheWrite: 0,
20022
- },
20023
- contextWindow: 256000,
20024
- maxTokens: 256000,
20025
- },
20026
20208
  "mistral/ministral-14b": {
20027
20209
  id: "mistral/ministral-14b",
20028
20210
  name: "Ministral 14B",
@@ -20034,10 +20216,10 @@ export const MODELS = {
20034
20216
  cost: {
20035
20217
  input: 0.19999999999999998,
20036
20218
  output: 0.19999999999999998,
20037
- cacheRead: 0,
20219
+ cacheRead: 0.02,
20038
20220
  cacheWrite: 0,
20039
20221
  },
20040
- contextWindow: 256000,
20222
+ contextWindow: 262144,
20041
20223
  maxTokens: 256000,
20042
20224
  },
20043
20225
  "mistral/ministral-3b": {
@@ -20051,10 +20233,10 @@ export const MODELS = {
20051
20233
  cost: {
20052
20234
  input: 0.09999999999999999,
20053
20235
  output: 0.09999999999999999,
20054
- cacheRead: 0,
20236
+ cacheRead: 0.01,
20055
20237
  cacheWrite: 0,
20056
20238
  },
20057
- contextWindow: 128000,
20239
+ contextWindow: 131072,
20058
20240
  maxTokens: 4000,
20059
20241
  },
20060
20242
  "mistral/ministral-8b": {
@@ -20068,10 +20250,10 @@ export const MODELS = {
20068
20250
  cost: {
20069
20251
  input: 0.15,
20070
20252
  output: 0.15,
20071
- cacheRead: 0,
20253
+ cacheRead: 0.015,
20072
20254
  cacheWrite: 0,
20073
20255
  },
20074
- contextWindow: 128000,
20256
+ contextWindow: 262144,
20075
20257
  maxTokens: 4000,
20076
20258
  },
20077
20259
  "mistral/mistral-large-3": {
@@ -20085,29 +20267,12 @@ export const MODELS = {
20085
20267
  cost: {
20086
20268
  input: 0.5,
20087
20269
  output: 1.5,
20088
- cacheRead: 0,
20270
+ cacheRead: 0.049999999999999996,
20089
20271
  cacheWrite: 0,
20090
20272
  },
20091
- contextWindow: 256000,
20273
+ contextWindow: 262144,
20092
20274
  maxTokens: 256000,
20093
20275
  },
20094
- "mistral/mistral-medium": {
20095
- id: "mistral/mistral-medium",
20096
- name: "Mistral Medium 3.1",
20097
- api: "anthropic-messages",
20098
- provider: "vercel-ai-gateway",
20099
- baseUrl: "https://ai-gateway.vercel.sh",
20100
- reasoning: false,
20101
- input: ["text", "image"],
20102
- cost: {
20103
- input: 0.39999999999999997,
20104
- output: 2,
20105
- cacheRead: 0,
20106
- cacheWrite: 0,
20107
- },
20108
- contextWindow: 128000,
20109
- maxTokens: 64000,
20110
- },
20111
20276
  "mistral/mistral-medium-3.5": {
20112
20277
  id: "mistral/mistral-medium-3.5",
20113
20278
  name: "Mistral Medium Latest",
@@ -20119,10 +20284,10 @@ export const MODELS = {
20119
20284
  cost: {
20120
20285
  input: 1.5,
20121
20286
  output: 7.5,
20122
- cacheRead: 0,
20287
+ cacheRead: 0.15,
20123
20288
  cacheWrite: 0,
20124
20289
  },
20125
- contextWindow: 256000,
20290
+ contextWindow: 262144,
20126
20291
  maxTokens: 256000,
20127
20292
  },
20128
20293
  "mistral/mistral-nemo": {
@@ -20132,15 +20297,15 @@ export const MODELS = {
20132
20297
  provider: "vercel-ai-gateway",
20133
20298
  baseUrl: "https://ai-gateway.vercel.sh",
20134
20299
  reasoning: false,
20135
- input: ["text", "image"],
20300
+ input: ["text"],
20136
20301
  cost: {
20137
- input: 0.15,
20138
- output: 0.15,
20302
+ input: 0.04,
20303
+ output: 0.16999999999999998,
20139
20304
  cacheRead: 0,
20140
20305
  cacheWrite: 0,
20141
20306
  },
20142
- contextWindow: 128000,
20143
- maxTokens: 128000,
20307
+ contextWindow: 60288,
20308
+ maxTokens: 16000,
20144
20309
  },
20145
20310
  "mistral/mistral-small": {
20146
20311
  id: "mistral/mistral-small",
@@ -20150,30 +20315,13 @@ export const MODELS = {
20150
20315
  baseUrl: "https://ai-gateway.vercel.sh",
20151
20316
  reasoning: false,
20152
20317
  input: ["text", "image"],
20153
- cost: {
20154
- input: 0.09999999999999999,
20155
- output: 0.3,
20156
- cacheRead: 0,
20157
- cacheWrite: 0,
20158
- },
20159
- contextWindow: 32000,
20160
- maxTokens: 4000,
20161
- },
20162
- "mistral/pixtral-12b": {
20163
- id: "mistral/pixtral-12b",
20164
- name: "Pixtral 12B 2409",
20165
- api: "anthropic-messages",
20166
- provider: "vercel-ai-gateway",
20167
- baseUrl: "https://ai-gateway.vercel.sh",
20168
- reasoning: false,
20169
- input: ["text", "image"],
20170
20318
  cost: {
20171
20319
  input: 0.15,
20172
- output: 0.15,
20173
- cacheRead: 0,
20320
+ output: 0.6,
20321
+ cacheRead: 0.015,
20174
20322
  cacheWrite: 0,
20175
20323
  },
20176
- contextWindow: 128000,
20324
+ contextWindow: 262144,
20177
20325
  maxTokens: 4000,
20178
20326
  },
20179
20327
  "moonshotai/kimi-k2": {
@@ -20221,11 +20369,11 @@ export const MODELS = {
20221
20369
  cost: {
20222
20370
  input: 0.6,
20223
20371
  output: 3,
20224
- cacheRead: 0.09999999999999999,
20372
+ cacheRead: 0,
20225
20373
  cacheWrite: 0,
20226
20374
  },
20227
- contextWindow: 262114,
20228
- maxTokens: 262114,
20375
+ contextWindow: 256000,
20376
+ maxTokens: 256000,
20229
20377
  },
20230
20378
  "moonshotai/kimi-k2.6": {
20231
20379
  id: "moonshotai/kimi-k2.6",
@@ -20322,8 +20470,8 @@ export const MODELS = {
20322
20470
  input: ["text"],
20323
20471
  cost: {
20324
20472
  input: 0.049999999999999996,
20325
- output: 0.24,
20326
- cacheRead: 0,
20473
+ output: 0.19999999999999998,
20474
+ cacheRead: 0.024999999999999998,
20327
20475
  cacheWrite: 0,
20328
20476
  },
20329
20477
  contextWindow: 262144,
@@ -21262,8 +21410,8 @@ export const MODELS = {
21262
21410
  reasoning: true,
21263
21411
  input: ["text"],
21264
21412
  cost: {
21265
- input: 0.049999999999999996,
21266
- output: 0.19999999999999998,
21413
+ input: 0.03,
21414
+ output: 0.14,
21267
21415
  cacheRead: 0,
21268
21416
  cacheWrite: 0,
21269
21417
  },