@fleetagent/pi-ai 0.2.4 → 0.2.6

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -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,
@@ -4220,9 +4288,27 @@ export const MODELS = {
4220
4288
  contextWindow: 1050000,
4221
4289
  maxTokens: 128000,
4222
4290
  },
4223
- "o3": {
4224
- id: "o3",
4225
- name: "o3",
4291
+ "gpt-6-astra": {
4292
+ id: "gpt-6-astra",
4293
+ name: "GPT-6 Astra",
4294
+ api: "openai-responses",
4295
+ provider: "cloudflare-ai-gateway",
4296
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
4297
+ reasoning: true,
4298
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
4299
+ input: ["text", "image"],
4300
+ cost: {
4301
+ input: 10,
4302
+ output: 50,
4303
+ cacheRead: 1,
4304
+ cacheWrite: 12,
4305
+ },
4306
+ contextWindow: 1050000,
4307
+ maxTokens: 128000,
4308
+ },
4309
+ "o3": {
4310
+ id: "o3",
4311
+ name: "o3",
4226
4312
  api: "openai-responses",
4227
4313
  provider: "cloudflare-ai-gateway",
4228
4314
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
@@ -4695,6 +4781,24 @@ export const MODELS = {
4695
4781
  contextWindow: 1000000,
4696
4782
  maxTokens: 384000,
4697
4783
  },
4784
+ "accounts/fireworks/models/deepseek-v4-pro": {
4785
+ id: "accounts/fireworks/models/deepseek-v4-pro",
4786
+ name: "DeepSeek V4 Pro",
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: 1.2,
4795
+ output: 1.2,
4796
+ cacheRead: 0.6,
4797
+ cacheWrite: 0,
4798
+ },
4799
+ contextWindow: 1000000,
4800
+ maxTokens: 384000,
4801
+ },
4698
4802
  "accounts/fireworks/models/deepseek-v4-pro-0813": {
4699
4803
  id: "accounts/fireworks/models/deepseek-v4-pro-0813",
4700
4804
  name: "DeepSeek V4 Pro 0813",
@@ -4875,41 +4979,41 @@ export const MODELS = {
4875
4979
  contextWindow: 1048576,
4876
4980
  maxTokens: 131072,
4877
4981
  },
4878
- "accounts/fireworks/models/minimax-m3": {
4879
- id: "accounts/fireworks/models/minimax-m3",
4880
- name: "MiniMax-M3",
4982
+ "accounts/fireworks/models/minimax-m2p7": {
4983
+ id: "accounts/fireworks/models/minimax-m2p7",
4984
+ name: "MiniMax-M2.7",
4881
4985
  api: "anthropic-messages",
4882
4986
  provider: "fireworks",
4883
4987
  baseUrl: "https://api.fireworks.ai/inference",
4884
4988
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4885
4989
  reasoning: true,
4886
- input: ["text", "image"],
4990
+ input: ["text"],
4887
4991
  cost: {
4888
- input: 0.3,
4992
+ input: 1.2,
4889
4993
  output: 1.2,
4890
- cacheRead: 0.06,
4994
+ cacheRead: 0.6,
4891
4995
  cacheWrite: 0,
4892
4996
  },
4893
- contextWindow: 512000,
4894
- maxTokens: 512000,
4997
+ contextWindow: 196608,
4998
+ maxTokens: 131072,
4895
4999
  },
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",
5000
+ "accounts/fireworks/models/minimax-m3": {
5001
+ id: "accounts/fireworks/models/minimax-m3",
5002
+ name: "MiniMax-M3",
4899
5003
  api: "anthropic-messages",
4900
5004
  provider: "fireworks",
4901
5005
  baseUrl: "https://api.fireworks.ai/inference",
4902
5006
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4903
- reasoning: false,
4904
- input: ["text", "image"],
5007
+ reasoning: true,
5008
+ input: ["text"],
4905
5009
  cost: {
4906
- input: 0,
4907
- output: 0,
4908
- cacheRead: 0,
5010
+ input: 0.3,
5011
+ output: 1.2,
5012
+ cacheRead: 0.06,
4909
5013
  cacheWrite: 0,
4910
5014
  },
4911
- contextWindow: 262144,
4912
- maxTokens: 262144,
5015
+ contextWindow: 512000,
5016
+ maxTokens: 512000,
4913
5017
  },
4914
5018
  "accounts/fireworks/models/muse-glimmer-30b": {
4915
5019
  id: "accounts/fireworks/models/muse-glimmer-30b",
@@ -4941,7 +5045,7 @@ export const MODELS = {
4941
5045
  cost: {
4942
5046
  input: 0.6,
4943
5047
  output: 2.4,
4944
- cacheRead: 0.119,
5048
+ cacheRead: 0.12,
4945
5049
  cacheWrite: 0,
4946
5050
  },
4947
5051
  contextWindow: 262144,
@@ -5009,7 +5113,7 @@ export const MODELS = {
5009
5113
  baseUrl: "https://api.fireworks.ai/inference",
5010
5114
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5011
5115
  reasoning: true,
5012
- input: ["text"],
5116
+ input: ["text", "image"],
5013
5117
  cost: {
5014
5118
  input: 2,
5015
5119
  output: 6,
@@ -5019,6 +5123,42 @@ export const MODELS = {
5019
5123
  contextWindow: 262144,
5020
5124
  maxTokens: 131072,
5021
5125
  },
5126
+ "accounts/fireworks/routers/deepseek-flash-latest": {
5127
+ id: "accounts/fireworks/routers/deepseek-flash-latest",
5128
+ name: "DeepSeek Flash 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", "image"],
5135
+ cost: {
5136
+ input: 0.22,
5137
+ output: 0.66,
5138
+ cacheRead: 0.007,
5139
+ cacheWrite: 0,
5140
+ },
5141
+ contextWindow: 1000000,
5142
+ maxTokens: 384000,
5143
+ },
5144
+ "accounts/fireworks/routers/deepseek-pro-latest": {
5145
+ id: "accounts/fireworks/routers/deepseek-pro-latest",
5146
+ name: "DeepSeek Pro Latest",
5147
+ api: "anthropic-messages",
5148
+ provider: "fireworks",
5149
+ baseUrl: "https://api.fireworks.ai/inference",
5150
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5151
+ reasoning: true,
5152
+ input: ["text"],
5153
+ cost: {
5154
+ input: 1.32,
5155
+ output: 3.96,
5156
+ cacheRead: 0.044,
5157
+ cacheWrite: 0,
5158
+ },
5159
+ contextWindow: 1000000,
5160
+ maxTokens: 384000,
5161
+ },
5022
5162
  "accounts/fireworks/routers/glm-5p2-fast": {
5023
5163
  id: "accounts/fireworks/routers/glm-5p2-fast",
5024
5164
  name: "GLM 5.2 Fast",
@@ -5055,6 +5195,78 @@ export const MODELS = {
5055
5195
  contextWindow: 1048572,
5056
5196
  maxTokens: 262144,
5057
5197
  },
5198
+ "accounts/fireworks/routers/glm-fast-latest": {
5199
+ id: "accounts/fireworks/routers/glm-fast-latest",
5200
+ name: "GLM 5.3 Fast (Latest)",
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"],
5207
+ cost: {
5208
+ input: 2.1,
5209
+ output: 6.6,
5210
+ cacheRead: 0.39,
5211
+ cacheWrite: 0,
5212
+ },
5213
+ contextWindow: 1048572,
5214
+ maxTokens: 262144,
5215
+ },
5216
+ "accounts/fireworks/routers/glm-flash-latest": {
5217
+ id: "accounts/fireworks/routers/glm-flash-latest",
5218
+ name: "GLM Flash Latest (GLM 5.3 Flash)",
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", "image"],
5225
+ cost: {
5226
+ input: 0.15,
5227
+ output: 0.5,
5228
+ cacheRead: 0.03,
5229
+ cacheWrite: 0,
5230
+ },
5231
+ contextWindow: 1048573,
5232
+ maxTokens: 131072,
5233
+ },
5234
+ "accounts/fireworks/routers/glm-latest": {
5235
+ id: "accounts/fireworks/routers/glm-latest",
5236
+ name: "GLM 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"],
5243
+ cost: {
5244
+ input: 1.4,
5245
+ output: 4.4,
5246
+ cacheRead: 0.26,
5247
+ cacheWrite: 0,
5248
+ },
5249
+ contextWindow: 1048573,
5250
+ maxTokens: 262144,
5251
+ },
5252
+ "accounts/fireworks/routers/kimi-fast-latest": {
5253
+ id: "accounts/fireworks/routers/kimi-fast-latest",
5254
+ name: "Kimi Fast Latest",
5255
+ api: "anthropic-messages",
5256
+ provider: "fireworks",
5257
+ baseUrl: "https://api.fireworks.ai/inference",
5258
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5259
+ reasoning: true,
5260
+ input: ["text", "image"],
5261
+ cost: {
5262
+ input: 4.5,
5263
+ output: 22.5,
5264
+ cacheRead: 0.45,
5265
+ cacheWrite: 0,
5266
+ },
5267
+ contextWindow: 1048576,
5268
+ maxTokens: 131072,
5269
+ },
5058
5270
  "accounts/fireworks/routers/kimi-k2p6-turbo": {
5059
5271
  id: "accounts/fireworks/routers/kimi-k2p6-turbo",
5060
5272
  name: "Kimi K2.6 Turbo",
@@ -5091,6 +5303,60 @@ export const MODELS = {
5091
5303
  contextWindow: 1048576,
5092
5304
  maxTokens: 131072,
5093
5305
  },
5306
+ "accounts/fireworks/routers/kimi-latest": {
5307
+ id: "accounts/fireworks/routers/kimi-latest",
5308
+ name: "Kimi 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", "image"],
5315
+ cost: {
5316
+ input: 3,
5317
+ output: 15,
5318
+ cacheRead: 0.3,
5319
+ cacheWrite: 0,
5320
+ },
5321
+ contextWindow: 1048576,
5322
+ maxTokens: 131072,
5323
+ },
5324
+ "accounts/fireworks/routers/minimax-latest": {
5325
+ id: "accounts/fireworks/routers/minimax-latest",
5326
+ name: "MiniMax Latest",
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"],
5333
+ cost: {
5334
+ input: 0.3,
5335
+ output: 1.2,
5336
+ cacheRead: 0.06,
5337
+ cacheWrite: 0,
5338
+ },
5339
+ contextWindow: 512000,
5340
+ maxTokens: 512000,
5341
+ },
5342
+ "accounts/fireworks/routers/qwen-max-latest": {
5343
+ id: "accounts/fireworks/routers/qwen-max-latest",
5344
+ name: "Qwen Max Latest (Qwen3.8 Max)",
5345
+ api: "anthropic-messages",
5346
+ provider: "fireworks",
5347
+ baseUrl: "https://api.fireworks.ai/inference",
5348
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
5349
+ reasoning: true,
5350
+ input: ["text", "image"],
5351
+ cost: {
5352
+ input: 2,
5353
+ output: 6,
5354
+ cacheRead: 0.25,
5355
+ cacheWrite: 0,
5356
+ },
5357
+ contextWindow: 262144,
5358
+ maxTokens: 131072,
5359
+ },
5094
5360
  },
5095
5361
  "github-copilot": {
5096
5362
  "claude-fable-5": {
@@ -7757,7 +8023,7 @@ export const MODELS = {
7757
8023
  },
7758
8024
  "kimi-for-coding": {
7759
8025
  id: "kimi-for-coding",
7760
- name: "Kimi K2.7 Code",
8026
+ name: "kimi-for-coding",
7761
8027
  api: "anthropic-messages",
7762
8028
  provider: "kimi-coding",
7763
8029
  baseUrl: "https://api.kimi.com/coding",
@@ -7770,7 +8036,7 @@ export const MODELS = {
7770
8036
  cacheRead: 0,
7771
8037
  cacheWrite: 0,
7772
8038
  },
7773
- contextWindow: 262144,
8039
+ contextWindow: 1048576,
7774
8040
  maxTokens: 32768,
7775
8041
  },
7776
8042
  "kimi-for-coding-highspeed": {
@@ -10698,6 +10964,23 @@ export const MODELS = {
10698
10964
  contextWindow: 262144,
10699
10965
  maxTokens: 65536,
10700
10966
  },
10967
+ "union-alpha": {
10968
+ id: "union-alpha",
10969
+ name: "Union Alpha Free",
10970
+ api: "anthropic-messages",
10971
+ provider: "opencode",
10972
+ baseUrl: "https://opencode.ai/zen",
10973
+ reasoning: true,
10974
+ input: ["text", "image"],
10975
+ cost: {
10976
+ input: 0,
10977
+ output: 0,
10978
+ cacheRead: 0,
10979
+ cacheWrite: 0,
10980
+ },
10981
+ contextWindow: 262144,
10982
+ maxTokens: 131072,
10983
+ },
10701
10984
  },
10702
10985
  "opencode-go": {
10703
10986
  "deepseek-v4-flash": {
@@ -11171,6 +11454,23 @@ export const MODELS = {
11171
11454
  contextWindow: 1000000,
11172
11455
  maxTokens: 131072,
11173
11456
  },
11457
+ "union-alpha": {
11458
+ id: "union-alpha",
11459
+ name: "Union Alpha Free",
11460
+ api: "anthropic-messages",
11461
+ provider: "opencode-go",
11462
+ baseUrl: "https://opencode.ai/zen/go",
11463
+ reasoning: true,
11464
+ input: ["text", "image"],
11465
+ cost: {
11466
+ input: 0,
11467
+ output: 0,
11468
+ cacheRead: 0,
11469
+ cacheWrite: 0,
11470
+ },
11471
+ contextWindow: 262144,
11472
+ maxTokens: 131072,
11473
+ },
11174
11474
  },
11175
11475
  "openrouter": {
11176
11476
  "aion-labs/aion-2.0": {
@@ -11621,7 +11921,7 @@ export const MODELS = {
11621
11921
  },
11622
11922
  "anthropic/claude-opus-5": {
11623
11923
  id: "anthropic/claude-opus-5",
11624
- name: "Claude Opus 5",
11924
+ name: "Anthropic: Claude Opus 5",
11625
11925
  api: "openai-completions",
11626
11926
  provider: "openrouter",
11627
11927
  baseUrl: "https://openrouter.ai/api/v1",
@@ -11638,7 +11938,7 @@ export const MODELS = {
11638
11938
  },
11639
11939
  "anthropic/claude-opus-5:batch": {
11640
11940
  id: "anthropic/claude-opus-5:batch",
11641
- name: "Claude Opus 5 (batch)",
11941
+ name: "Anthropic: Claude Opus 5 (batch)",
11642
11942
  api: "openai-completions",
11643
11943
  provider: "openrouter",
11644
11944
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12106,9 +12406,9 @@ export const MODELS = {
12106
12406
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12107
12407
  input: ["text"],
12108
12408
  cost: {
12109
- input: 0.06678,
12110
- output: 0.13356,
12111
- cacheRead: 0.013356,
12409
+ input: 0.07,
12410
+ output: 0.14,
12411
+ cacheRead: 0.014,
12112
12412
  cacheWrite: 0,
12113
12413
  },
12114
12414
  contextWindow: 1048576,
@@ -12125,9 +12425,9 @@ export const MODELS = {
12125
12425
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12126
12426
  input: ["text"],
12127
12427
  cost: {
12128
- input: 0.04,
12129
- output: 0.08,
12130
- cacheRead: 0.008,
12428
+ input: 0.06,
12429
+ output: 0.12,
12430
+ cacheRead: 0.012,
12131
12431
  cacheWrite: 0,
12132
12432
  },
12133
12433
  contextWindow: 1310720,
@@ -12201,13 +12501,13 @@ export const MODELS = {
12201
12501
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12202
12502
  input: ["text"],
12203
12503
  cost: {
12204
- input: 0.788916,
12205
- output: 1.577832,
12206
- cacheRead: 0.065743,
12504
+ input: 1.5999999999999999,
12505
+ output: 3.1999999999999997,
12506
+ cacheRead: 0.135,
12207
12507
  cacheWrite: 0,
12208
12508
  },
12209
12509
  contextWindow: 1048576,
12210
- maxTokens: 384000,
12510
+ maxTokens: 393216,
12211
12511
  },
12212
12512
  "deepseek/deepseek-v4-pro-0813": {
12213
12513
  id: "deepseek/deepseek-v4-pro-0813",
@@ -12220,9 +12520,9 @@ export const MODELS = {
12220
12520
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12221
12521
  input: ["text"],
12222
12522
  cost: {
12223
- input: 0.57816,
12224
- output: 1.73448,
12225
- cacheRead: 0.018396000000000003,
12523
+ input: 0.57948,
12524
+ output: 1.73844,
12525
+ cacheRead: 0.018438,
12226
12526
  cacheWrite: 0,
12227
12527
  },
12228
12528
  contextWindow: 1048576,
@@ -12385,39 +12685,22 @@ export const MODELS = {
12385
12685
  contextWindow: 1048576,
12386
12686
  maxTokens: 65536,
12387
12687
  },
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",
12688
+ "google/gemini-2.5-pro:batch": {
12689
+ id: "google/gemini-2.5-pro:batch",
12690
+ name: "Google: Gemini 2.5 Pro (batch)",
12391
12691
  api: "openai-completions",
12392
12692
  provider: "openrouter",
12393
12693
  baseUrl: "https://openrouter.ai/api/v1",
12394
12694
  reasoning: true,
12395
12695
  input: ["text", "image"],
12396
12696
  cost: {
12397
- input: 1.25,
12398
- output: 10,
12697
+ input: 0.625,
12698
+ output: 5,
12399
12699
  cacheRead: 0.125,
12400
- cacheWrite: 0.375,
12700
+ cacheWrite: 0,
12401
12701
  },
12402
12702
  contextWindow: 1048576,
12403
- maxTokens: 65535,
12404
- },
12405
- "google/gemini-2.5-pro:batch": {
12406
- id: "google/gemini-2.5-pro:batch",
12407
- name: "Google: Gemini 2.5 Pro (batch)",
12408
- api: "openai-completions",
12409
- provider: "openrouter",
12410
- baseUrl: "https://openrouter.ai/api/v1",
12411
- reasoning: true,
12412
- input: ["text", "image"],
12413
- cost: {
12414
- input: 0.625,
12415
- output: 5,
12416
- cacheRead: 0.125,
12417
- cacheWrite: 0,
12418
- },
12419
- contextWindow: 1048576,
12420
- maxTokens: 65536,
12703
+ maxTokens: 65536,
12421
12704
  },
12422
12705
  "google/gemini-3-flash-preview": {
12423
12706
  id: "google/gemini-3-flash-preview",
@@ -12785,13 +13068,13 @@ export const MODELS = {
12785
13068
  reasoning: true,
12786
13069
  input: ["text", "image"],
12787
13070
  cost: {
12788
- input: 0.041999999999999996,
12789
- output: 0.22,
12790
- cacheRead: 0,
13071
+ input: 0.09,
13072
+ output: 0.3,
13073
+ cacheRead: 0.049999999999999996,
12791
13074
  cacheWrite: 0,
12792
13075
  },
12793
13076
  contextWindow: 262144,
12794
- maxTokens: 32768,
13077
+ maxTokens: 235929,
12795
13078
  },
12796
13079
  "google/gemma-4-26b-a4b-it:free": {
12797
13080
  id: "google/gemma-4-26b-a4b-it:free",
@@ -12827,23 +13110,6 @@ export const MODELS = {
12827
13110
  contextWindow: 262144,
12828
13111
  maxTokens: 16384,
12829
13112
  },
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
13113
  "google/gemma-4-31b-it:free": {
12848
13114
  id: "google/gemma-4-31b-it:free",
12849
13115
  name: "Google: Gemma 4 31B (free)",
@@ -13093,13 +13359,13 @@ export const MODELS = {
13093
13359
  reasoning: false,
13094
13360
  input: ["text"],
13095
13361
  cost: {
13096
- input: 0.72,
13097
- output: 0.72,
13362
+ input: 0.39999999999999997,
13363
+ output: 0.39999999999999997,
13098
13364
  cacheRead: 0,
13099
13365
  cacheWrite: 0,
13100
13366
  },
13101
13367
  contextWindow: 131072,
13102
- maxTokens: 8192,
13368
+ maxTokens: 16384,
13103
13369
  },
13104
13370
  "meta-llama/llama-3.1-8b-instruct": {
13105
13371
  id: "meta-llama/llama-3.1-8b-instruct",
@@ -13144,13 +13410,13 @@ export const MODELS = {
13144
13410
  reasoning: false,
13145
13411
  input: ["text", "image"],
13146
13412
  cost: {
13147
- input: 0.19999999999999998,
13148
- output: 0.696,
13413
+ input: 0.1875,
13414
+ output: 0.6525,
13149
13415
  cacheRead: 0,
13150
13416
  cacheWrite: 0,
13151
13417
  },
13152
13418
  contextWindow: 1048576,
13153
- maxTokens: 115200,
13419
+ maxTokens: 16384,
13154
13420
  },
13155
13421
  "meta-llama/llama-4-scout": {
13156
13422
  id: "meta-llama/llama-4-scout",
@@ -13297,7 +13563,7 @@ export const MODELS = {
13297
13563
  reasoning: true,
13298
13564
  input: ["text"],
13299
13565
  cost: {
13300
- input: 0.55,
13566
+ input: 0.39999999999999997,
13301
13567
  output: 2.2,
13302
13568
  cacheRead: 0,
13303
13569
  cacheWrite: 0,
@@ -13756,8 +14022,8 @@ export const MODELS = {
13756
14022
  reasoning: false,
13757
14023
  input: ["text", "image"],
13758
14024
  cost: {
13759
- input: 0.075,
13760
- output: 0.19999999999999998,
14025
+ input: 0.09375,
14026
+ output: 0.25,
13761
14027
  cacheRead: 0,
13762
14028
  cacheWrite: 0,
13763
14029
  },
@@ -13813,7 +14079,7 @@ export const MODELS = {
13813
14079
  cacheWrite: 0,
13814
14080
  },
13815
14081
  contextWindow: 131072,
13816
- maxTokens: 100352,
14082
+ maxTokens: 98304,
13817
14083
  },
13818
14084
  "moonshotai/kimi-k2-0905": {
13819
14085
  id: "moonshotai/kimi-k2-0905",
@@ -13830,7 +14096,7 @@ export const MODELS = {
13830
14096
  cacheWrite: 0,
13831
14097
  },
13832
14098
  contextWindow: 262144,
13833
- maxTokens: 100352,
14099
+ maxTokens: 98304,
13834
14100
  },
13835
14101
  "moonshotai/kimi-k2-thinking": {
13836
14102
  id: "moonshotai/kimi-k2-thinking",
@@ -13843,11 +14109,11 @@ export const MODELS = {
13843
14109
  cost: {
13844
14110
  input: 0.6,
13845
14111
  output: 2.5,
13846
- cacheRead: 0,
14112
+ cacheRead: 0.15,
13847
14113
  cacheWrite: 0,
13848
14114
  },
13849
14115
  contextWindow: 262144,
13850
- maxTokens: 235929,
14116
+ maxTokens: 98304,
13851
14117
  },
13852
14118
  "moonshotai/kimi-k2.5": {
13853
14119
  id: "moonshotai/kimi-k2.5",
@@ -13893,9 +14159,9 @@ export const MODELS = {
13893
14159
  reasoning: true,
13894
14160
  input: ["text", "image"],
13895
14161
  cost: {
13896
- input: 0.71,
13897
- output: 3.5,
13898
- cacheRead: 0.15,
14162
+ input: 0.7062,
14163
+ output: 3.21,
14164
+ cacheRead: 0.18,
13899
14165
  cacheWrite: 0,
13900
14166
  },
13901
14167
  contextWindow: 262144,
@@ -13910,9 +14176,9 @@ export const MODELS = {
13910
14176
  reasoning: true,
13911
14177
  input: ["text", "image"],
13912
14178
  cost: {
13913
- input: 2.30272875,
13914
- output: 11.550195,
13915
- cacheRead: 0.26316900000000004,
14179
+ input: 3,
14180
+ output: 15,
14181
+ cacheRead: 0.3,
13916
14182
  cacheWrite: 0,
13917
14183
  },
13918
14184
  contextWindow: 1048576,
@@ -13978,9 +14244,9 @@ export const MODELS = {
13978
14244
  reasoning: true,
13979
14245
  input: ["text"],
13980
14246
  cost: {
13981
- input: 0.049999999999999996,
13982
- output: 0.19999999999999998,
13983
- cacheRead: 0.03,
14247
+ input: 0.06,
14248
+ output: 0.24,
14249
+ cacheRead: 0,
13984
14250
  cacheWrite: 0,
13985
14251
  },
13986
14252
  contextWindow: 262144,
@@ -14012,13 +14278,13 @@ export const MODELS = {
14012
14278
  reasoning: true,
14013
14279
  input: ["text"],
14014
14280
  cost: {
14015
- input: 0.08499999999999999,
14016
- output: 0.39999999999999997,
14281
+ input: 0.08,
14282
+ output: 0.44999999999999996,
14017
14283
  cacheRead: 0,
14018
14284
  cacheWrite: 0,
14019
14285
  },
14020
14286
  contextWindow: 262144,
14021
- maxTokens: 16384,
14287
+ maxTokens: 235929,
14022
14288
  },
14023
14289
  "nvidia/nemotron-3-super-120b-a12b:free": {
14024
14290
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -14207,23 +14473,6 @@ export const MODELS = {
14207
14473
  contextWindow: 128000,
14208
14474
  maxTokens: 4096,
14209
14475
  },
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
14476
  "openai/gpt-4-turbo:batch": {
14228
14477
  id: "openai/gpt-4-turbo:batch",
14229
14478
  name: "OpenAI: GPT-4 Turbo (batch)",
@@ -15815,13 +16064,13 @@ export const MODELS = {
15815
16064
  reasoning: true,
15816
16065
  input: ["text"],
15817
16066
  cost: {
15818
- input: 0.22749999999999998,
15819
- output: 0.9099999999999999,
16067
+ input: 0.12,
16068
+ output: 0.24,
15820
16069
  cacheRead: 0,
15821
16070
  cacheWrite: 0,
15822
16071
  },
15823
16072
  contextWindow: 131072,
15824
- maxTokens: 8192,
16073
+ maxTokens: 16384,
15825
16074
  },
15826
16075
  "qwen/qwen3-235b-a22b": {
15827
16076
  id: "qwen/qwen3-235b-a22b",
@@ -15900,13 +16149,13 @@ export const MODELS = {
15900
16149
  reasoning: false,
15901
16150
  input: ["text"],
15902
16151
  cost: {
15903
- input: 0.09,
15904
- output: 0.3,
16152
+ input: 0.04815,
16153
+ output: 0.19305,
15905
16154
  cacheRead: 0,
15906
16155
  cacheWrite: 0,
15907
16156
  },
15908
16157
  contextWindow: 262144,
15909
- maxTokens: 235929,
16158
+ maxTokens: 32000,
15910
16159
  },
15911
16160
  "qwen/qwen3-30b-a3b-thinking-2507": {
15912
16161
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -16110,7 +16359,7 @@ export const MODELS = {
16110
16359
  cacheWrite: 0,
16111
16360
  },
16112
16361
  contextWindow: 262144,
16113
- maxTokens: 235929,
16362
+ maxTokens: 32768,
16114
16363
  },
16115
16364
  "qwen/qwen3-vl-235b-a22b-instruct": {
16116
16365
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -16155,13 +16404,13 @@ export const MODELS = {
16155
16404
  reasoning: false,
16156
16405
  input: ["text", "image"],
16157
16406
  cost: {
16158
- input: 0.15,
16159
- output: 0.6,
16407
+ input: 0.13,
16408
+ output: 0.52,
16160
16409
  cacheRead: 0,
16161
16410
  cacheWrite: 0,
16162
16411
  },
16163
16412
  contextWindow: 262144,
16164
- maxTokens: 16384,
16413
+ maxTokens: 32768,
16165
16414
  },
16166
16415
  "qwen/qwen3-vl-30b-a3b-thinking": {
16167
16416
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -16274,13 +16523,13 @@ export const MODELS = {
16274
16523
  reasoning: true,
16275
16524
  input: ["text", "image"],
16276
16525
  cost: {
16277
- input: 0.3125,
16278
- output: 1.25,
16279
- cacheRead: 0.15625,
16526
+ input: 0.1625,
16527
+ output: 1.3,
16528
+ cacheRead: 0,
16280
16529
  cacheWrite: 0,
16281
16530
  },
16282
16531
  contextWindow: 262144,
16283
- maxTokens: 16384,
16532
+ maxTokens: 65536,
16284
16533
  },
16285
16534
  "qwen/qwen3.5-397b-a17b": {
16286
16535
  id: "qwen/qwen3.5-397b-a17b",
@@ -16724,6 +16973,23 @@ export const MODELS = {
16724
16973
  contextWindow: 131072,
16725
16974
  maxTokens: 16384,
16726
16975
  },
16976
+ "stealth/union-alpha": {
16977
+ id: "stealth/union-alpha",
16978
+ name: "Union Alpha",
16979
+ api: "openai-completions",
16980
+ provider: "openrouter",
16981
+ baseUrl: "https://openrouter.ai/api/v1",
16982
+ reasoning: false,
16983
+ input: ["text", "image"],
16984
+ cost: {
16985
+ input: 0,
16986
+ output: 0,
16987
+ cacheRead: 0,
16988
+ cacheWrite: 0,
16989
+ },
16990
+ contextWindow: 262144,
16991
+ maxTokens: 131072,
16992
+ },
16727
16993
  "stepfun/step-3.5-flash": {
16728
16994
  id: "stepfun/step-3.5-flash",
16729
16995
  name: "StepFun: Step 3.5 Flash",
@@ -16767,9 +17033,9 @@ export const MODELS = {
16767
17033
  reasoning: true,
16768
17034
  input: ["text"],
16769
17035
  cost: {
16770
- input: 0.13199999999999998,
16771
- output: 0.5279999999999999,
16772
- cacheRead: 0.032999999999999995,
17036
+ input: 0.0825,
17037
+ output: 0.33,
17038
+ cacheRead: 0.020625,
16773
17039
  cacheWrite: 0,
16774
17040
  },
16775
17041
  contextWindow: 262144,
@@ -16809,23 +17075,6 @@ export const MODELS = {
16809
17075
  contextWindow: 1048576,
16810
17076
  maxTokens: 64000,
16811
17077
  },
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
17078
  "thinkingmachines/inkling": {
16830
17079
  id: "thinkingmachines/inkling",
16831
17080
  name: "Thinking Machines: Inkling",
@@ -16860,23 +17109,6 @@ export const MODELS = {
16860
17109
  contextWindow: 1048576,
16861
17110
  maxTokens: 262144,
16862
17111
  },
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
17112
  "thinkingmachines/inkling-small:free": {
16881
17113
  id: "thinkingmachines/inkling-small:free",
16882
17114
  name: "Thinking Machines: Inkling Small (free)",
@@ -17277,13 +17509,13 @@ export const MODELS = {
17277
17509
  reasoning: true,
17278
17510
  input: ["text"],
17279
17511
  cost: {
17280
- input: 0.6,
17281
- output: 2,
17282
- cacheRead: 0.15,
17512
+ input: 1.4,
17513
+ output: 4.4,
17514
+ cacheRead: 0.14,
17283
17515
  cacheWrite: 0,
17284
17516
  },
17285
17517
  contextWindow: 1048576,
17286
- maxTokens: 182476,
17518
+ maxTokens: 131072,
17287
17519
  },
17288
17520
  "z-ai/glm-5.2:batch": {
17289
17521
  id: "z-ai/glm-5.2:batch",
@@ -17317,7 +17549,7 @@ export const MODELS = {
17317
17549
  cacheWrite: 0,
17318
17550
  },
17319
17551
  contextWindow: 1310720,
17320
- maxTokens: 943718,
17552
+ maxTokens: 943717,
17321
17553
  },
17322
17554
  "z-ai/glm-5.3-flash": {
17323
17555
  id: "z-ai/glm-5.3-flash",
@@ -17328,9 +17560,9 @@ export const MODELS = {
17328
17560
  reasoning: true,
17329
17561
  input: ["text", "image"],
17330
17562
  cost: {
17331
- input: 0.075,
17332
- output: 0.25,
17333
- cacheRead: 0.015,
17563
+ input: 0.09,
17564
+ output: 0.3,
17565
+ cacheRead: 0.018,
17334
17566
  cacheWrite: 0,
17335
17567
  },
17336
17568
  contextWindow: 1310720,
@@ -17455,6 +17687,40 @@ export const MODELS = {
17455
17687
  contextWindow: 1000000,
17456
17688
  maxTokens: 128000,
17457
17689
  },
17690
+ "~deepseek/deepseek-flash-latest": {
17691
+ id: "~deepseek/deepseek-flash-latest",
17692
+ name: "DeepSeek: DeepSeek Flash Latest",
17693
+ api: "openai-completions",
17694
+ provider: "openrouter",
17695
+ baseUrl: "https://openrouter.ai/api/v1",
17696
+ reasoning: true,
17697
+ input: ["text", "image"],
17698
+ cost: {
17699
+ input: 0.15,
17700
+ output: 0.6,
17701
+ cacheRead: 0.015,
17702
+ cacheWrite: 0,
17703
+ },
17704
+ contextWindow: 1048576,
17705
+ maxTokens: 393216,
17706
+ },
17707
+ "~deepseek/deepseek-pro-latest": {
17708
+ id: "~deepseek/deepseek-pro-latest",
17709
+ name: "DeepSeek: DeepSeek Pro Latest",
17710
+ api: "openai-completions",
17711
+ provider: "openrouter",
17712
+ baseUrl: "https://openrouter.ai/api/v1",
17713
+ reasoning: true,
17714
+ input: ["text"],
17715
+ cost: {
17716
+ input: 0.57948,
17717
+ output: 1.73844,
17718
+ cacheRead: 0.018438,
17719
+ cacheWrite: 0,
17720
+ },
17721
+ contextWindow: 1048576,
17722
+ maxTokens: 393216,
17723
+ },
17458
17724
  "~deepseek/deepseek-v4-flash-latest": {
17459
17725
  id: "~deepseek/deepseek-v4-flash-latest",
17460
17726
  name: "DeepSeek: DeepSeek V4 Flash Latest",
@@ -17467,8 +17733,8 @@ export const MODELS = {
17467
17733
  input: ["text"],
17468
17734
  cost: {
17469
17735
  input: 0.03,
17470
- output: 0.07,
17471
- cacheRead: 0.003,
17736
+ output: 0.13,
17737
+ cacheRead: 0.01,
17472
17738
  cacheWrite: 0,
17473
17739
  },
17474
17740
  contextWindow: 1310720,
@@ -17517,9 +17783,9 @@ export const MODELS = {
17517
17783
  reasoning: true,
17518
17784
  input: ["text", "image"],
17519
17785
  cost: {
17520
- input: 2.125,
17521
- output: 11.9,
17522
- cacheRead: 0.2465,
17786
+ input: 1.875,
17787
+ output: 10.5,
17788
+ cacheRead: 0.2175,
17523
17789
  cacheWrite: 0,
17524
17790
  },
17525
17791
  contextWindow: 1048576,
@@ -17653,13 +17919,13 @@ export const MODELS = {
17653
17919
  reasoning: true,
17654
17920
  input: ["text"],
17655
17921
  cost: {
17656
- input: 0.8727,
17657
- output: 3.36,
17658
- cacheRead: 0.16390000000000002,
17922
+ input: 0.8775,
17923
+ output: 2.9699999999999998,
17924
+ cacheRead: 0.1755,
17659
17925
  cacheWrite: 0,
17660
17926
  },
17661
17927
  contextWindow: 1310720,
17662
- maxTokens: 943718,
17928
+ maxTokens: 235929,
17663
17929
  },
17664
17930
  },
17665
17931
  "together": {
@@ -17832,6 +18098,25 @@ export const MODELS = {
17832
18098
  contextWindow: 1048576,
17833
18099
  maxTokens: 384000,
17834
18100
  },
18101
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
18102
+ id: "deepseek-ai/DeepSeek-V4.1-Flash",
18103
+ name: "DeepSeek V4.1 Flash",
18104
+ api: "openai-completions",
18105
+ provider: "together",
18106
+ baseUrl: "https://api.together.ai/v1",
18107
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
18108
+ reasoning: true,
18109
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
18110
+ input: ["text", "image"],
18111
+ cost: {
18112
+ input: 0.3,
18113
+ output: 1.2,
18114
+ cacheRead: 0.006,
18115
+ cacheWrite: 0,
18116
+ },
18117
+ contextWindow: 1048576,
18118
+ maxTokens: 384000,
18119
+ },
17835
18120
  "google/gemma-4-31B-it": {
17836
18121
  id: "google/gemma-4-31B-it",
17837
18122
  name: "Gemma 4 31B Instruct",
@@ -18140,7 +18425,7 @@ export const MODELS = {
18140
18425
  cost: {
18141
18426
  input: 1.3,
18142
18427
  output: 7.8,
18143
- cacheRead: 0.26,
18428
+ cacheRead: 0.13,
18144
18429
  cacheWrite: 1.625,
18145
18430
  },
18146
18431
  contextWindow: 240000,
@@ -18296,8 +18581,8 @@ export const MODELS = {
18296
18581
  cacheRead: 0,
18297
18582
  cacheWrite: 0,
18298
18583
  },
18299
- contextWindow: 131072,
18300
- maxTokens: 32768,
18584
+ contextWindow: 262114,
18585
+ maxTokens: 262114,
18301
18586
  },
18302
18587
  "alibaba/qwen3-next-80b-a3b-thinking": {
18303
18588
  id: "alibaba/qwen3-next-80b-a3b-thinking",
@@ -18313,8 +18598,8 @@ export const MODELS = {
18313
18598
  cacheRead: 0,
18314
18599
  cacheWrite: 0,
18315
18600
  },
18316
- contextWindow: 131072,
18317
- maxTokens: 32768,
18601
+ contextWindow: 262144,
18602
+ maxTokens: 262144,
18318
18603
  },
18319
18604
  "alibaba/qwen3-vl-235b-a22b-instruct": {
18320
18605
  id: "alibaba/qwen3-vl-235b-a22b-instruct",
@@ -18378,7 +18663,7 @@ export const MODELS = {
18378
18663
  cost: {
18379
18664
  input: 0.09999999999999999,
18380
18665
  output: 0.39999999999999997,
18381
- cacheRead: 0.001,
18666
+ cacheRead: 0.01,
18382
18667
  cacheWrite: 0.125,
18383
18668
  },
18384
18669
  contextWindow: 1000000,
@@ -18394,7 +18679,7 @@ export const MODELS = {
18394
18679
  input: ["text", "image"],
18395
18680
  cost: {
18396
18681
  input: 0.39999999999999997,
18397
- output: 2.4,
18682
+ output: 2.5,
18398
18683
  cacheRead: 0.04,
18399
18684
  cacheWrite: 0.5,
18400
18685
  },
@@ -18429,7 +18714,7 @@ export const MODELS = {
18429
18714
  cost: {
18430
18715
  input: 0.5,
18431
18716
  output: 3,
18432
- cacheRead: 0.09999999999999999,
18717
+ cacheRead: 0.049999999999999996,
18433
18718
  cacheWrite: 0.625,
18434
18719
  },
18435
18720
  contextWindow: 1000000,
@@ -18529,7 +18814,7 @@ export const MODELS = {
18529
18814
  reasoning: true,
18530
18815
  input: ["text", "image"],
18531
18816
  cost: {
18532
- input: 0.16,
18817
+ input: 0.15,
18533
18818
  output: 0.47,
18534
18819
  cacheRead: 0.016,
18535
18820
  cacheWrite: 0.19999999999999998,
@@ -18549,9 +18834,9 @@ export const MODELS = {
18549
18834
  input: 2,
18550
18835
  output: 6,
18551
18836
  cacheRead: 0.25,
18552
- cacheWrite: 2.5,
18837
+ cacheWrite: 0,
18553
18838
  },
18554
- contextWindow: 1000000,
18839
+ contextWindow: 262144,
18555
18840
  maxTokens: 128000,
18556
18841
  },
18557
18842
  "alibaba/qwen3.8-max-0902": {
@@ -18969,6 +19254,23 @@ export const MODELS = {
18969
19254
  contextWindow: 256000,
18970
19255
  maxTokens: 64000,
18971
19256
  },
19257
+ "bytedance/seed-2.1-turbo": {
19258
+ id: "bytedance/seed-2.1-turbo",
19259
+ name: "Seed 2.1 Turbo",
19260
+ api: "anthropic-messages",
19261
+ provider: "vercel-ai-gateway",
19262
+ baseUrl: "https://ai-gateway.vercel.sh",
19263
+ reasoning: true,
19264
+ input: ["text", "image"],
19265
+ cost: {
19266
+ input: 0.5,
19267
+ output: 2.5,
19268
+ cacheRead: 0.09999999999999999,
19269
+ cacheWrite: 0,
19270
+ },
19271
+ contextWindow: 262144,
19272
+ maxTokens: 262144,
19273
+ },
18972
19274
  "cohere/command-a": {
18973
19275
  id: "cohere/command-a",
18974
19276
  name: "Command A",
@@ -19454,9 +19756,9 @@ export const MODELS = {
19454
19756
  reasoning: true,
19455
19757
  input: ["text"],
19456
19758
  cost: {
19457
- input: 0.06,
19458
- output: 0.18,
19459
- cacheRead: 0.012,
19759
+ input: 0.020999999999999998,
19760
+ output: 0.063,
19761
+ cacheRead: 0.004200000000000001,
19460
19762
  cacheWrite: 0,
19461
19763
  },
19462
19764
  contextWindow: 256000,
@@ -19581,74 +19883,6 @@ export const MODELS = {
19581
19883
  contextWindow: 1000000,
19582
19884
  maxTokens: 32000,
19583
19885
  },
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
19886
  "meta/llama-3.1-70b": {
19653
19887
  id: "meta/llama-3.1-70b",
19654
19888
  name: "Llama 3.1 70B Instruct",
@@ -19983,46 +20217,12 @@ export const MODELS = {
19983
20217
  cost: {
19984
20218
  input: 0.3,
19985
20219
  output: 0.8999999999999999,
19986
- cacheRead: 0,
20220
+ cacheRead: 0.03,
19987
20221
  cacheWrite: 0,
19988
20222
  },
19989
20223
  contextWindow: 128000,
19990
20224
  maxTokens: 4000,
19991
20225
  },
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
20226
  "mistral/ministral-14b": {
20027
20227
  id: "mistral/ministral-14b",
20028
20228
  name: "Ministral 14B",
@@ -20034,10 +20234,10 @@ export const MODELS = {
20034
20234
  cost: {
20035
20235
  input: 0.19999999999999998,
20036
20236
  output: 0.19999999999999998,
20037
- cacheRead: 0,
20237
+ cacheRead: 0.02,
20038
20238
  cacheWrite: 0,
20039
20239
  },
20040
- contextWindow: 256000,
20240
+ contextWindow: 262144,
20041
20241
  maxTokens: 256000,
20042
20242
  },
20043
20243
  "mistral/ministral-3b": {
@@ -20051,10 +20251,10 @@ export const MODELS = {
20051
20251
  cost: {
20052
20252
  input: 0.09999999999999999,
20053
20253
  output: 0.09999999999999999,
20054
- cacheRead: 0,
20254
+ cacheRead: 0.01,
20055
20255
  cacheWrite: 0,
20056
20256
  },
20057
- contextWindow: 128000,
20257
+ contextWindow: 131072,
20058
20258
  maxTokens: 4000,
20059
20259
  },
20060
20260
  "mistral/ministral-8b": {
@@ -20068,10 +20268,10 @@ export const MODELS = {
20068
20268
  cost: {
20069
20269
  input: 0.15,
20070
20270
  output: 0.15,
20071
- cacheRead: 0,
20271
+ cacheRead: 0.015,
20072
20272
  cacheWrite: 0,
20073
20273
  },
20074
- contextWindow: 128000,
20274
+ contextWindow: 262144,
20075
20275
  maxTokens: 4000,
20076
20276
  },
20077
20277
  "mistral/mistral-large-3": {
@@ -20085,29 +20285,12 @@ export const MODELS = {
20085
20285
  cost: {
20086
20286
  input: 0.5,
20087
20287
  output: 1.5,
20088
- cacheRead: 0,
20288
+ cacheRead: 0.049999999999999996,
20089
20289
  cacheWrite: 0,
20090
20290
  },
20091
- contextWindow: 256000,
20291
+ contextWindow: 262144,
20092
20292
  maxTokens: 256000,
20093
20293
  },
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
20294
  "mistral/mistral-medium-3.5": {
20112
20295
  id: "mistral/mistral-medium-3.5",
20113
20296
  name: "Mistral Medium Latest",
@@ -20119,10 +20302,10 @@ export const MODELS = {
20119
20302
  cost: {
20120
20303
  input: 1.5,
20121
20304
  output: 7.5,
20122
- cacheRead: 0,
20305
+ cacheRead: 0.15,
20123
20306
  cacheWrite: 0,
20124
20307
  },
20125
- contextWindow: 256000,
20308
+ contextWindow: 262144,
20126
20309
  maxTokens: 256000,
20127
20310
  },
20128
20311
  "mistral/mistral-nemo": {
@@ -20132,15 +20315,15 @@ export const MODELS = {
20132
20315
  provider: "vercel-ai-gateway",
20133
20316
  baseUrl: "https://ai-gateway.vercel.sh",
20134
20317
  reasoning: false,
20135
- input: ["text", "image"],
20318
+ input: ["text"],
20136
20319
  cost: {
20137
- input: 0.15,
20138
- output: 0.15,
20320
+ input: 0.04,
20321
+ output: 0.16999999999999998,
20139
20322
  cacheRead: 0,
20140
20323
  cacheWrite: 0,
20141
20324
  },
20142
- contextWindow: 128000,
20143
- maxTokens: 128000,
20325
+ contextWindow: 60288,
20326
+ maxTokens: 16000,
20144
20327
  },
20145
20328
  "mistral/mistral-small": {
20146
20329
  id: "mistral/mistral-small",
@@ -20150,30 +20333,13 @@ export const MODELS = {
20150
20333
  baseUrl: "https://ai-gateway.vercel.sh",
20151
20334
  reasoning: false,
20152
20335
  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
20336
  cost: {
20171
20337
  input: 0.15,
20172
- output: 0.15,
20173
- cacheRead: 0,
20338
+ output: 0.6,
20339
+ cacheRead: 0.015,
20174
20340
  cacheWrite: 0,
20175
20341
  },
20176
- contextWindow: 128000,
20342
+ contextWindow: 262144,
20177
20343
  maxTokens: 4000,
20178
20344
  },
20179
20345
  "moonshotai/kimi-k2": {
@@ -20221,11 +20387,11 @@ export const MODELS = {
20221
20387
  cost: {
20222
20388
  input: 0.6,
20223
20389
  output: 3,
20224
- cacheRead: 0.09999999999999999,
20390
+ cacheRead: 0,
20225
20391
  cacheWrite: 0,
20226
20392
  },
20227
- contextWindow: 262114,
20228
- maxTokens: 262114,
20393
+ contextWindow: 256000,
20394
+ maxTokens: 256000,
20229
20395
  },
20230
20396
  "moonshotai/kimi-k2.6": {
20231
20397
  id: "moonshotai/kimi-k2.6",
@@ -20322,8 +20488,8 @@ export const MODELS = {
20322
20488
  input: ["text"],
20323
20489
  cost: {
20324
20490
  input: 0.049999999999999996,
20325
- output: 0.24,
20326
- cacheRead: 0,
20491
+ output: 0.19999999999999998,
20492
+ cacheRead: 0.024999999999999998,
20327
20493
  cacheWrite: 0,
20328
20494
  },
20329
20495
  contextWindow: 262144,
@@ -21262,8 +21428,8 @@ export const MODELS = {
21262
21428
  reasoning: true,
21263
21429
  input: ["text"],
21264
21430
  cost: {
21265
- input: 0.049999999999999996,
21266
- output: 0.19999999999999998,
21431
+ input: 0.03,
21432
+ output: 0.14,
21267
21433
  cacheRead: 0,
21268
21434
  cacheWrite: 0,
21269
21435
  },