@fleetagent/pi-ai 0.2.2 → 0.2.3

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.
@@ -276,6 +276,57 @@ export const MODELS = {
276
276
  contextWindow: 1000000,
277
277
  maxTokens: 128000,
278
278
  },
279
+ "apac.amazon.nova-lite-v1:0": {
280
+ id: "apac.amazon.nova-lite-v1:0",
281
+ name: "Nova Lite (APAC)",
282
+ api: "bedrock-converse-stream",
283
+ provider: "amazon-bedrock",
284
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
285
+ reasoning: false,
286
+ input: ["text", "image"],
287
+ cost: {
288
+ input: 0.063,
289
+ output: 0.252,
290
+ cacheRead: 0.01575,
291
+ cacheWrite: 0,
292
+ },
293
+ contextWindow: 300000,
294
+ maxTokens: 8192,
295
+ },
296
+ "apac.amazon.nova-micro-v1:0": {
297
+ id: "apac.amazon.nova-micro-v1:0",
298
+ name: "Nova Micro (APAC)",
299
+ api: "bedrock-converse-stream",
300
+ provider: "amazon-bedrock",
301
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
302
+ reasoning: false,
303
+ input: ["text"],
304
+ cost: {
305
+ input: 0.037,
306
+ output: 0.148,
307
+ cacheRead: 0.00925,
308
+ cacheWrite: 0,
309
+ },
310
+ contextWindow: 128000,
311
+ maxTokens: 8192,
312
+ },
313
+ "apac.amazon.nova-pro-v1:0": {
314
+ id: "apac.amazon.nova-pro-v1:0",
315
+ name: "Nova Pro (APAC)",
316
+ api: "bedrock-converse-stream",
317
+ provider: "amazon-bedrock",
318
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
319
+ reasoning: false,
320
+ input: ["text", "image"],
321
+ cost: {
322
+ input: 0.84,
323
+ output: 3.36,
324
+ cacheRead: 0.21,
325
+ cacheWrite: 0,
326
+ },
327
+ contextWindow: 300000,
328
+ maxTokens: 8192,
329
+ },
279
330
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
280
331
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
281
332
  name: "Claude Haiku 4.5 (AU)",
@@ -311,6 +362,24 @@ export const MODELS = {
311
362
  contextWindow: 1000000,
312
363
  maxTokens: 128000,
313
364
  },
365
+ "au.anthropic.claude-opus-4-7": {
366
+ id: "au.anthropic.claude-opus-4-7",
367
+ name: "Claude Opus 4.7 (AU)",
368
+ api: "bedrock-converse-stream",
369
+ provider: "amazon-bedrock",
370
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
371
+ reasoning: true,
372
+ thinkingLevelMap: { "xhigh": "xhigh" },
373
+ input: ["text", "image"],
374
+ cost: {
375
+ input: 5.5,
376
+ output: 27.5,
377
+ cacheRead: 0.55,
378
+ cacheWrite: 6.875,
379
+ },
380
+ contextWindow: 1000000,
381
+ maxTokens: 128000,
382
+ },
314
383
  "au.anthropic.claude-opus-4-8": {
315
384
  id: "au.anthropic.claude-opus-4-8",
316
385
  name: "Claude Opus 4.8 (AU)",
@@ -396,6 +465,23 @@ export const MODELS = {
396
465
  contextWindow: 1000000,
397
466
  maxTokens: 128000,
398
467
  },
468
+ "ca.amazon.nova-lite-v1:0": {
469
+ id: "ca.amazon.nova-lite-v1:0",
470
+ name: "Nova Lite (CA)",
471
+ api: "bedrock-converse-stream",
472
+ provider: "amazon-bedrock",
473
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
474
+ reasoning: false,
475
+ input: ["text", "image"],
476
+ cost: {
477
+ input: 0.064,
478
+ output: 0.256,
479
+ cacheRead: 0.016,
480
+ cacheWrite: 0,
481
+ },
482
+ contextWindow: 300000,
483
+ maxTokens: 8192,
484
+ },
399
485
  "deepseek.r1-v1:0": {
400
486
  id: "deepseek.r1-v1:0",
401
487
  name: "DeepSeek-R1",
@@ -447,6 +533,74 @@ export const MODELS = {
447
533
  contextWindow: 163840,
448
534
  maxTokens: 81920,
449
535
  },
536
+ "eu.amazon.nova-2-lite-v1:0": {
537
+ id: "eu.amazon.nova-2-lite-v1:0",
538
+ name: "Nova 2 Lite (EU)",
539
+ api: "bedrock-converse-stream",
540
+ provider: "amazon-bedrock",
541
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
542
+ reasoning: true,
543
+ input: ["text", "image"],
544
+ cost: {
545
+ input: 0.374,
546
+ output: 3.157,
547
+ cacheRead: 0.0935,
548
+ cacheWrite: 0,
549
+ },
550
+ contextWindow: 1000000,
551
+ maxTokens: 64000,
552
+ },
553
+ "eu.amazon.nova-lite-v1:0": {
554
+ id: "eu.amazon.nova-lite-v1:0",
555
+ name: "Nova Lite (EU)",
556
+ api: "bedrock-converse-stream",
557
+ provider: "amazon-bedrock",
558
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
559
+ reasoning: false,
560
+ input: ["text", "image"],
561
+ cost: {
562
+ input: 0.069,
563
+ output: 0.276,
564
+ cacheRead: 0.01725,
565
+ cacheWrite: 0,
566
+ },
567
+ contextWindow: 300000,
568
+ maxTokens: 8192,
569
+ },
570
+ "eu.amazon.nova-micro-v1:0": {
571
+ id: "eu.amazon.nova-micro-v1:0",
572
+ name: "Nova Micro (EU)",
573
+ api: "bedrock-converse-stream",
574
+ provider: "amazon-bedrock",
575
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
576
+ reasoning: false,
577
+ input: ["text"],
578
+ cost: {
579
+ input: 0.04,
580
+ output: 0.16,
581
+ cacheRead: 0.01,
582
+ cacheWrite: 0,
583
+ },
584
+ contextWindow: 128000,
585
+ maxTokens: 8192,
586
+ },
587
+ "eu.amazon.nova-pro-v1:0": {
588
+ id: "eu.amazon.nova-pro-v1:0",
589
+ name: "Nova Pro (EU)",
590
+ api: "bedrock-converse-stream",
591
+ provider: "amazon-bedrock",
592
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
593
+ reasoning: false,
594
+ input: ["text", "image"],
595
+ cost: {
596
+ input: 0.92,
597
+ output: 3.68,
598
+ cacheRead: 0.23,
599
+ cacheWrite: 0,
600
+ },
601
+ contextWindow: 300000,
602
+ maxTokens: 8192,
603
+ },
450
604
  "eu.anthropic.claude-fable-5": {
451
605
  id: "eu.anthropic.claude-fable-5",
452
606
  name: "Claude Fable 5 (EU)",
@@ -619,6 +773,40 @@ export const MODELS = {
619
773
  contextWindow: 1000000,
620
774
  maxTokens: 128000,
621
775
  },
776
+ "eu.mistral.pixtral-large-2502-v1:0": {
777
+ id: "eu.mistral.pixtral-large-2502-v1:0",
778
+ name: "Pixtral Large (25.02) (EU)",
779
+ api: "bedrock-converse-stream",
780
+ provider: "amazon-bedrock",
781
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
782
+ reasoning: false,
783
+ input: ["text", "image"],
784
+ cost: {
785
+ input: 2,
786
+ output: 6,
787
+ cacheRead: 0,
788
+ cacheWrite: 0,
789
+ },
790
+ contextWindow: 128000,
791
+ maxTokens: 8192,
792
+ },
793
+ "global.amazon.nova-2-lite-v1:0": {
794
+ id: "global.amazon.nova-2-lite-v1:0",
795
+ name: "Nova 2 Lite (Global)",
796
+ api: "bedrock-converse-stream",
797
+ provider: "amazon-bedrock",
798
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
799
+ reasoning: true,
800
+ input: ["text", "image"],
801
+ cost: {
802
+ input: 0.3,
803
+ output: 2.5,
804
+ cacheRead: 0.075,
805
+ cacheWrite: 0,
806
+ },
807
+ contextWindow: 1000000,
808
+ maxTokens: 64000,
809
+ },
622
810
  "global.anthropic.claude-fable-5": {
623
811
  id: "global.anthropic.claude-fable-5",
624
812
  name: "Claude Fable 5 (Global)",
@@ -862,6 +1050,24 @@ export const MODELS = {
862
1050
  contextWindow: 1050000,
863
1051
  maxTokens: 128000,
864
1052
  },
1053
+ "global.openai.gpt-6-astra": {
1054
+ id: "global.openai.gpt-6-astra",
1055
+ name: "GPT-6 Astra (Global)",
1056
+ api: "bedrock-converse-stream",
1057
+ provider: "amazon-bedrock",
1058
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1059
+ reasoning: true,
1060
+ thinkingLevelMap: { "xhigh": "xhigh" },
1061
+ input: ["text", "image"],
1062
+ cost: {
1063
+ input: 10,
1064
+ output: 50,
1065
+ cacheRead: 1,
1066
+ cacheWrite: 12.5,
1067
+ },
1068
+ contextWindow: 1050000,
1069
+ maxTokens: 128000,
1070
+ },
865
1071
  "global.xai.grok-4.6": {
866
1072
  id: "global.xai.grok-4.6",
867
1073
  name: "Grok 4.6 (Global)",
@@ -913,6 +1119,23 @@ export const MODELS = {
913
1119
  contextWindow: 128000,
914
1120
  maxTokens: 4096,
915
1121
  },
1122
+ "jp.amazon.nova-2-lite-v1:0": {
1123
+ id: "jp.amazon.nova-2-lite-v1:0",
1124
+ name: "Nova 2 Lite (JP)",
1125
+ api: "bedrock-converse-stream",
1126
+ provider: "amazon-bedrock",
1127
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1128
+ reasoning: true,
1129
+ input: ["text", "image"],
1130
+ cost: {
1131
+ input: 0.396,
1132
+ output: 3.311,
1133
+ cacheRead: 0.099,
1134
+ cacheWrite: 0,
1135
+ },
1136
+ contextWindow: 1000000,
1137
+ maxTokens: 64000,
1138
+ },
916
1139
  "jp.anthropic.claude-haiku-4-5-20251001-v1:0": {
917
1140
  id: "jp.anthropic.claude-haiku-4-5-20251001-v1:0",
918
1141
  name: "Claude Haiku 4.5 (JP)",
@@ -1514,6 +1737,24 @@ export const MODELS = {
1514
1737
  contextWindow: 1050000,
1515
1738
  maxTokens: 128000,
1516
1739
  },
1740
+ "openai.gpt-6-astra": {
1741
+ id: "openai.gpt-6-astra",
1742
+ name: "GPT-6 Astra",
1743
+ api: "bedrock-converse-stream",
1744
+ provider: "amazon-bedrock",
1745
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1746
+ reasoning: true,
1747
+ thinkingLevelMap: { "xhigh": "xhigh" },
1748
+ input: ["text", "image"],
1749
+ cost: {
1750
+ input: 11,
1751
+ output: 55,
1752
+ cacheRead: 1.1,
1753
+ cacheWrite: 13.75,
1754
+ },
1755
+ contextWindow: 1050000,
1756
+ maxTokens: 128000,
1757
+ },
1517
1758
  "openai.gpt-oss-120b": {
1518
1759
  id: "openai.gpt-oss-120b",
1519
1760
  name: "gpt-oss-120b",
@@ -1735,6 +1976,74 @@ export const MODELS = {
1735
1976
  contextWindow: 262000,
1736
1977
  maxTokens: 262000,
1737
1978
  },
1979
+ "us.amazon.nova-2-lite-v1:0": {
1980
+ id: "us.amazon.nova-2-lite-v1:0",
1981
+ name: "Nova 2 Lite (US)",
1982
+ api: "bedrock-converse-stream",
1983
+ provider: "amazon-bedrock",
1984
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1985
+ reasoning: true,
1986
+ input: ["text", "image"],
1987
+ cost: {
1988
+ input: 0.33,
1989
+ output: 2.75,
1990
+ cacheRead: 0.0825,
1991
+ cacheWrite: 0,
1992
+ },
1993
+ contextWindow: 1000000,
1994
+ maxTokens: 64000,
1995
+ },
1996
+ "us.amazon.nova-lite-v1:0": {
1997
+ id: "us.amazon.nova-lite-v1:0",
1998
+ name: "Nova Lite (US)",
1999
+ api: "bedrock-converse-stream",
2000
+ provider: "amazon-bedrock",
2001
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2002
+ reasoning: false,
2003
+ input: ["text", "image"],
2004
+ cost: {
2005
+ input: 0.06,
2006
+ output: 0.24,
2007
+ cacheRead: 0.015,
2008
+ cacheWrite: 0,
2009
+ },
2010
+ contextWindow: 300000,
2011
+ maxTokens: 8192,
2012
+ },
2013
+ "us.amazon.nova-micro-v1:0": {
2014
+ id: "us.amazon.nova-micro-v1:0",
2015
+ name: "Nova Micro (US)",
2016
+ api: "bedrock-converse-stream",
2017
+ provider: "amazon-bedrock",
2018
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2019
+ reasoning: false,
2020
+ input: ["text"],
2021
+ cost: {
2022
+ input: 0.035,
2023
+ output: 0.14,
2024
+ cacheRead: 0.00875,
2025
+ cacheWrite: 0,
2026
+ },
2027
+ contextWindow: 128000,
2028
+ maxTokens: 8192,
2029
+ },
2030
+ "us.amazon.nova-pro-v1:0": {
2031
+ id: "us.amazon.nova-pro-v1:0",
2032
+ name: "Nova Pro (US)",
2033
+ api: "bedrock-converse-stream",
2034
+ provider: "amazon-bedrock",
2035
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2036
+ reasoning: false,
2037
+ input: ["text", "image"],
2038
+ cost: {
2039
+ input: 0.8,
2040
+ output: 3.2,
2041
+ cacheRead: 0.2,
2042
+ cacheWrite: 0,
2043
+ },
2044
+ contextWindow: 300000,
2045
+ maxTokens: 8192,
2046
+ },
1738
2047
  "us.anthropic.claude-fable-5": {
1739
2048
  id: "us.anthropic.claude-fable-5",
1740
2049
  name: "Claude Fable 5 (US)",
@@ -1958,6 +2267,57 @@ export const MODELS = {
1958
2267
  contextWindow: 128000,
1959
2268
  maxTokens: 32768,
1960
2269
  },
2270
+ "us.meta.llama3-1-70b-instruct-v1:0": {
2271
+ id: "us.meta.llama3-1-70b-instruct-v1:0",
2272
+ name: "Llama 3.1 70B Instruct (US)",
2273
+ api: "bedrock-converse-stream",
2274
+ provider: "amazon-bedrock",
2275
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2276
+ reasoning: false,
2277
+ input: ["text"],
2278
+ cost: {
2279
+ input: 0.72,
2280
+ output: 0.72,
2281
+ cacheRead: 0,
2282
+ cacheWrite: 0,
2283
+ },
2284
+ contextWindow: 128000,
2285
+ maxTokens: 4096,
2286
+ },
2287
+ "us.meta.llama3-1-8b-instruct-v1:0": {
2288
+ id: "us.meta.llama3-1-8b-instruct-v1:0",
2289
+ name: "Llama 3.1 8B Instruct (US)",
2290
+ api: "bedrock-converse-stream",
2291
+ provider: "amazon-bedrock",
2292
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2293
+ reasoning: false,
2294
+ input: ["text"],
2295
+ cost: {
2296
+ input: 0.22,
2297
+ output: 0.22,
2298
+ cacheRead: 0,
2299
+ cacheWrite: 0,
2300
+ },
2301
+ contextWindow: 128000,
2302
+ maxTokens: 4096,
2303
+ },
2304
+ "us.meta.llama3-3-70b-instruct-v1:0": {
2305
+ id: "us.meta.llama3-3-70b-instruct-v1:0",
2306
+ name: "Llama 3.3 70B Instruct (US)",
2307
+ api: "bedrock-converse-stream",
2308
+ provider: "amazon-bedrock",
2309
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2310
+ reasoning: false,
2311
+ input: ["text"],
2312
+ cost: {
2313
+ input: 0.72,
2314
+ output: 0.72,
2315
+ cacheRead: 0,
2316
+ cacheWrite: 0,
2317
+ },
2318
+ contextWindow: 128000,
2319
+ maxTokens: 4096,
2320
+ },
1961
2321
  "us.meta.llama4-maverick-17b-instruct-v1:0": {
1962
2322
  id: "us.meta.llama4-maverick-17b-instruct-v1:0",
1963
2323
  name: "Llama 4 Maverick 17B Instruct (US)",
@@ -1992,6 +2352,129 @@ export const MODELS = {
1992
2352
  contextWindow: 3500000,
1993
2353
  maxTokens: 16384,
1994
2354
  },
2355
+ "us.mistral.pixtral-large-2502-v1:0": {
2356
+ id: "us.mistral.pixtral-large-2502-v1:0",
2357
+ name: "Pixtral Large (25.02) (US)",
2358
+ api: "bedrock-converse-stream",
2359
+ provider: "amazon-bedrock",
2360
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2361
+ reasoning: false,
2362
+ input: ["text", "image"],
2363
+ cost: {
2364
+ input: 2,
2365
+ output: 6,
2366
+ cacheRead: 0,
2367
+ cacheWrite: 0,
2368
+ },
2369
+ contextWindow: 128000,
2370
+ maxTokens: 8192,
2371
+ },
2372
+ "us.openai.gpt-5.6-luna": {
2373
+ id: "us.openai.gpt-5.6-luna",
2374
+ name: "GPT-5.6 Luna (US)",
2375
+ api: "bedrock-converse-stream",
2376
+ provider: "amazon-bedrock",
2377
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2378
+ reasoning: true,
2379
+ thinkingLevelMap: { "xhigh": "xhigh" },
2380
+ input: ["text", "image"],
2381
+ cost: {
2382
+ input: 0.22,
2383
+ output: 1.32,
2384
+ cacheRead: 0.022,
2385
+ cacheWrite: 0.275,
2386
+ },
2387
+ contextWindow: 1050000,
2388
+ maxTokens: 128000,
2389
+ },
2390
+ "us.openai.gpt-5.6-sol": {
2391
+ id: "us.openai.gpt-5.6-sol",
2392
+ name: "GPT-5.6 Sol (US)",
2393
+ api: "bedrock-converse-stream",
2394
+ provider: "amazon-bedrock",
2395
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2396
+ reasoning: true,
2397
+ thinkingLevelMap: { "xhigh": "xhigh" },
2398
+ input: ["text", "image"],
2399
+ cost: {
2400
+ input: 4.4,
2401
+ output: 22,
2402
+ cacheRead: 0.44,
2403
+ cacheWrite: 5.5,
2404
+ },
2405
+ contextWindow: 1050000,
2406
+ maxTokens: 128000,
2407
+ },
2408
+ "us.openai.gpt-5.6-terra": {
2409
+ id: "us.openai.gpt-5.6-terra",
2410
+ name: "GPT-5.6 Terra (US)",
2411
+ api: "bedrock-converse-stream",
2412
+ provider: "amazon-bedrock",
2413
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2414
+ reasoning: true,
2415
+ thinkingLevelMap: { "xhigh": "xhigh" },
2416
+ input: ["text", "image"],
2417
+ cost: {
2418
+ input: 2.2,
2419
+ output: 13.2,
2420
+ cacheRead: 0.22,
2421
+ cacheWrite: 2.75,
2422
+ },
2423
+ contextWindow: 1050000,
2424
+ maxTokens: 128000,
2425
+ },
2426
+ "us.openai.gpt-6-astra": {
2427
+ id: "us.openai.gpt-6-astra",
2428
+ name: "GPT-6 Astra (US)",
2429
+ api: "bedrock-converse-stream",
2430
+ provider: "amazon-bedrock",
2431
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2432
+ reasoning: true,
2433
+ thinkingLevelMap: { "xhigh": "xhigh" },
2434
+ input: ["text", "image"],
2435
+ cost: {
2436
+ input: 11,
2437
+ output: 55,
2438
+ cacheRead: 1.1,
2439
+ cacheWrite: 13.75,
2440
+ },
2441
+ contextWindow: 1050000,
2442
+ maxTokens: 128000,
2443
+ },
2444
+ "us.writer.palmyra-x4-v1:0": {
2445
+ id: "us.writer.palmyra-x4-v1:0",
2446
+ name: "Palmyra X4 (US)",
2447
+ api: "bedrock-converse-stream",
2448
+ provider: "amazon-bedrock",
2449
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2450
+ reasoning: true,
2451
+ input: ["text"],
2452
+ cost: {
2453
+ input: 2.5,
2454
+ output: 10,
2455
+ cacheRead: 0,
2456
+ cacheWrite: 0,
2457
+ },
2458
+ contextWindow: 122880,
2459
+ maxTokens: 8192,
2460
+ },
2461
+ "us.writer.palmyra-x5-v1:0": {
2462
+ id: "us.writer.palmyra-x5-v1:0",
2463
+ name: "Palmyra X5 (US)",
2464
+ api: "bedrock-converse-stream",
2465
+ provider: "amazon-bedrock",
2466
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2467
+ reasoning: true,
2468
+ input: ["text"],
2469
+ cost: {
2470
+ input: 0.6,
2471
+ output: 6,
2472
+ cacheRead: 0,
2473
+ cacheWrite: 0,
2474
+ },
2475
+ contextWindow: 1040000,
2476
+ maxTokens: 8192,
2477
+ },
1995
2478
  "us.xai.grok-4.6": {
1996
2479
  id: "us.xai.grok-4.6",
1997
2480
  name: "Grok 4.6 (US)",
@@ -2153,7 +2636,9 @@ export const MODELS = {
2153
2636
  api: "anthropic-messages",
2154
2637
  provider: "anthropic",
2155
2638
  baseUrl: "https://api.anthropic.com",
2639
+ compat: { "supportsTemperature": false },
2156
2640
  reasoning: true,
2641
+ thinkingLevelMap: { "off": "low", "minimal": "low", "xhigh": "xhigh" },
2157
2642
  input: ["text", "image"],
2158
2643
  cost: {
2159
2644
  input: 10,
@@ -10037,6 +10522,23 @@ export const MODELS = {
10037
10522
  },
10038
10523
  },
10039
10524
  "opencode-go": {
10525
+ "deepseek-flash": {
10526
+ id: "deepseek-flash",
10527
+ name: "DeepSeek V4.1 Flash",
10528
+ api: "openai-completions",
10529
+ provider: "opencode-go",
10530
+ baseUrl: "https://opencode.ai/zen/go/v1",
10531
+ reasoning: true,
10532
+ input: ["text", "image"],
10533
+ cost: {
10534
+ input: 0.15,
10535
+ output: 0.6,
10536
+ cacheRead: 0.003,
10537
+ cacheWrite: 0,
10538
+ },
10539
+ contextWindow: 1000000,
10540
+ maxTokens: 384000,
10541
+ },
10040
10542
  "deepseek-v4-flash": {
10041
10543
  id: "deepseek-v4-flash",
10042
10544
  name: "DeepSeek V4 Flash",
@@ -10048,9 +10550,9 @@ export const MODELS = {
10048
10550
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10049
10551
  input: ["text"],
10050
10552
  cost: {
10051
- input: 0.22,
10052
- output: 0.66,
10053
- cacheRead: 0.007,
10553
+ input: 0.15,
10554
+ output: 0.6,
10555
+ cacheRead: 0.003,
10054
10556
  cacheWrite: 0,
10055
10557
  },
10056
10558
  contextWindow: 1000000,
@@ -10067,9 +10569,9 @@ export const MODELS = {
10067
10569
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10068
10570
  input: ["text", "image"],
10069
10571
  cost: {
10070
- input: 0.22,
10071
- output: 0.66,
10072
- cacheRead: 0.007,
10572
+ input: 0.15,
10573
+ output: 0.6,
10574
+ cacheRead: 0.003,
10073
10575
  cacheWrite: 0,
10074
10576
  },
10075
10577
  contextWindow: 1000000,
@@ -10147,16 +10649,16 @@ export const MODELS = {
10147
10649
  },
10148
10650
  "glm-5.3-flash": {
10149
10651
  id: "glm-5.3-flash",
10150
- name: "GLM-5.3-Flash (2x usage)",
10652
+ name: "GLM-5.3-Flash",
10151
10653
  api: "openai-completions",
10152
10654
  provider: "opencode-go",
10153
10655
  baseUrl: "https://opencode.ai/zen/go/v1",
10154
10656
  reasoning: true,
10155
10657
  input: ["text", "image"],
10156
10658
  cost: {
10157
- input: 0.075,
10158
- output: 0.25,
10159
- cacheRead: 0.015,
10659
+ input: 0.15,
10660
+ output: 0.5,
10661
+ cacheRead: 0.03,
10160
10662
  cacheWrite: 0,
10161
10663
  },
10162
10664
  contextWindow: 1000000,
@@ -10403,23 +10905,6 @@ export const MODELS = {
10403
10905
  contextWindow: 1048576,
10404
10906
  maxTokens: 131072,
10405
10907
  },
10406
- "omen-alpha": {
10407
- id: "omen-alpha",
10408
- name: "Omen Alpha",
10409
- api: "openai-completions",
10410
- provider: "opencode-go",
10411
- baseUrl: "https://opencode.ai/zen/go/v1",
10412
- reasoning: true,
10413
- input: ["text", "image"],
10414
- cost: {
10415
- input: 0.2,
10416
- output: 0.66,
10417
- cacheRead: 0.04,
10418
- cacheWrite: 0,
10419
- },
10420
- contextWindow: 500000,
10421
- maxTokens: 128000,
10422
- },
10423
10908
  "qwen3.6-plus": {
10424
10909
  id: "qwen3.6-plus",
10425
10910
  name: "Qwen3.6 Plus",
@@ -11303,13 +11788,13 @@ export const MODELS = {
11303
11788
  reasoning: false,
11304
11789
  input: ["text"],
11305
11790
  cost: {
11306
- input: 0.32,
11307
- output: 0.8899999999999999,
11791
+ input: 0.2574,
11792
+ output: 1.0287,
11308
11793
  cacheRead: 0,
11309
11794
  cacheWrite: 0,
11310
11795
  },
11311
11796
  contextWindow: 163840,
11312
- maxTokens: 16384,
11797
+ maxTokens: 16000,
11313
11798
  },
11314
11799
  "deepseek/deepseek-chat-v3-0324": {
11315
11800
  id: "deepseek/deepseek-chat-v3-0324",
@@ -11479,9 +11964,9 @@ export const MODELS = {
11479
11964
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11480
11965
  input: ["text"],
11481
11966
  cost: {
11482
- input: 0.14,
11483
- output: 0.28,
11484
- cacheRead: 0.03,
11967
+ input: 0.11,
11968
+ output: 0.33,
11969
+ cacheRead: 0.0035,
11485
11970
  cacheWrite: 0,
11486
11971
  },
11487
11972
  contextWindow: 1048576,
@@ -11498,13 +11983,32 @@ export const MODELS = {
11498
11983
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11499
11984
  input: ["text", "image"],
11500
11985
  cost: {
11501
- input: 0.44,
11502
- output: 1.32,
11503
- cacheRead: 0.014,
11986
+ input: 0.22,
11987
+ output: 0.66,
11988
+ cacheRead: 0.007,
11504
11989
  cacheWrite: 0,
11505
11990
  },
11506
11991
  contextWindow: 1048576,
11507
- maxTokens: 384000,
11992
+ maxTokens: 943718,
11993
+ },
11994
+ "deepseek/deepseek-v4-flash-vision-exp:batch": {
11995
+ id: "deepseek/deepseek-v4-flash-vision-exp:batch",
11996
+ name: "DeepSeek: DeepSeek V4 Flash Vision Exp (batch)",
11997
+ api: "openai-completions",
11998
+ provider: "openrouter",
11999
+ baseUrl: "https://openrouter.ai/api/v1",
12000
+ compat: { "requiresReasoningContentOnAssistantMessages": true },
12001
+ reasoning: true,
12002
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12003
+ input: ["text", "image"],
12004
+ cost: {
12005
+ input: 0.11,
12006
+ output: 0.33,
12007
+ cacheRead: 0.0035,
12008
+ cacheWrite: 0,
12009
+ },
12010
+ contextWindow: 1048576,
12011
+ maxTokens: 943718,
11508
12012
  },
11509
12013
  "deepseek/deepseek-v4-pro": {
11510
12014
  id: "deepseek/deepseek-v4-pro",
@@ -11555,9 +12059,9 @@ export const MODELS = {
11555
12059
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11556
12060
  input: ["text"],
11557
12061
  cost: {
11558
- input: 1.32,
11559
- output: 3.9600000000000004,
11560
- cacheRead: 0.13,
12062
+ input: 0.66,
12063
+ output: 1.9800000000000002,
12064
+ cacheRead: 0.022,
11561
12065
  cacheWrite: 0,
11562
12066
  },
11563
12067
  contextWindow: 1048576,
@@ -12167,9 +12671,9 @@ export const MODELS = {
12167
12671
  reasoning: true,
12168
12672
  input: ["text"],
12169
12673
  cost: {
12170
- input: 0.09999999999999999,
12171
- output: 0.15,
12172
- cacheRead: 0.049999999999999996,
12674
+ input: 0.06,
12675
+ output: 0.25,
12676
+ cacheRead: 0.015,
12173
12677
  cacheWrite: 0,
12174
12678
  },
12175
12679
  contextWindow: 131072,
@@ -12193,9 +12697,9 @@ export const MODELS = {
12193
12697
  contextWindow: 128000,
12194
12698
  maxTokens: 50000,
12195
12699
  },
12196
- "inception/mercury-2.5-preview": {
12197
- id: "inception/mercury-2.5-preview",
12198
- name: "Inception: Mercury 2.5 Preview",
12700
+ "inception/mercury-2.5": {
12701
+ id: "inception/mercury-2.5",
12702
+ name: "Inception: Mercury 2.5",
12199
12703
  api: "openai-completions",
12200
12704
  provider: "openrouter",
12201
12705
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12458,9 +12962,9 @@ export const MODELS = {
12458
12962
  reasoning: true,
12459
12963
  input: ["text", "image"],
12460
12964
  cost: {
12461
- input: 0.35,
12462
- output: 1.5,
12463
- cacheRead: 0.04,
12965
+ input: 0.175,
12966
+ output: 0.75,
12967
+ cacheRead: 0.02,
12464
12968
  cacheWrite: 0,
12465
12969
  },
12466
12970
  contextWindow: 131072,
@@ -12560,7 +13064,7 @@ export const MODELS = {
12560
13064
  reasoning: true,
12561
13065
  input: ["text"],
12562
13066
  cost: {
12563
- input: 0.39999999999999997,
13067
+ input: 0.55,
12564
13068
  output: 2.2,
12565
13069
  cacheRead: 0,
12566
13070
  cacheWrite: 0,
@@ -12611,13 +13115,13 @@ export const MODELS = {
12611
13115
  reasoning: true,
12612
13116
  input: ["text"],
12613
13117
  cost: {
12614
- input: 0.27,
12615
- output: 1.08,
12616
- cacheRead: 0.027,
13118
+ input: 0.3,
13119
+ output: 1.2,
13120
+ cacheRead: 0.03,
12617
13121
  cacheWrite: 0,
12618
13122
  },
12619
13123
  contextWindow: 204800,
12620
- maxTokens: 128000,
13124
+ maxTokens: 131072,
12621
13125
  },
12622
13126
  "minimax/minimax-m2.7": {
12623
13127
  id: "minimax/minimax-m2.7",
@@ -12687,6 +13191,23 @@ export const MODELS = {
12687
13191
  contextWindow: 256000,
12688
13192
  maxTokens: 204800,
12689
13193
  },
13194
+ "mistralai/codestral-2508:batch": {
13195
+ id: "mistralai/codestral-2508:batch",
13196
+ name: "Mistral: Codestral 2508 (batch)",
13197
+ api: "openai-completions",
13198
+ provider: "openrouter",
13199
+ baseUrl: "https://openrouter.ai/api/v1",
13200
+ reasoning: false,
13201
+ input: ["text"],
13202
+ cost: {
13203
+ input: 0.15,
13204
+ output: 0.44999999999999996,
13205
+ cacheRead: 0.015,
13206
+ cacheWrite: 0,
13207
+ },
13208
+ contextWindow: 256000,
13209
+ maxTokens: 204800,
13210
+ },
12690
13211
  "mistralai/devstral-2512": {
12691
13212
  id: "mistralai/devstral-2512",
12692
13213
  name: "Mistral: Devstral 2 2512",
@@ -12755,6 +13276,23 @@ export const MODELS = {
12755
13276
  contextWindow: 262144,
12756
13277
  maxTokens: 209715,
12757
13278
  },
13279
+ "mistralai/ministral-8b-2512:batch": {
13280
+ id: "mistralai/ministral-8b-2512:batch",
13281
+ name: "Mistral: Ministral 3 8B 2512 (batch)",
13282
+ api: "openai-completions",
13283
+ provider: "openrouter",
13284
+ baseUrl: "https://openrouter.ai/api/v1",
13285
+ reasoning: false,
13286
+ input: ["text", "image"],
13287
+ cost: {
13288
+ input: 0.075,
13289
+ output: 0.075,
13290
+ cacheRead: 0.0075,
13291
+ cacheWrite: 0,
13292
+ },
13293
+ contextWindow: 262144,
13294
+ maxTokens: 209715,
13295
+ },
12758
13296
  "mistralai/mistral-large": {
12759
13297
  id: "mistralai/mistral-large",
12760
13298
  name: "Mistral Large",
@@ -12806,6 +13344,23 @@ export const MODELS = {
12806
13344
  contextWindow: 262144,
12807
13345
  maxTokens: 209715,
12808
13346
  },
13347
+ "mistralai/mistral-large-2512:batch": {
13348
+ id: "mistralai/mistral-large-2512:batch",
13349
+ name: "Mistral: Mistral Large 3 2512 (batch)",
13350
+ api: "openai-completions",
13351
+ provider: "openrouter",
13352
+ baseUrl: "https://openrouter.ai/api/v1",
13353
+ reasoning: false,
13354
+ input: ["text", "image"],
13355
+ cost: {
13356
+ input: 0.25,
13357
+ output: 0.75,
13358
+ cacheRead: 0.024999999999999998,
13359
+ cacheWrite: 0,
13360
+ },
13361
+ contextWindow: 262144,
13362
+ maxTokens: 209715,
13363
+ },
12809
13364
  "mistralai/mistral-medium-3": {
12810
13365
  id: "mistralai/mistral-medium-3",
12811
13366
  name: "Mistral: Mistral Medium 3",
@@ -12874,6 +13429,23 @@ export const MODELS = {
12874
13429
  contextWindow: 131072,
12875
13430
  maxTokens: 104857,
12876
13431
  },
13432
+ "mistralai/mistral-medium-3.1:batch": {
13433
+ id: "mistralai/mistral-medium-3.1:batch",
13434
+ name: "Mistral: Mistral Medium 3.1 (batch)",
13435
+ api: "openai-completions",
13436
+ provider: "openrouter",
13437
+ baseUrl: "https://openrouter.ai/api/v1",
13438
+ reasoning: false,
13439
+ input: ["text", "image"],
13440
+ cost: {
13441
+ input: 0.19999999999999998,
13442
+ output: 1,
13443
+ cacheRead: 0.02,
13444
+ cacheWrite: 0,
13445
+ },
13446
+ contextWindow: 131072,
13447
+ maxTokens: 104857,
13448
+ },
12877
13449
  "mistralai/mistral-nemo": {
12878
13450
  id: "mistralai/mistral-nemo",
12879
13451
  name: "Mistral: Mistral Nemo",
@@ -12925,6 +13497,23 @@ export const MODELS = {
12925
13497
  contextWindow: 262144,
12926
13498
  maxTokens: 209715,
12927
13499
  },
13500
+ "mistralai/mistral-small-2603:batch": {
13501
+ id: "mistralai/mistral-small-2603:batch",
13502
+ name: "Mistral: Mistral Small 4 (batch)",
13503
+ api: "openai-completions",
13504
+ provider: "openrouter",
13505
+ baseUrl: "https://openrouter.ai/api/v1",
13506
+ reasoning: true,
13507
+ input: ["text", "image"],
13508
+ cost: {
13509
+ input: 0.075,
13510
+ output: 0.3,
13511
+ cacheRead: 0.0075,
13512
+ cacheWrite: 0,
13513
+ },
13514
+ contextWindow: 262144,
13515
+ maxTokens: 209715,
13516
+ },
12928
13517
  "mistralai/mistral-small-3.2-24b-instruct": {
12929
13518
  id: "mistralai/mistral-small-3.2-24b-instruct",
12930
13519
  name: "Mistral: Mistral Small 3.2 24B",
@@ -13021,11 +13610,11 @@ export const MODELS = {
13021
13610
  cost: {
13022
13611
  input: 0.6,
13023
13612
  output: 2.5,
13024
- cacheRead: 0,
13613
+ cacheRead: 0.15,
13025
13614
  cacheWrite: 0,
13026
13615
  },
13027
13616
  contextWindow: 262144,
13028
- maxTokens: 235929,
13617
+ maxTokens: 100352,
13029
13618
  },
13030
13619
  "moonshotai/kimi-k2.5": {
13031
13620
  id: "moonshotai/kimi-k2.5",
@@ -13113,35 +13702,35 @@ export const MODELS = {
13113
13702
  contextWindow: 1048576,
13114
13703
  maxTokens: 943718,
13115
13704
  },
13116
- "nex-agi/nex-n2-mini": {
13117
- id: "nex-agi/nex-n2-mini",
13118
- name: "Nex AGI: Nex-N2-Mini",
13705
+ "nex-agi/nex-n2.5-mini:free": {
13706
+ id: "nex-agi/nex-n2.5-mini:free",
13707
+ name: "Nex AGI: Nex-N2.5-Mini (free)",
13119
13708
  api: "openai-completions",
13120
13709
  provider: "openrouter",
13121
13710
  baseUrl: "https://openrouter.ai/api/v1",
13122
13711
  reasoning: true,
13123
13712
  input: ["text", "image"],
13124
13713
  cost: {
13125
- input: 0.024999999999999998,
13126
- output: 0.09999999999999999,
13127
- cacheRead: 0.0025,
13714
+ input: 0,
13715
+ output: 0,
13716
+ cacheRead: 0,
13128
13717
  cacheWrite: 0,
13129
13718
  },
13130
13719
  contextWindow: 262144,
13131
13720
  maxTokens: 235929,
13132
13721
  },
13133
- "nex-agi/nex-n2-pro": {
13134
- id: "nex-agi/nex-n2-pro",
13135
- name: "Nex AGI: Nex-N2-Pro",
13722
+ "nex-agi/nex-n2.5-pro:free": {
13723
+ id: "nex-agi/nex-n2.5-pro:free",
13724
+ name: "Nex AGI: Nex-N2.5-Pro (free)",
13136
13725
  api: "openai-completions",
13137
13726
  provider: "openrouter",
13138
13727
  baseUrl: "https://openrouter.ai/api/v1",
13139
13728
  reasoning: true,
13140
13729
  input: ["text", "image"],
13141
13730
  cost: {
13142
- input: 0.25,
13143
- output: 1,
13144
- cacheRead: 0.024999999999999998,
13731
+ input: 0,
13732
+ output: 0,
13733
+ cacheRead: 0,
13145
13734
  cacheWrite: 0,
13146
13735
  },
13147
13736
  contextWindow: 262144,
@@ -13195,7 +13784,7 @@ export const MODELS = {
13195
13784
  cacheRead: 0,
13196
13785
  cacheWrite: 0,
13197
13786
  },
13198
- contextWindow: 1000000,
13787
+ contextWindow: 262144,
13199
13788
  maxTokens: 16384,
13200
13789
  },
13201
13790
  "nvidia/nemotron-3-super-120b-a12b:free": {
@@ -15027,8 +15616,8 @@ export const MODELS = {
15027
15616
  reasoning: false,
15028
15617
  input: ["text"],
15029
15618
  cost: {
15030
- input: 0.09,
15031
- output: 0.55,
15619
+ input: 0.22,
15620
+ output: 0.88,
15032
15621
  cacheRead: 0,
15033
15622
  cacheWrite: 0,
15034
15623
  },
@@ -15078,13 +15667,13 @@ export const MODELS = {
15078
15667
  reasoning: false,
15079
15668
  input: ["text"],
15080
15669
  cost: {
15081
- input: 0.04815,
15082
- output: 0.19305,
15670
+ input: 0.09,
15671
+ output: 0.3,
15083
15672
  cacheRead: 0,
15084
15673
  cacheWrite: 0,
15085
15674
  },
15086
15675
  contextWindow: 262144,
15087
- maxTokens: 32000,
15676
+ maxTokens: 235929,
15088
15677
  },
15089
15678
  "qwen/qwen3-30b-a3b-thinking-2507": {
15090
15679
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -15265,13 +15854,13 @@ export const MODELS = {
15265
15854
  reasoning: false,
15266
15855
  input: ["text"],
15267
15856
  cost: {
15268
- input: 0.09999999999999999,
15857
+ input: 0.09,
15269
15858
  output: 1.1,
15270
- cacheRead: 0.07,
15859
+ cacheRead: 0,
15271
15860
  cacheWrite: 0,
15272
15861
  },
15273
15862
  contextWindow: 262144,
15274
- maxTokens: 235929,
15863
+ maxTokens: 16384,
15275
15864
  },
15276
15865
  "qwen/qwen3-next-80b-a3b-thinking": {
15277
15866
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -15418,13 +16007,13 @@ export const MODELS = {
15418
16007
  reasoning: true,
15419
16008
  input: ["text", "image"],
15420
16009
  cost: {
15421
- input: 0.29,
15422
- output: 2.4,
16010
+ input: 0.26,
16011
+ output: 2.08,
15423
16012
  cacheRead: 0,
15424
16013
  cacheWrite: 0,
15425
16014
  },
15426
16015
  contextWindow: 262144,
15427
- maxTokens: 81920,
16016
+ maxTokens: 65536,
15428
16017
  },
15429
16018
  "qwen/qwen3.5-27b": {
15430
16019
  id: "qwen/qwen3.5-27b",
@@ -15469,13 +16058,13 @@ export const MODELS = {
15469
16058
  reasoning: true,
15470
16059
  input: ["text", "image"],
15471
16060
  cost: {
15472
- input: 0.39,
15473
- output: 2.34,
15474
- cacheRead: 0,
16061
+ input: 0.55,
16062
+ output: 3.5,
16063
+ cacheRead: 0.22499999999999998,
15475
16064
  cacheWrite: 0,
15476
16065
  },
15477
16066
  contextWindow: 262144,
15478
- maxTokens: 65536,
16067
+ maxTokens: 235929,
15479
16068
  },
15480
16069
  "qwen/qwen3.5-9b": {
15481
16070
  id: "qwen/qwen3.5-9b",
@@ -15713,7 +16302,7 @@ export const MODELS = {
15713
16302
  cacheWrite: 0,
15714
16303
  },
15715
16304
  contextWindow: 1048576,
15716
- maxTokens: 262144,
16305
+ maxTokens: 131072,
15717
16306
  },
15718
16307
  "qwen/qwen3.8-2.4t-a95b:batch": {
15719
16308
  id: "qwen/qwen3.8-2.4t-a95b:batch",
@@ -15985,7 +16574,7 @@ export const MODELS = {
15985
16574
  cacheWrite: 0,
15986
16575
  },
15987
16576
  contextWindow: 1048576,
15988
- maxTokens: 471859,
16577
+ maxTokens: 32768,
15989
16578
  },
15990
16579
  "thinkingmachines/inkling-small": {
15991
16580
  id: "thinkingmachines/inkling-small",
@@ -16302,13 +16891,13 @@ export const MODELS = {
16302
16891
  reasoning: true,
16303
16892
  input: ["text"],
16304
16893
  cost: {
16305
- input: 0.55,
16306
- output: 2.2,
16307
- cacheRead: 0.11,
16894
+ input: 0.43,
16895
+ output: 1.75,
16896
+ cacheRead: 0.08,
16308
16897
  cacheWrite: 0,
16309
16898
  },
16310
16899
  contextWindow: 204800,
16311
- maxTokens: 131072,
16900
+ maxTokens: 16384,
16312
16901
  },
16313
16902
  "z-ai/glm-4.6v": {
16314
16903
  id: "z-ai/glm-4.6v",
@@ -16429,6 +17018,23 @@ export const MODELS = {
16429
17018
  contextWindow: 1048576,
16430
17019
  maxTokens: 131072,
16431
17020
  },
17021
+ "z-ai/glm-5.2:batch": {
17022
+ id: "z-ai/glm-5.2:batch",
17023
+ name: "Z.ai: GLM 5.2 (batch)",
17024
+ api: "openai-completions",
17025
+ provider: "openrouter",
17026
+ baseUrl: "https://openrouter.ai/api/v1",
17027
+ reasoning: true,
17028
+ input: ["text"],
17029
+ cost: {
17030
+ input: 0.7,
17031
+ output: 2.2,
17032
+ cacheRead: 0.07,
17033
+ cacheWrite: 0,
17034
+ },
17035
+ contextWindow: 1048576,
17036
+ maxTokens: 943718,
17037
+ },
16432
17038
  "z-ai/glm-5.3": {
16433
17039
  id: "z-ai/glm-5.3",
16434
17040
  name: "Z.ai: GLM 5.3",
@@ -16472,13 +17078,30 @@ export const MODELS = {
16472
17078
  reasoning: true,
16473
17079
  input: ["text", "image"],
16474
17080
  cost: {
16475
- input: 0.15,
16476
- output: 0.5,
16477
- cacheRead: 0.03,
17081
+ input: 0.075,
17082
+ output: 0.25,
17083
+ cacheRead: 0.015,
16478
17084
  cacheWrite: 0,
16479
17085
  },
16480
- contextWindow: 1048575,
16481
- maxTokens: 943717,
17086
+ contextWindow: 1048576,
17087
+ maxTokens: 943718,
17088
+ },
17089
+ "z-ai/glm-5.3:batch": {
17090
+ id: "z-ai/glm-5.3:batch",
17091
+ name: "Z.ai: GLM 5.3 (batch)",
17092
+ api: "openai-completions",
17093
+ provider: "openrouter",
17094
+ baseUrl: "https://openrouter.ai/api/v1",
17095
+ reasoning: true,
17096
+ input: ["text"],
17097
+ cost: {
17098
+ input: 0.7,
17099
+ output: 2.2,
17100
+ cacheRead: 0.13,
17101
+ cacheWrite: 0,
17102
+ },
17103
+ contextWindow: 1048576,
17104
+ maxTokens: 943718,
16482
17105
  },
16483
17106
  "z-ai/glm-5v-turbo": {
16484
17107
  id: "z-ai/glm-5v-turbo",
@@ -16627,9 +17250,9 @@ export const MODELS = {
16627
17250
  reasoning: true,
16628
17251
  input: ["text", "image"],
16629
17252
  cost: {
16630
- input: 2.5,
16631
- output: 14,
16632
- cacheRead: 0.29,
17253
+ input: 2.4,
17254
+ output: 12,
17255
+ cacheRead: 0.24,
16633
17256
  cacheWrite: 0,
16634
17257
  },
16635
17258
  contextWindow: 1048576,
@@ -16695,9 +17318,9 @@ export const MODELS = {
16695
17318
  reasoning: true,
16696
17319
  input: ["text", "image"],
16697
17320
  cost: {
16698
- input: 0.07125000000000001,
16699
- output: 0.2375,
16700
- cacheRead: 0.01425,
17321
+ input: 0.075,
17322
+ output: 0.25,
17323
+ cacheRead: 0.015,
16701
17324
  cacheWrite: 0,
16702
17325
  },
16703
17326
  contextWindow: 1310720,
@@ -16712,13 +17335,13 @@ export const MODELS = {
16712
17335
  reasoning: true,
16713
17336
  input: ["text"],
16714
17337
  cost: {
16715
- input: 1.12,
16716
- output: 3.52,
16717
- cacheRead: 0.20800000000000002,
17338
+ input: 1.092,
17339
+ output: 3.432,
17340
+ cacheRead: 0.20279999999999998,
16718
17341
  cacheWrite: 0,
16719
17342
  },
16720
17343
  contextWindow: 1310720,
16721
- maxTokens: 943718,
17344
+ maxTokens: 131072,
16722
17345
  },
16723
17346
  },
16724
17347
  "together": {
@@ -18122,9 +18745,9 @@ export const MODELS = {
18122
18745
  reasoning: false,
18123
18746
  input: ["text"],
18124
18747
  cost: {
18125
- input: 0.28,
18126
- output: 0.42,
18127
- cacheRead: 0.028,
18748
+ input: 0.62,
18749
+ output: 1.85,
18750
+ cacheRead: 0,
18128
18751
  cacheWrite: 0,
18129
18752
  },
18130
18753
  contextWindow: 128000,
@@ -18232,6 +18855,23 @@ export const MODELS = {
18232
18855
  contextWindow: 1000000,
18233
18856
  maxTokens: 384000,
18234
18857
  },
18858
+ "deepseek/deepseek-v4.1-flash": {
18859
+ id: "deepseek/deepseek-v4.1-flash",
18860
+ name: "DeepSeek V4.1 Flash",
18861
+ api: "anthropic-messages",
18862
+ provider: "vercel-ai-gateway",
18863
+ baseUrl: "https://ai-gateway.vercel.sh",
18864
+ reasoning: true,
18865
+ input: ["text", "image"],
18866
+ cost: {
18867
+ input: 0.15,
18868
+ output: 0.6,
18869
+ cacheRead: 0.003,
18870
+ cacheWrite: 0,
18871
+ },
18872
+ contextWindow: 1000000,
18873
+ maxTokens: 384000,
18874
+ },
18235
18875
  "google/gemini-2.5-flash": {
18236
18876
  id: "google/gemini-2.5-flash",
18237
18877
  name: "Gemini 2.5 Flash",
@@ -18470,6 +19110,23 @@ export const MODELS = {
18470
19110
  contextWindow: 128000,
18471
19111
  maxTokens: 128000,
18472
19112
  },
19113
+ "inception/mercury-2.5": {
19114
+ id: "inception/mercury-2.5",
19115
+ name: "Mercury 2.5",
19116
+ api: "anthropic-messages",
19117
+ provider: "vercel-ai-gateway",
19118
+ baseUrl: "https://ai-gateway.vercel.sh",
19119
+ reasoning: true,
19120
+ input: ["text"],
19121
+ cost: {
19122
+ input: 0.04,
19123
+ output: 0.15,
19124
+ cacheRead: 0.004,
19125
+ cacheWrite: 0,
19126
+ },
19127
+ contextWindow: 260000,
19128
+ maxTokens: 65536,
19129
+ },
18473
19130
  "inception/mercury-coder-small": {
18474
19131
  id: "inception/mercury-coder-small",
18475
19132
  name: "Mercury Coder Small Beta",
@@ -20839,23 +21496,6 @@ export const MODELS = {
20839
21496
  contextWindow: 1050000,
20840
21497
  maxTokens: 131000,
20841
21498
  },
20842
- "xiaomi/mimo-v2.5-pro-ultraspeed": {
20843
- id: "xiaomi/mimo-v2.5-pro-ultraspeed",
20844
- name: "MiMo V2.5 Pro UltraSpeed",
20845
- api: "anthropic-messages",
20846
- provider: "vercel-ai-gateway",
20847
- baseUrl: "https://ai-gateway.vercel.sh",
20848
- reasoning: true,
20849
- input: ["text"],
20850
- cost: {
20851
- input: 1.305,
20852
- output: 2.61,
20853
- cacheRead: 0.0108,
20854
- cacheWrite: 0,
20855
- },
20856
- contextWindow: 1048576,
20857
- maxTokens: 131072,
20858
- },
20859
21499
  "zai/glm-4.5": {
20860
21500
  id: "zai/glm-4.5",
20861
21501
  name: "GLM 4.5",
@@ -21069,9 +21709,9 @@ export const MODELS = {
21069
21709
  reasoning: true,
21070
21710
  input: ["text"],
21071
21711
  cost: {
21072
- input: 0.7,
21073
- output: 2.2,
21074
- cacheRead: 0.13,
21712
+ input: 1.4,
21713
+ output: 4.4,
21714
+ cacheRead: 0.14,
21075
21715
  cacheWrite: 0,
21076
21716
  },
21077
21717
  contextWindow: 1000000,
@@ -21111,23 +21751,6 @@ export const MODELS = {
21111
21751
  contextWindow: 1000000,
21112
21752
  maxTokens: 131000,
21113
21753
  },
21114
- "zai/glm-5.3-promo-50": {
21115
- id: "zai/glm-5.3-promo-50",
21116
- name: "GLM 5.3 (50% off)",
21117
- api: "anthropic-messages",
21118
- provider: "vercel-ai-gateway",
21119
- baseUrl: "https://ai-gateway.vercel.sh",
21120
- reasoning: true,
21121
- input: ["text"],
21122
- cost: {
21123
- input: 0.7,
21124
- output: 2.2,
21125
- cacheRead: 0.13,
21126
- cacheWrite: 0,
21127
- },
21128
- contextWindow: 1048576,
21129
- maxTokens: 1048576,
21130
- },
21131
21754
  "zai/glm-5v-turbo": {
21132
21755
  id: "zai/glm-5v-turbo",
21133
21756
  name: "GLM 5V Turbo",