@fleetagent/pi-ai 0.1.8 → 0.1.9

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.
@@ -191,6 +191,23 @@ export const MODELS = {
191
191
  contextWindow: 1000000,
192
192
  maxTokens: 128000,
193
193
  },
194
+ "anthropic.claude-opus-5": {
195
+ id: "anthropic.claude-opus-5",
196
+ name: "Claude Opus 5",
197
+ api: "bedrock-converse-stream",
198
+ provider: "amazon-bedrock",
199
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
200
+ reasoning: true,
201
+ input: ["text", "image"],
202
+ cost: {
203
+ input: 5,
204
+ output: 25,
205
+ cacheRead: 0.5,
206
+ cacheWrite: 6.25,
207
+ },
208
+ contextWindow: 1000000,
209
+ maxTokens: 128000,
210
+ },
194
211
  "anthropic.claude-sonnet-4-5-20250929-v1:0": {
195
212
  id: "anthropic.claude-sonnet-4-5-20250929-v1:0",
196
213
  name: "Claude Sonnet 4.5",
@@ -294,6 +311,23 @@ export const MODELS = {
294
311
  contextWindow: 1000000,
295
312
  maxTokens: 128000,
296
313
  },
314
+ "au.anthropic.claude-opus-5": {
315
+ id: "au.anthropic.claude-opus-5",
316
+ name: "Claude Opus 5 (AU)",
317
+ api: "bedrock-converse-stream",
318
+ provider: "amazon-bedrock",
319
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
320
+ reasoning: true,
321
+ input: ["text", "image"],
322
+ cost: {
323
+ input: 5,
324
+ output: 25,
325
+ cacheRead: 0.5,
326
+ cacheWrite: 6.25,
327
+ },
328
+ contextWindow: 1000000,
329
+ maxTokens: 128000,
330
+ },
297
331
  "au.anthropic.claude-sonnet-4-5-20250929-v1:0": {
298
332
  id: "au.anthropic.claude-sonnet-4-5-20250929-v1:0",
299
333
  name: "Claude Sonnet 4.5 (AU)",
@@ -500,6 +534,23 @@ export const MODELS = {
500
534
  contextWindow: 1000000,
501
535
  maxTokens: 128000,
502
536
  },
537
+ "eu.anthropic.claude-opus-5": {
538
+ id: "eu.anthropic.claude-opus-5",
539
+ name: "Claude Opus 5 (EU)",
540
+ api: "bedrock-converse-stream",
541
+ provider: "amazon-bedrock",
542
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
543
+ reasoning: true,
544
+ input: ["text", "image"],
545
+ cost: {
546
+ input: 5.5,
547
+ output: 27.5,
548
+ cacheRead: 0.55,
549
+ cacheWrite: 6.875,
550
+ },
551
+ contextWindow: 1000000,
552
+ maxTokens: 128000,
553
+ },
503
554
  "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": {
504
555
  id: "eu.anthropic.claude-sonnet-4-5-20250929-v1:0",
505
556
  name: "Claude Sonnet 4.5 (EU)",
@@ -655,6 +706,23 @@ export const MODELS = {
655
706
  contextWindow: 1000000,
656
707
  maxTokens: 128000,
657
708
  },
709
+ "global.anthropic.claude-opus-5": {
710
+ id: "global.anthropic.claude-opus-5",
711
+ name: "Claude Opus 5 (Global)",
712
+ api: "bedrock-converse-stream",
713
+ provider: "amazon-bedrock",
714
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
715
+ reasoning: true,
716
+ input: ["text", "image"],
717
+ cost: {
718
+ input: 5,
719
+ output: 25,
720
+ cacheRead: 0.5,
721
+ cacheWrite: 6.25,
722
+ },
723
+ contextWindow: 1000000,
724
+ maxTokens: 128000,
725
+ },
658
726
  "global.anthropic.claude-sonnet-4-5-20250929-v1:0": {
659
727
  id: "global.anthropic.claude-sonnet-4-5-20250929-v1:0",
660
728
  name: "Claude Sonnet 4.5 (Global)",
@@ -792,6 +860,23 @@ export const MODELS = {
792
860
  contextWindow: 1000000,
793
861
  maxTokens: 128000,
794
862
  },
863
+ "jp.anthropic.claude-opus-5": {
864
+ id: "jp.anthropic.claude-opus-5",
865
+ name: "Claude Opus 5 (JP)",
866
+ api: "bedrock-converse-stream",
867
+ provider: "amazon-bedrock",
868
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
869
+ reasoning: true,
870
+ input: ["text", "image"],
871
+ cost: {
872
+ input: 5,
873
+ output: 25,
874
+ cacheRead: 0.5,
875
+ cacheWrite: 6.25,
876
+ },
877
+ contextWindow: 1000000,
878
+ maxTokens: 128000,
879
+ },
795
880
  "jp.anthropic.claude-sonnet-4-5-20250929-v1:0": {
796
881
  id: "jp.anthropic.claude-sonnet-4-5-20250929-v1:0",
797
882
  name: "Claude Sonnet 4.5 (JP)",
@@ -1270,6 +1355,60 @@ export const MODELS = {
1270
1355
  contextWindow: 272000,
1271
1356
  maxTokens: 128000,
1272
1357
  },
1358
+ "openai.gpt-5.6-luna": {
1359
+ id: "openai.gpt-5.6-luna",
1360
+ name: "GPT-5.6 Luna",
1361
+ api: "bedrock-converse-stream",
1362
+ provider: "amazon-bedrock",
1363
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1364
+ reasoning: true,
1365
+ thinkingLevelMap: { "xhigh": "xhigh" },
1366
+ input: ["text", "image"],
1367
+ cost: {
1368
+ input: 1,
1369
+ output: 6,
1370
+ cacheRead: 0.1,
1371
+ cacheWrite: 1.25,
1372
+ },
1373
+ contextWindow: 272000,
1374
+ maxTokens: 128000,
1375
+ },
1376
+ "openai.gpt-5.6-sol": {
1377
+ id: "openai.gpt-5.6-sol",
1378
+ name: "GPT-5.6 Sol",
1379
+ api: "bedrock-converse-stream",
1380
+ provider: "amazon-bedrock",
1381
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1382
+ reasoning: true,
1383
+ thinkingLevelMap: { "xhigh": "xhigh" },
1384
+ input: ["text", "image"],
1385
+ cost: {
1386
+ input: 5,
1387
+ output: 30,
1388
+ cacheRead: 0.5,
1389
+ cacheWrite: 6.25,
1390
+ },
1391
+ contextWindow: 272000,
1392
+ maxTokens: 128000,
1393
+ },
1394
+ "openai.gpt-5.6-terra": {
1395
+ id: "openai.gpt-5.6-terra",
1396
+ name: "GPT-5.6 Terra",
1397
+ api: "bedrock-converse-stream",
1398
+ provider: "amazon-bedrock",
1399
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1400
+ reasoning: true,
1401
+ thinkingLevelMap: { "xhigh": "xhigh" },
1402
+ input: ["text", "image"],
1403
+ cost: {
1404
+ input: 2.5,
1405
+ output: 15,
1406
+ cacheRead: 0.25,
1407
+ cacheWrite: 3.125,
1408
+ },
1409
+ contextWindow: 272000,
1410
+ maxTokens: 128000,
1411
+ },
1273
1412
  "openai.gpt-oss-120b": {
1274
1413
  id: "openai.gpt-oss-120b",
1275
1414
  name: "gpt-oss-120b",
@@ -1612,6 +1751,23 @@ export const MODELS = {
1612
1751
  contextWindow: 1000000,
1613
1752
  maxTokens: 128000,
1614
1753
  },
1754
+ "us.anthropic.claude-opus-5": {
1755
+ id: "us.anthropic.claude-opus-5",
1756
+ name: "Claude Opus 5 (US)",
1757
+ api: "bedrock-converse-stream",
1758
+ provider: "amazon-bedrock",
1759
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1760
+ reasoning: true,
1761
+ input: ["text", "image"],
1762
+ cost: {
1763
+ input: 5,
1764
+ output: 25,
1765
+ cacheRead: 0.5,
1766
+ cacheWrite: 6.25,
1767
+ },
1768
+ contextWindow: 1000000,
1769
+ maxTokens: 128000,
1770
+ },
1615
1771
  "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
1616
1772
  id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
1617
1773
  name: "Claude Sonnet 4.5 (US)",
@@ -1993,6 +2149,23 @@ export const MODELS = {
1993
2149
  contextWindow: 1000000,
1994
2150
  maxTokens: 128000,
1995
2151
  },
2152
+ "claude-opus-5": {
2153
+ id: "claude-opus-5",
2154
+ name: "Claude Opus 5",
2155
+ api: "anthropic-messages",
2156
+ provider: "anthropic",
2157
+ baseUrl: "https://api.anthropic.com",
2158
+ reasoning: true,
2159
+ input: ["text", "image"],
2160
+ cost: {
2161
+ input: 5,
2162
+ output: 25,
2163
+ cacheRead: 0.5,
2164
+ cacheWrite: 6.25,
2165
+ },
2166
+ contextWindow: 1000000,
2167
+ maxTokens: 128000,
2168
+ },
1996
2169
  "claude-sonnet-4-5": {
1997
2170
  id: "claude-sonnet-4-5",
1998
2171
  name: "Claude Sonnet 4.5 (latest)",
@@ -2270,24 +2443,6 @@ export const MODELS = {
2270
2443
  contextWindow: 128000,
2271
2444
  maxTokens: 16384,
2272
2445
  },
2273
- "gpt-5-codex": {
2274
- id: "gpt-5-codex",
2275
- name: "GPT-5-Codex",
2276
- api: "azure-openai-responses",
2277
- provider: "azure-openai-responses",
2278
- baseUrl: "",
2279
- reasoning: true,
2280
- thinkingLevelMap: { "off": null },
2281
- input: ["text", "image"],
2282
- cost: {
2283
- input: 1.25,
2284
- output: 10,
2285
- cacheRead: 0.125,
2286
- cacheWrite: 0,
2287
- },
2288
- contextWindow: 400000,
2289
- maxTokens: 128000,
2290
- },
2291
2446
  "gpt-5-mini": {
2292
2447
  id: "gpt-5-mini",
2293
2448
  name: "GPT-5 Mini",
@@ -2360,24 +2515,6 @@ export const MODELS = {
2360
2515
  contextWindow: 400000,
2361
2516
  maxTokens: 128000,
2362
2517
  },
2363
- "gpt-5.1-chat-latest": {
2364
- id: "gpt-5.1-chat-latest",
2365
- name: "GPT-5.1 Chat",
2366
- api: "azure-openai-responses",
2367
- provider: "azure-openai-responses",
2368
- baseUrl: "",
2369
- reasoning: true,
2370
- thinkingLevelMap: { "off": null },
2371
- input: ["text", "image"],
2372
- cost: {
2373
- input: 1.25,
2374
- output: 10,
2375
- cacheRead: 0.125,
2376
- cacheWrite: 0,
2377
- },
2378
- contextWindow: 128000,
2379
- maxTokens: 16384,
2380
- },
2381
2518
  "gpt-5.1-codex": {
2382
2519
  id: "gpt-5.1-codex",
2383
2520
  name: "GPT-5.1 Codex",
@@ -2389,9 +2526,9 @@ export const MODELS = {
2389
2526
  input: ["text", "image"],
2390
2527
  cost: {
2391
2528
  input: 1.25,
2392
- output: 10,
2529
+ output: 5,
2393
2530
  cacheRead: 0.125,
2394
- cacheWrite: 0,
2531
+ cacheWrite: 1.25,
2395
2532
  },
2396
2533
  contextWindow: 400000,
2397
2534
  maxTokens: 128000,
@@ -2414,24 +2551,6 @@ export const MODELS = {
2414
2551
  contextWindow: 400000,
2415
2552
  maxTokens: 128000,
2416
2553
  },
2417
- "gpt-5.1-codex-mini": {
2418
- id: "gpt-5.1-codex-mini",
2419
- name: "GPT-5.1 Codex mini",
2420
- api: "azure-openai-responses",
2421
- provider: "azure-openai-responses",
2422
- baseUrl: "",
2423
- reasoning: true,
2424
- thinkingLevelMap: { "off": null },
2425
- input: ["text", "image"],
2426
- cost: {
2427
- input: 0.25,
2428
- output: 2,
2429
- cacheRead: 0.025,
2430
- cacheWrite: 0,
2431
- },
2432
- contextWindow: 400000,
2433
- maxTokens: 128000,
2434
- },
2435
2554
  "gpt-5.2": {
2436
2555
  id: "gpt-5.2",
2437
2556
  name: "GPT-5.2",
@@ -2468,24 +2587,6 @@ export const MODELS = {
2468
2587
  contextWindow: 128000,
2469
2588
  maxTokens: 16384,
2470
2589
  },
2471
- "gpt-5.2-codex": {
2472
- id: "gpt-5.2-codex",
2473
- name: "GPT-5.2 Codex",
2474
- api: "azure-openai-responses",
2475
- provider: "azure-openai-responses",
2476
- baseUrl: "",
2477
- reasoning: true,
2478
- thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
2479
- input: ["text", "image"],
2480
- cost: {
2481
- input: 1.75,
2482
- output: 14,
2483
- cacheRead: 0.175,
2484
- cacheWrite: 0,
2485
- },
2486
- contextWindow: 400000,
2487
- maxTokens: 128000,
2488
- },
2489
2590
  "gpt-5.2-pro": {
2490
2591
  id: "gpt-5.2-pro",
2491
2592
  name: "GPT-5.2 Pro",
@@ -2806,23 +2907,6 @@ export const MODELS = {
2806
2907
  contextWindow: 200000,
2807
2908
  maxTokens: 100000,
2808
2909
  },
2809
- "o3-deep-research": {
2810
- id: "o3-deep-research",
2811
- name: "o3-deep-research",
2812
- api: "azure-openai-responses",
2813
- provider: "azure-openai-responses",
2814
- baseUrl: "",
2815
- reasoning: true,
2816
- input: ["text", "image"],
2817
- cost: {
2818
- input: 10,
2819
- output: 40,
2820
- cacheRead: 2.5,
2821
- cacheWrite: 0,
2822
- },
2823
- contextWindow: 200000,
2824
- maxTokens: 100000,
2825
- },
2826
2910
  "o3-mini": {
2827
2911
  id: "o3-mini",
2828
2912
  name: "o3-mini",
@@ -2874,23 +2958,6 @@ export const MODELS = {
2874
2958
  contextWindow: 200000,
2875
2959
  maxTokens: 100000,
2876
2960
  },
2877
- "o4-mini-deep-research": {
2878
- id: "o4-mini-deep-research",
2879
- name: "o4-mini-deep-research",
2880
- api: "azure-openai-responses",
2881
- provider: "azure-openai-responses",
2882
- baseUrl: "",
2883
- reasoning: true,
2884
- input: ["text", "image"],
2885
- cost: {
2886
- input: 2,
2887
- output: 8,
2888
- cacheRead: 0.5,
2889
- cacheWrite: 0,
2890
- },
2891
- contextWindow: 200000,
2892
- maxTokens: 100000,
2893
- },
2894
2961
  },
2895
2962
  "cerebras": {
2896
2963
  "gemma-4-31b": {
@@ -2938,7 +3005,7 @@ export const MODELS = {
2938
3005
  cost: {
2939
3006
  input: 2.25,
2940
3007
  output: 2.75,
2941
- cacheRead: 0,
3008
+ cacheRead: 2.25,
2942
3009
  cacheWrite: 0,
2943
3010
  },
2944
3011
  contextWindow: 131072,
@@ -3189,20 +3256,37 @@ export const MODELS = {
3189
3256
  contextWindow: 1000000,
3190
3257
  maxTokens: 128000,
3191
3258
  },
3192
- "claude-sonnet-4": {
3193
- id: "claude-sonnet-4",
3194
- name: "Claude Sonnet 4 (latest)",
3259
+ "claude-opus-5": {
3260
+ id: "claude-opus-5",
3261
+ name: "Claude Opus 5",
3195
3262
  api: "anthropic-messages",
3196
3263
  provider: "cloudflare-ai-gateway",
3197
3264
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3198
3265
  reasoning: true,
3199
3266
  input: ["text", "image"],
3200
3267
  cost: {
3201
- input: 3,
3202
- output: 15,
3203
- cacheRead: 0.3,
3204
- cacheWrite: 3.75,
3205
- },
3268
+ input: 5,
3269
+ output: 25,
3270
+ cacheRead: 0.5,
3271
+ cacheWrite: 6.25,
3272
+ },
3273
+ contextWindow: 1000000,
3274
+ maxTokens: 128000,
3275
+ },
3276
+ "claude-sonnet-4": {
3277
+ id: "claude-sonnet-4",
3278
+ name: "Claude Sonnet 4 (latest)",
3279
+ api: "anthropic-messages",
3280
+ provider: "cloudflare-ai-gateway",
3281
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3282
+ reasoning: true,
3283
+ input: ["text", "image"],
3284
+ cost: {
3285
+ input: 3,
3286
+ output: 15,
3287
+ cacheRead: 0.3,
3288
+ cacheWrite: 3.75,
3289
+ },
3206
3290
  contextWindow: 200000,
3207
3291
  maxTokens: 64000,
3208
3292
  },
@@ -3452,6 +3536,60 @@ export const MODELS = {
3452
3536
  contextWindow: 1050000,
3453
3537
  maxTokens: 128000,
3454
3538
  },
3539
+ "gpt-5.6-luna": {
3540
+ id: "gpt-5.6-luna",
3541
+ name: "GPT-5.6 Luna",
3542
+ api: "openai-responses",
3543
+ provider: "cloudflare-ai-gateway",
3544
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3545
+ reasoning: true,
3546
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3547
+ input: ["text", "image"],
3548
+ cost: {
3549
+ input: 1,
3550
+ output: 6,
3551
+ cacheRead: 0.1,
3552
+ cacheWrite: 0,
3553
+ },
3554
+ contextWindow: 1050000,
3555
+ maxTokens: 128000,
3556
+ },
3557
+ "gpt-5.6-sol": {
3558
+ id: "gpt-5.6-sol",
3559
+ name: "GPT-5.6 Sol",
3560
+ api: "openai-responses",
3561
+ provider: "cloudflare-ai-gateway",
3562
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3563
+ reasoning: true,
3564
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3565
+ input: ["text", "image"],
3566
+ cost: {
3567
+ input: 5,
3568
+ output: 30,
3569
+ cacheRead: 0.5,
3570
+ cacheWrite: 0,
3571
+ },
3572
+ contextWindow: 1050000,
3573
+ maxTokens: 128000,
3574
+ },
3575
+ "gpt-5.6-terra": {
3576
+ id: "gpt-5.6-terra",
3577
+ name: "GPT-5.6 Terra",
3578
+ api: "openai-responses",
3579
+ provider: "cloudflare-ai-gateway",
3580
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3581
+ reasoning: true,
3582
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3583
+ input: ["text", "image"],
3584
+ cost: {
3585
+ input: 2.5,
3586
+ output: 15,
3587
+ cacheRead: 0.25,
3588
+ cacheWrite: 0,
3589
+ },
3590
+ contextWindow: 1050000,
3591
+ maxTokens: 128000,
3592
+ },
3455
3593
  "o1": {
3456
3594
  id: "o1",
3457
3595
  name: "o1",
@@ -3609,6 +3747,24 @@ export const MODELS = {
3609
3747
  contextWindow: 131072,
3610
3748
  maxTokens: 131072,
3611
3749
  },
3750
+ "workers-ai/@cf/zai-org/glm-5.2": {
3751
+ id: "workers-ai/@cf/zai-org/glm-5.2",
3752
+ name: "Glm 5.2",
3753
+ api: "openai-completions",
3754
+ provider: "cloudflare-ai-gateway",
3755
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3756
+ compat: { "sendSessionAffinityHeaders": true },
3757
+ reasoning: true,
3758
+ input: ["text"],
3759
+ cost: {
3760
+ input: 1.4,
3761
+ output: 4.4,
3762
+ cacheRead: 0.26,
3763
+ cacheWrite: 0,
3764
+ },
3765
+ contextWindow: 262144,
3766
+ maxTokens: 262144,
3767
+ },
3612
3768
  },
3613
3769
  "cloudflare-workers-ai": {
3614
3770
  "@cf/google/gemma-4-26b-a4b-it": {
@@ -3923,24 +4079,6 @@ export const MODELS = {
3923
4079
  contextWindow: 1000000,
3924
4080
  maxTokens: 384000,
3925
4081
  },
3926
- "accounts/fireworks/models/glm-5p1": {
3927
- id: "accounts/fireworks/models/glm-5p1",
3928
- name: "GLM 5.1",
3929
- api: "anthropic-messages",
3930
- provider: "fireworks",
3931
- baseUrl: "https://api.fireworks.ai/inference",
3932
- compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
3933
- reasoning: true,
3934
- input: ["text"],
3935
- cost: {
3936
- input: 1.4,
3937
- output: 4.4,
3938
- cacheRead: 0.26,
3939
- cacheWrite: 0,
3940
- },
3941
- contextWindow: 202800,
3942
- maxTokens: 131072,
3943
- },
3944
4082
  "accounts/fireworks/models/glm-5p2": {
3945
4083
  id: "accounts/fireworks/models/glm-5p2",
3946
4084
  name: "GLM 5.2",
@@ -4057,7 +4195,7 @@ export const MODELS = {
4057
4195
  baseUrl: "https://api.fireworks.ai/inference",
4058
4196
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4059
4197
  reasoning: true,
4060
- input: ["text"],
4198
+ input: ["text", "image"],
4061
4199
  cost: {
4062
4200
  input: 0.3,
4063
4201
  output: 1.2,
@@ -4085,24 +4223,6 @@ export const MODELS = {
4085
4223
  contextWindow: 262144,
4086
4224
  maxTokens: 65536,
4087
4225
  },
4088
- "accounts/fireworks/routers/glm-5p1-fast": {
4089
- id: "accounts/fireworks/routers/glm-5p1-fast",
4090
- name: "GLM 5.1 Fast",
4091
- api: "anthropic-messages",
4092
- provider: "fireworks",
4093
- baseUrl: "https://api.fireworks.ai/inference",
4094
- compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4095
- reasoning: true,
4096
- input: ["text"],
4097
- cost: {
4098
- input: 2.8,
4099
- output: 8.8,
4100
- cacheRead: 0.52,
4101
- cacheWrite: 0,
4102
- },
4103
- contextWindow: 202800,
4104
- maxTokens: 131072,
4105
- },
4106
4226
  "accounts/fireworks/routers/glm-5p2-fast": {
4107
4227
  id: "accounts/fireworks/routers/glm-5p2-fast",
4108
4228
  name: "GLM 5.2 Fast",
@@ -4292,6 +4412,25 @@ export const MODELS = {
4292
4412
  contextWindow: 200000,
4293
4413
  maxTokens: 64000,
4294
4414
  },
4415
+ "claude-opus-5": {
4416
+ id: "claude-opus-5",
4417
+ name: "Claude Opus 5",
4418
+ api: "openai-completions",
4419
+ provider: "github-copilot",
4420
+ baseUrl: "https://api.individual.githubcopilot.com",
4421
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4422
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4423
+ reasoning: true,
4424
+ input: ["text", "image"],
4425
+ cost: {
4426
+ input: 5,
4427
+ output: 25,
4428
+ cacheRead: 0.5,
4429
+ cacheWrite: 6.25,
4430
+ },
4431
+ contextWindow: 1000000,
4432
+ maxTokens: 64000,
4433
+ },
4295
4434
  "claude-sonnet-4": {
4296
4435
  id: "claude-sonnet-4",
4297
4436
  name: "Claude Sonnet 4 (latest)",
@@ -4629,7 +4768,7 @@ export const MODELS = {
4629
4768
  input: 1,
4630
4769
  output: 6,
4631
4770
  cacheRead: 0.1,
4632
- cacheWrite: 0,
4771
+ cacheWrite: 1.25,
4633
4772
  },
4634
4773
  contextWindow: 1050000,
4635
4774
  maxTokens: 128000,
@@ -4648,7 +4787,7 @@ export const MODELS = {
4648
4787
  input: 5,
4649
4788
  output: 30,
4650
4789
  cacheRead: 0.5,
4651
- cacheWrite: 0,
4790
+ cacheWrite: 6.25,
4652
4791
  },
4653
4792
  contextWindow: 1050000,
4654
4793
  maxTokens: 128000,
@@ -4667,7 +4806,7 @@ export const MODELS = {
4667
4806
  input: 2.5,
4668
4807
  output: 15,
4669
4808
  cacheRead: 0.25,
4670
- cacheWrite: 0,
4809
+ cacheWrite: 3.125,
4671
4810
  },
4672
4811
  contextWindow: 1050000,
4673
4812
  maxTokens: 128000,
@@ -4712,6 +4851,40 @@ export const MODELS = {
4712
4851
  },
4713
4852
  },
4714
4853
  "google": {
4854
+ "deep-research-max-preview-04-2026": {
4855
+ id: "deep-research-max-preview-04-2026",
4856
+ name: "Deep Research Max Preview (Apr-21-2026)",
4857
+ api: "google-generative-ai",
4858
+ provider: "google",
4859
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4860
+ reasoning: true,
4861
+ input: ["text", "image"],
4862
+ cost: {
4863
+ input: 2,
4864
+ output: 12,
4865
+ cacheRead: 0.2,
4866
+ cacheWrite: 0,
4867
+ },
4868
+ contextWindow: 131072,
4869
+ maxTokens: 65536,
4870
+ },
4871
+ "deep-research-preview-04-2026": {
4872
+ id: "deep-research-preview-04-2026",
4873
+ name: "Deep Research Preview (Apr-21-2026)",
4874
+ api: "google-generative-ai",
4875
+ provider: "google",
4876
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4877
+ reasoning: true,
4878
+ input: ["text", "image"],
4879
+ cost: {
4880
+ input: 2,
4881
+ output: 12,
4882
+ cacheRead: 0.2,
4883
+ cacheWrite: 0,
4884
+ },
4885
+ contextWindow: 131072,
4886
+ maxTokens: 65536,
4887
+ },
4715
4888
  "gemini-2.0-flash": {
4716
4889
  id: "gemini-2.0-flash",
4717
4890
  name: "Gemini 2.0 Flash",
@@ -4746,6 +4919,23 @@ export const MODELS = {
4746
4919
  contextWindow: 1048576,
4747
4920
  maxTokens: 8192,
4748
4921
  },
4922
+ "gemini-2.5-computer-use-preview-10-2025": {
4923
+ id: "gemini-2.5-computer-use-preview-10-2025",
4924
+ name: "Gemini 2.5 Computer Use Preview 10-2025",
4925
+ api: "google-generative-ai",
4926
+ provider: "google",
4927
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4928
+ reasoning: true,
4929
+ input: ["text", "image"],
4930
+ cost: {
4931
+ input: 1.25,
4932
+ output: 10,
4933
+ cacheRead: 0,
4934
+ cacheWrite: 0,
4935
+ },
4936
+ contextWindow: 131072,
4937
+ maxTokens: 65536,
4938
+ },
4749
4939
  "gemini-2.5-flash": {
4750
4940
  id: "gemini-2.5-flash",
4751
4941
  name: "Gemini 2.5 Flash",
@@ -4851,6 +5041,24 @@ export const MODELS = {
4851
5041
  contextWindow: 1048576,
4852
5042
  maxTokens: 65536,
4853
5043
  },
5044
+ "gemini-3.1-flash-lite-image": {
5045
+ id: "gemini-3.1-flash-lite-image",
5046
+ name: "Nano Banana 2 Lite",
5047
+ api: "google-generative-ai",
5048
+ provider: "google",
5049
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5050
+ reasoning: true,
5051
+ thinkingLevelMap: { "off": null },
5052
+ input: ["text", "image"],
5053
+ cost: {
5054
+ input: 0.25,
5055
+ output: 30,
5056
+ cacheRead: 0,
5057
+ cacheWrite: 0,
5058
+ },
5059
+ contextWindow: 65536,
5060
+ maxTokens: 65536,
5061
+ },
4854
5062
  "gemini-3.1-flash-lite-preview": {
4855
5063
  id: "gemini-3.1-flash-lite-preview",
4856
5064
  name: "Gemini 3.1 Flash Lite Preview",
@@ -4869,6 +5077,24 @@ export const MODELS = {
4869
5077
  contextWindow: 1048576,
4870
5078
  maxTokens: 65536,
4871
5079
  },
5080
+ "gemini-3.1-flash-live-preview": {
5081
+ id: "gemini-3.1-flash-live-preview",
5082
+ name: "Gemini 3.1 Flash Live Preview",
5083
+ api: "google-generative-ai",
5084
+ provider: "google",
5085
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5086
+ reasoning: true,
5087
+ thinkingLevelMap: { "off": null },
5088
+ input: ["text", "image"],
5089
+ cost: {
5090
+ input: 0.75,
5091
+ output: 4.5,
5092
+ cacheRead: 0,
5093
+ cacheWrite: 0,
5094
+ },
5095
+ contextWindow: 131072,
5096
+ maxTokens: 65536,
5097
+ },
4872
5098
  "gemini-3.1-pro-preview": {
4873
5099
  id: "gemini-3.1-pro-preview",
4874
5100
  name: "Gemini 3.1 Pro Preview",
@@ -4923,66 +5149,119 @@ export const MODELS = {
4923
5149
  contextWindow: 1048576,
4924
5150
  maxTokens: 65536,
4925
5151
  },
4926
- "gemini-flash-latest": {
4927
- id: "gemini-flash-latest",
4928
- name: "Gemini Flash Latest",
5152
+ "gemini-3.5-flash-lite": {
5153
+ id: "gemini-3.5-flash-lite",
5154
+ name: "Gemini 3.5 Flash Lite",
4929
5155
  api: "google-generative-ai",
4930
5156
  provider: "google",
4931
5157
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4932
5158
  reasoning: true,
5159
+ thinkingLevelMap: { "off": null },
4933
5160
  input: ["text", "image"],
4934
5161
  cost: {
4935
5162
  input: 0.3,
4936
5163
  output: 2.5,
4937
- cacheRead: 0.075,
5164
+ cacheRead: 0.03,
4938
5165
  cacheWrite: 0,
4939
5166
  },
4940
5167
  contextWindow: 1048576,
4941
5168
  maxTokens: 65536,
4942
5169
  },
4943
- "gemini-flash-lite-latest": {
4944
- id: "gemini-flash-lite-latest",
4945
- name: "Gemini Flash-Lite Latest",
5170
+ "gemini-3.6-flash": {
5171
+ id: "gemini-3.6-flash",
5172
+ name: "Gemini 3.6 Flash",
4946
5173
  api: "google-generative-ai",
4947
5174
  provider: "google",
4948
5175
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4949
5176
  reasoning: true,
5177
+ thinkingLevelMap: { "off": null },
4950
5178
  input: ["text", "image"],
4951
5179
  cost: {
4952
- input: 0.1,
4953
- output: 0.4,
4954
- cacheRead: 0.025,
5180
+ input: 1.5,
5181
+ output: 7.5,
5182
+ cacheRead: 0.15,
4955
5183
  cacheWrite: 0,
4956
5184
  },
4957
5185
  contextWindow: 1048576,
4958
5186
  maxTokens: 65536,
4959
5187
  },
4960
- "gemma-4-26b-a4b-it": {
4961
- id: "gemma-4-26b-a4b-it",
4962
- name: "Gemma 4 26B A4B IT",
5188
+ "gemini-flash-latest": {
5189
+ id: "gemini-flash-latest",
5190
+ name: "Gemini Flash Latest",
4963
5191
  api: "google-generative-ai",
4964
5192
  provider: "google",
4965
5193
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4966
5194
  reasoning: true,
4967
- thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4968
5195
  input: ["text", "image"],
4969
5196
  cost: {
4970
- input: 0,
4971
- output: 0,
4972
- cacheRead: 0,
5197
+ input: 1.5,
5198
+ output: 9,
5199
+ cacheRead: 0.15,
4973
5200
  cacheWrite: 0,
4974
5201
  },
4975
- contextWindow: 262144,
4976
- maxTokens: 32768,
5202
+ contextWindow: 1048576,
5203
+ maxTokens: 65536,
4977
5204
  },
4978
- "gemma-4-31b-it": {
4979
- id: "gemma-4-31b-it",
4980
- name: "Gemma 4 31B IT",
5205
+ "gemini-flash-lite-latest": {
5206
+ id: "gemini-flash-lite-latest",
5207
+ name: "Gemini Flash-Lite Latest",
4981
5208
  api: "google-generative-ai",
4982
5209
  provider: "google",
4983
5210
  baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4984
5211
  reasoning: true,
4985
- thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
5212
+ input: ["text", "image"],
5213
+ cost: {
5214
+ input: 0.25,
5215
+ output: 1.5,
5216
+ cacheRead: 0.025,
5217
+ cacheWrite: 0,
5218
+ },
5219
+ contextWindow: 1048576,
5220
+ maxTokens: 65536,
5221
+ },
5222
+ "gemini-robotics-er-1.6-preview": {
5223
+ id: "gemini-robotics-er-1.6-preview",
5224
+ name: "Gemini Robotics-ER 1.6 Preview",
5225
+ api: "google-generative-ai",
5226
+ provider: "google",
5227
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5228
+ reasoning: true,
5229
+ input: ["text", "image"],
5230
+ cost: {
5231
+ input: 1,
5232
+ output: 5,
5233
+ cacheRead: 0,
5234
+ cacheWrite: 0,
5235
+ },
5236
+ contextWindow: 131072,
5237
+ maxTokens: 65536,
5238
+ },
5239
+ "gemma-4-26b-a4b-it": {
5240
+ id: "gemma-4-26b-a4b-it",
5241
+ name: "Gemma 4 26B A4B IT",
5242
+ api: "google-generative-ai",
5243
+ provider: "google",
5244
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5245
+ reasoning: true,
5246
+ thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
5247
+ input: ["text", "image"],
5248
+ cost: {
5249
+ input: 0,
5250
+ output: 0,
5251
+ cacheRead: 0,
5252
+ cacheWrite: 0,
5253
+ },
5254
+ contextWindow: 262144,
5255
+ maxTokens: 32768,
5256
+ },
5257
+ "gemma-4-31b-it": {
5258
+ id: "gemma-4-31b-it",
5259
+ name: "Gemma 4 31B IT",
5260
+ api: "google-generative-ai",
5261
+ provider: "google",
5262
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5263
+ reasoning: true,
5264
+ thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4986
5265
  input: ["text", "image"],
4987
5266
  cost: {
4988
5267
  input: 0,
@@ -5722,6 +6001,24 @@ export const MODELS = {
5722
6001
  contextWindow: 262144,
5723
6002
  maxTokens: 4096,
5724
6003
  },
6004
+ "XiaomiMiMo/MiMo-V2.5": {
6005
+ id: "XiaomiMiMo/MiMo-V2.5",
6006
+ name: "MiMo-V2.5",
6007
+ api: "openai-completions",
6008
+ provider: "huggingface",
6009
+ baseUrl: "https://router.huggingface.co/v1",
6010
+ compat: { "supportsDeveloperRole": false },
6011
+ reasoning: true,
6012
+ input: ["text"],
6013
+ cost: {
6014
+ input: 0.4,
6015
+ output: 2,
6016
+ cacheRead: 0,
6017
+ cacheWrite: 0,
6018
+ },
6019
+ contextWindow: 262144,
6020
+ maxTokens: 131072,
6021
+ },
5725
6022
  "XiaomiMiMo/MiMo-V2.5-Pro": {
5726
6023
  id: "XiaomiMiMo/MiMo-V2.5-Pro",
5727
6024
  name: "MiMo-V2.5-Pro",
@@ -6228,9 +6525,27 @@ export const MODELS = {
6228
6525
  },
6229
6526
  },
6230
6527
  "kimi-coding": {
6231
- "k2p7": {
6232
- id: "k2p7",
6233
- name: "Kimi K2.7 Code",
6528
+ "k3": {
6529
+ id: "k3",
6530
+ name: "Kimi K3",
6531
+ api: "anthropic-messages",
6532
+ provider: "kimi-coding",
6533
+ baseUrl: "https://api.kimi.com/coding",
6534
+ headers: { "User-Agent": "KimiCLI/1.5" },
6535
+ reasoning: true,
6536
+ input: ["text", "image"],
6537
+ cost: {
6538
+ input: 0,
6539
+ output: 0,
6540
+ cacheRead: 0,
6541
+ cacheWrite: 0,
6542
+ },
6543
+ contextWindow: 1048576,
6544
+ maxTokens: 131072,
6545
+ },
6546
+ "k3-256k": {
6547
+ id: "k3-256k",
6548
+ name: "Kimi K3-256K",
6234
6549
  api: "anthropic-messages",
6235
6550
  provider: "kimi-coding",
6236
6551
  baseUrl: "https://api.kimi.com/coding",
@@ -6244,11 +6559,11 @@ export const MODELS = {
6244
6559
  cacheWrite: 0,
6245
6560
  },
6246
6561
  contextWindow: 262144,
6247
- maxTokens: 32768,
6562
+ maxTokens: 131072,
6248
6563
  },
6249
6564
  "kimi-for-coding": {
6250
6565
  id: "kimi-for-coding",
6251
- name: "Kimi For Coding",
6566
+ name: "Kimi K2.7 Code",
6252
6567
  api: "anthropic-messages",
6253
6568
  provider: "kimi-coding",
6254
6569
  baseUrl: "https://api.kimi.com/coding",
@@ -6264,15 +6579,15 @@ export const MODELS = {
6264
6579
  contextWindow: 262144,
6265
6580
  maxTokens: 32768,
6266
6581
  },
6267
- "kimi-k2-thinking": {
6268
- id: "kimi-k2-thinking",
6269
- name: "Kimi K2 Thinking",
6582
+ "kimi-for-coding-highspeed": {
6583
+ id: "kimi-for-coding-highspeed",
6584
+ name: "Kimi For Coding HighSpeed",
6270
6585
  api: "anthropic-messages",
6271
6586
  provider: "kimi-coding",
6272
6587
  baseUrl: "https://api.kimi.com/coding",
6273
6588
  headers: { "User-Agent": "KimiCLI/1.5" },
6274
6589
  reasoning: true,
6275
- input: ["text"],
6590
+ input: ["text", "image"],
6276
6591
  cost: {
6277
6592
  input: 0,
6278
6593
  output: 0,
@@ -7030,6 +7345,24 @@ export const MODELS = {
7030
7345
  contextWindow: 262144,
7031
7346
  maxTokens: 262144,
7032
7347
  },
7348
+ "kimi-k3": {
7349
+ id: "kimi-k3",
7350
+ name: "Kimi K3",
7351
+ api: "openai-completions",
7352
+ provider: "moonshotai",
7353
+ baseUrl: "https://api.moonshot.ai/v1",
7354
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7355
+ reasoning: true,
7356
+ input: ["text", "image"],
7357
+ cost: {
7358
+ input: 3,
7359
+ output: 15,
7360
+ cacheRead: 0.3,
7361
+ cacheWrite: 0,
7362
+ },
7363
+ contextWindow: 1048576,
7364
+ maxTokens: 131072,
7365
+ },
7033
7366
  },
7034
7367
  "moonshotai-cn": {
7035
7368
  "kimi-k2-0711-preview": {
@@ -7194,6 +7527,24 @@ export const MODELS = {
7194
7527
  contextWindow: 262144,
7195
7528
  maxTokens: 262144,
7196
7529
  },
7530
+ "kimi-k3": {
7531
+ id: "kimi-k3",
7532
+ name: "Kimi K3",
7533
+ api: "openai-completions",
7534
+ provider: "moonshotai-cn",
7535
+ baseUrl: "https://api.moonshot.cn/v1",
7536
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7537
+ reasoning: true,
7538
+ input: ["text", "image"],
7539
+ cost: {
7540
+ input: 3,
7541
+ output: 15,
7542
+ cacheRead: 0.3,
7543
+ cacheWrite: 0,
7544
+ },
7545
+ contextWindow: 1048576,
7546
+ maxTokens: 131072,
7547
+ },
7197
7548
  },
7198
7549
  "openai": {
7199
7550
  "gpt-4": {
@@ -7402,24 +7753,6 @@ export const MODELS = {
7402
7753
  contextWindow: 128000,
7403
7754
  maxTokens: 16384,
7404
7755
  },
7405
- "gpt-5-codex": {
7406
- id: "gpt-5-codex",
7407
- name: "GPT-5-Codex",
7408
- api: "openai-responses",
7409
- provider: "openai",
7410
- baseUrl: "https://api.openai.com/v1",
7411
- reasoning: true,
7412
- thinkingLevelMap: { "off": null },
7413
- input: ["text", "image"],
7414
- cost: {
7415
- input: 1.25,
7416
- output: 10,
7417
- cacheRead: 0.125,
7418
- cacheWrite: 0,
7419
- },
7420
- contextWindow: 400000,
7421
- maxTokens: 128000,
7422
- },
7423
7756
  "gpt-5-mini": {
7424
7757
  id: "gpt-5-mini",
7425
7758
  name: "GPT-5 Mini",
@@ -7492,24 +7825,6 @@ export const MODELS = {
7492
7825
  contextWindow: 400000,
7493
7826
  maxTokens: 128000,
7494
7827
  },
7495
- "gpt-5.1-chat-latest": {
7496
- id: "gpt-5.1-chat-latest",
7497
- name: "GPT-5.1 Chat",
7498
- api: "openai-responses",
7499
- provider: "openai",
7500
- baseUrl: "https://api.openai.com/v1",
7501
- reasoning: true,
7502
- thinkingLevelMap: { "off": null },
7503
- input: ["text", "image"],
7504
- cost: {
7505
- input: 1.25,
7506
- output: 10,
7507
- cacheRead: 0.125,
7508
- cacheWrite: 0,
7509
- },
7510
- contextWindow: 128000,
7511
- maxTokens: 16384,
7512
- },
7513
7828
  "gpt-5.1-codex": {
7514
7829
  id: "gpt-5.1-codex",
7515
7830
  name: "GPT-5.1 Codex",
@@ -7521,9 +7836,9 @@ export const MODELS = {
7521
7836
  input: ["text", "image"],
7522
7837
  cost: {
7523
7838
  input: 1.25,
7524
- output: 10,
7839
+ output: 5,
7525
7840
  cacheRead: 0.125,
7526
- cacheWrite: 0,
7841
+ cacheWrite: 1.25,
7527
7842
  },
7528
7843
  contextWindow: 400000,
7529
7844
  maxTokens: 128000,
@@ -7546,24 +7861,6 @@ export const MODELS = {
7546
7861
  contextWindow: 400000,
7547
7862
  maxTokens: 128000,
7548
7863
  },
7549
- "gpt-5.1-codex-mini": {
7550
- id: "gpt-5.1-codex-mini",
7551
- name: "GPT-5.1 Codex mini",
7552
- api: "openai-responses",
7553
- provider: "openai",
7554
- baseUrl: "https://api.openai.com/v1",
7555
- reasoning: true,
7556
- thinkingLevelMap: { "off": null },
7557
- input: ["text", "image"],
7558
- cost: {
7559
- input: 0.25,
7560
- output: 2,
7561
- cacheRead: 0.025,
7562
- cacheWrite: 0,
7563
- },
7564
- contextWindow: 400000,
7565
- maxTokens: 128000,
7566
- },
7567
7864
  "gpt-5.2": {
7568
7865
  id: "gpt-5.2",
7569
7866
  name: "GPT-5.2",
@@ -7600,24 +7897,6 @@ export const MODELS = {
7600
7897
  contextWindow: 128000,
7601
7898
  maxTokens: 16384,
7602
7899
  },
7603
- "gpt-5.2-codex": {
7604
- id: "gpt-5.2-codex",
7605
- name: "GPT-5.2 Codex",
7606
- api: "openai-responses",
7607
- provider: "openai",
7608
- baseUrl: "https://api.openai.com/v1",
7609
- reasoning: true,
7610
- thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
7611
- input: ["text", "image"],
7612
- cost: {
7613
- input: 1.75,
7614
- output: 14,
7615
- cacheRead: 0.175,
7616
- cacheWrite: 0,
7617
- },
7618
- contextWindow: 400000,
7619
- maxTokens: 128000,
7620
- },
7621
7900
  "gpt-5.2-pro": {
7622
7901
  id: "gpt-5.2-pro",
7623
7902
  name: "GPT-5.2 Pro",
@@ -7938,23 +8217,6 @@ export const MODELS = {
7938
8217
  contextWindow: 200000,
7939
8218
  maxTokens: 100000,
7940
8219
  },
7941
- "o3-deep-research": {
7942
- id: "o3-deep-research",
7943
- name: "o3-deep-research",
7944
- api: "openai-responses",
7945
- provider: "openai",
7946
- baseUrl: "https://api.openai.com/v1",
7947
- reasoning: true,
7948
- input: ["text", "image"],
7949
- cost: {
7950
- input: 10,
7951
- output: 40,
7952
- cacheRead: 2.5,
7953
- cacheWrite: 0,
7954
- },
7955
- contextWindow: 200000,
7956
- maxTokens: 100000,
7957
- },
7958
8220
  "o3-mini": {
7959
8221
  id: "o3-mini",
7960
8222
  name: "o3-mini",
@@ -8006,23 +8268,6 @@ export const MODELS = {
8006
8268
  contextWindow: 200000,
8007
8269
  maxTokens: 100000,
8008
8270
  },
8009
- "o4-mini-deep-research": {
8010
- id: "o4-mini-deep-research",
8011
- name: "o4-mini-deep-research",
8012
- api: "openai-responses",
8013
- provider: "openai",
8014
- baseUrl: "https://api.openai.com/v1",
8015
- reasoning: true,
8016
- input: ["text", "image"],
8017
- cost: {
8018
- input: 2,
8019
- output: 8,
8020
- cacheRead: 0.5,
8021
- cacheWrite: 0,
8022
- },
8023
- contextWindow: 200000,
8024
- maxTokens: 100000,
8025
- },
8026
8271
  },
8027
8272
  "openai-codex": {
8028
8273
  "gpt-5.2": {
@@ -8348,6 +8593,23 @@ export const MODELS = {
8348
8593
  contextWindow: 1000000,
8349
8594
  maxTokens: 128000,
8350
8595
  },
8596
+ "claude-opus-5": {
8597
+ id: "claude-opus-5",
8598
+ name: "Claude Opus 5",
8599
+ api: "anthropic-messages",
8600
+ provider: "opencode",
8601
+ baseUrl: "https://opencode.ai/zen",
8602
+ reasoning: true,
8603
+ input: ["text", "image"],
8604
+ cost: {
8605
+ input: 5,
8606
+ output: 25,
8607
+ cacheRead: 0.5,
8608
+ cacheWrite: 6.25,
8609
+ },
8610
+ contextWindow: 1000000,
8611
+ maxTokens: 128000,
8612
+ },
8351
8613
  "claude-sonnet-4": {
8352
8614
  id: "claude-sonnet-4",
8353
8615
  name: "Claude Sonnet 4",
@@ -8528,26 +8790,62 @@ export const MODELS = {
8528
8790
  contextWindow: 1048576,
8529
8791
  maxTokens: 65536,
8530
8792
  },
8531
- "glm-5": {
8532
- id: "glm-5",
8533
- name: "GLM-5",
8534
- api: "openai-completions",
8793
+ "gemini-3.5-flash-lite": {
8794
+ id: "gemini-3.5-flash-lite",
8795
+ name: "Gemini 3.5 Flash Lite",
8796
+ api: "google-generative-ai",
8535
8797
  provider: "opencode",
8536
8798
  baseUrl: "https://opencode.ai/zen/v1",
8537
8799
  reasoning: true,
8538
- input: ["text"],
8800
+ thinkingLevelMap: { "off": null },
8801
+ input: ["text", "image"],
8539
8802
  cost: {
8540
- input: 1,
8541
- output: 3.2,
8542
- cacheRead: 0.2,
8803
+ input: 0.3,
8804
+ output: 2.5,
8805
+ cacheRead: 0.03,
8543
8806
  cacheWrite: 0,
8544
8807
  },
8545
- contextWindow: 204800,
8546
- maxTokens: 131072,
8808
+ contextWindow: 1048576,
8809
+ maxTokens: 65536,
8547
8810
  },
8548
- "glm-5.1": {
8549
- id: "glm-5.1",
8550
- name: "GLM-5.1",
8811
+ "gemini-3.6-flash": {
8812
+ id: "gemini-3.6-flash",
8813
+ name: "Gemini 3.6 Flash",
8814
+ api: "google-generative-ai",
8815
+ provider: "opencode",
8816
+ baseUrl: "https://opencode.ai/zen/v1",
8817
+ reasoning: true,
8818
+ thinkingLevelMap: { "off": null },
8819
+ input: ["text", "image"],
8820
+ cost: {
8821
+ input: 1.5,
8822
+ output: 7.5,
8823
+ cacheRead: 0.15,
8824
+ cacheWrite: 0,
8825
+ },
8826
+ contextWindow: 1048576,
8827
+ maxTokens: 65536,
8828
+ },
8829
+ "glm-5": {
8830
+ id: "glm-5",
8831
+ name: "GLM-5",
8832
+ api: "openai-completions",
8833
+ provider: "opencode",
8834
+ baseUrl: "https://opencode.ai/zen/v1",
8835
+ reasoning: true,
8836
+ input: ["text"],
8837
+ cost: {
8838
+ input: 1,
8839
+ output: 3.2,
8840
+ cacheRead: 0.2,
8841
+ cacheWrite: 0,
8842
+ },
8843
+ contextWindow: 204800,
8844
+ maxTokens: 131072,
8845
+ },
8846
+ "glm-5.1": {
8847
+ id: "glm-5.1",
8848
+ name: "GLM-5.1",
8551
8849
  api: "openai-completions",
8552
8850
  provider: "opencode",
8553
8851
  baseUrl: "https://opencode.ai/zen/v1",
@@ -8924,7 +9222,7 @@ export const MODELS = {
8924
9222
  "grok-4.5": {
8925
9223
  id: "grok-4.5",
8926
9224
  name: "Grok 4.5",
8927
- api: "openai-completions",
9225
+ api: "openai-responses",
8928
9226
  provider: "opencode",
8929
9227
  baseUrl: "https://opencode.ai/zen/v1",
8930
9228
  reasoning: true,
@@ -8957,23 +9255,6 @@ export const MODELS = {
8957
9255
  contextWindow: 256000,
8958
9256
  maxTokens: 256000,
8959
9257
  },
8960
- "hy3-free": {
8961
- id: "hy3-free",
8962
- name: "Hy3 Free",
8963
- api: "openai-completions",
8964
- provider: "opencode",
8965
- baseUrl: "https://opencode.ai/zen/v1",
8966
- reasoning: true,
8967
- input: ["text"],
8968
- cost: {
8969
- input: 0,
8970
- output: 0,
8971
- cacheRead: 0,
8972
- cacheWrite: 0,
8973
- },
8974
- contextWindow: 190000,
8975
- maxTokens: 64000,
8976
- },
8977
9258
  "kimi-k2.5": {
8978
9259
  id: "kimi-k2.5",
8979
9260
  name: "Kimi K2.5",
@@ -9026,6 +9307,40 @@ export const MODELS = {
9026
9307
  contextWindow: 262144,
9027
9308
  maxTokens: 262144,
9028
9309
  },
9310
+ "laguna-s-2.1-free": {
9311
+ id: "laguna-s-2.1-free",
9312
+ name: "Laguna S 2.1 Free",
9313
+ api: "openai-completions",
9314
+ provider: "opencode",
9315
+ baseUrl: "https://opencode.ai/zen/v1",
9316
+ reasoning: true,
9317
+ input: ["text"],
9318
+ cost: {
9319
+ input: 0,
9320
+ output: 0,
9321
+ cacheRead: 0,
9322
+ cacheWrite: 0,
9323
+ },
9324
+ contextWindow: 256000,
9325
+ maxTokens: 32000,
9326
+ },
9327
+ "ling-3.0-flash-free": {
9328
+ id: "ling-3.0-flash-free",
9329
+ name: "Ling-3.0-flash Free",
9330
+ api: "openai-completions",
9331
+ provider: "opencode",
9332
+ baseUrl: "https://opencode.ai/zen/v1",
9333
+ reasoning: true,
9334
+ input: ["text"],
9335
+ cost: {
9336
+ input: 0,
9337
+ output: 0,
9338
+ cacheRead: 0,
9339
+ cacheWrite: 0,
9340
+ },
9341
+ contextWindow: 262144,
9342
+ maxTokens: 32768,
9343
+ },
9029
9344
  "mimo-v2.5-free": {
9030
9345
  id: "mimo-v2.5-free",
9031
9346
  name: "MiMo V2.5 Free",
@@ -9194,9 +9509,9 @@ export const MODELS = {
9194
9509
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9195
9510
  input: ["text"],
9196
9511
  cost: {
9197
- input: 1.74,
9198
- output: 3.48,
9199
- cacheRead: 0.0145,
9512
+ input: 0.435,
9513
+ output: 0.87,
9514
+ cacheRead: 0.003625,
9200
9515
  cacheWrite: 0,
9201
9516
  },
9202
9517
  contextWindow: 1000000,
@@ -9236,6 +9551,40 @@ export const MODELS = {
9236
9551
  contextWindow: 1000000,
9237
9552
  maxTokens: 131072,
9238
9553
  },
9554
+ "grok-4.5": {
9555
+ id: "grok-4.5",
9556
+ name: "Grok 4.5",
9557
+ api: "openai-responses",
9558
+ provider: "opencode-go",
9559
+ baseUrl: "https://opencode.ai/zen/go/v1",
9560
+ reasoning: true,
9561
+ input: ["text", "image"],
9562
+ cost: {
9563
+ input: 2,
9564
+ output: 6,
9565
+ cacheRead: 0.5,
9566
+ cacheWrite: 0,
9567
+ },
9568
+ contextWindow: 500000,
9569
+ maxTokens: 500000,
9570
+ },
9571
+ "hy3": {
9572
+ id: "hy3",
9573
+ name: "Hy3",
9574
+ api: "openai-completions",
9575
+ provider: "opencode-go",
9576
+ baseUrl: "https://opencode.ai/zen/go/v1",
9577
+ reasoning: true,
9578
+ input: ["text"],
9579
+ cost: {
9580
+ input: 0.14,
9581
+ output: 0.58,
9582
+ cacheRead: 0.035,
9583
+ cacheWrite: 0,
9584
+ },
9585
+ contextWindow: 256000,
9586
+ maxTokens: 64000,
9587
+ },
9239
9588
  "kimi-k2.6": {
9240
9589
  id: "kimi-k2.6",
9241
9590
  name: "Kimi K2.6",
@@ -9272,6 +9621,23 @@ export const MODELS = {
9272
9621
  contextWindow: 262144,
9273
9622
  maxTokens: 262144,
9274
9623
  },
9624
+ "kimi-k3": {
9625
+ id: "kimi-k3",
9626
+ name: "Kimi K3 (2x usage)",
9627
+ api: "openai-completions",
9628
+ provider: "opencode-go",
9629
+ baseUrl: "https://opencode.ai/zen/go/v1",
9630
+ reasoning: true,
9631
+ input: ["text", "image"],
9632
+ cost: {
9633
+ input: 3,
9634
+ output: 15,
9635
+ cacheRead: 0.3,
9636
+ cacheWrite: 0,
9637
+ },
9638
+ contextWindow: 1048576,
9639
+ maxTokens: 131072,
9640
+ },
9275
9641
  "mimo-v2.5": {
9276
9642
  id: "mimo-v2.5",
9277
9643
  name: "MiMo V2.5",
@@ -9298,9 +9664,9 @@ export const MODELS = {
9298
9664
  reasoning: true,
9299
9665
  input: ["text"],
9300
9666
  cost: {
9301
- input: 1.74,
9302
- output: 3.48,
9303
- cacheRead: 0.0145,
9667
+ input: 0.435,
9668
+ output: 0.87,
9669
+ cacheRead: 0.003625,
9304
9670
  cacheWrite: 0,
9305
9671
  },
9306
9672
  contextWindow: 1048576,
@@ -9737,6 +10103,40 @@ export const MODELS = {
9737
10103
  contextWindow: 1000000,
9738
10104
  maxTokens: 128000,
9739
10105
  },
10106
+ "anthropic/claude-opus-5": {
10107
+ id: "anthropic/claude-opus-5",
10108
+ name: "Claude Opus 5",
10109
+ api: "openai-completions",
10110
+ provider: "openrouter",
10111
+ baseUrl: "https://openrouter.ai/api/v1",
10112
+ reasoning: true,
10113
+ input: ["text", "image"],
10114
+ cost: {
10115
+ input: 5,
10116
+ output: 25,
10117
+ cacheRead: 0.5,
10118
+ cacheWrite: 6.25,
10119
+ },
10120
+ contextWindow: 1000000,
10121
+ maxTokens: 128000,
10122
+ },
10123
+ "anthropic/claude-opus-5-fast": {
10124
+ id: "anthropic/claude-opus-5-fast",
10125
+ name: "Claude Opus 5 (Fast)",
10126
+ api: "openai-completions",
10127
+ provider: "openrouter",
10128
+ baseUrl: "https://openrouter.ai/api/v1",
10129
+ reasoning: true,
10130
+ input: ["text", "image"],
10131
+ cost: {
10132
+ input: 10,
10133
+ output: 50,
10134
+ cacheRead: 1,
10135
+ cacheWrite: 12.5,
10136
+ },
10137
+ contextWindow: 1000000,
10138
+ maxTokens: 128000,
10139
+ },
9740
10140
  "anthropic/claude-sonnet-4": {
9741
10141
  id: "anthropic/claude-sonnet-4",
9742
10142
  name: "Anthropic: Claude Sonnet 4",
@@ -9814,13 +10214,13 @@ export const MODELS = {
9814
10214
  reasoning: true,
9815
10215
  input: ["text"],
9816
10216
  cost: {
9817
- input: 0.25,
9818
- output: 0.7999999999999999,
10217
+ input: 0.22,
10218
+ output: 0.85,
9819
10219
  cacheRead: 0.06,
9820
10220
  cacheWrite: 0,
9821
10221
  },
9822
10222
  contextWindow: 262144,
9823
- maxTokens: 80000,
10223
+ maxTokens: 262144,
9824
10224
  },
9825
10225
  "arcee-ai/virtuoso-large": {
9826
10226
  id: "arcee-ai/virtuoso-large",
@@ -9989,7 +10389,7 @@ export const MODELS = {
9989
10389
  cacheRead: 0,
9990
10390
  cacheWrite: 0,
9991
10391
  },
9992
- contextWindow: 131072,
10392
+ contextWindow: 163840,
9993
10393
  maxTokens: 16000,
9994
10394
  },
9995
10395
  "deepseek/deepseek-chat-v3-0324": {
@@ -10001,13 +10401,13 @@ export const MODELS = {
10001
10401
  reasoning: false,
10002
10402
  input: ["text"],
10003
10403
  cost: {
10004
- input: 0.24,
10005
- output: 0.8999999999999999,
10404
+ input: 0.27,
10405
+ output: 1.12,
10006
10406
  cacheRead: 0.135,
10007
10407
  cacheWrite: 0,
10008
10408
  },
10009
10409
  contextWindow: 163840,
10010
- maxTokens: 16384,
10410
+ maxTokens: 65536,
10011
10411
  },
10012
10412
  "deepseek/deepseek-chat-v3.1": {
10013
10413
  id: "deepseek/deepseek-chat-v3.1",
@@ -10018,8 +10418,8 @@ export const MODELS = {
10018
10418
  reasoning: true,
10019
10419
  input: ["text"],
10020
10420
  cost: {
10021
- input: 0.21,
10022
- output: 0.7899999999999999,
10421
+ input: 0.25,
10422
+ output: 0.95,
10023
10423
  cacheRead: 0.13,
10024
10424
  cacheWrite: 0,
10025
10425
  },
@@ -10070,8 +10470,8 @@ export const MODELS = {
10070
10470
  input: ["text"],
10071
10471
  cost: {
10072
10472
  input: 0.27,
10073
- output: 0.95,
10074
- cacheRead: 0.13,
10473
+ output: 1,
10474
+ cacheRead: 0.135,
10075
10475
  cacheWrite: 0,
10076
10476
  },
10077
10477
  contextWindow: 163840,
@@ -10086,13 +10486,13 @@ export const MODELS = {
10086
10486
  reasoning: true,
10087
10487
  input: ["text"],
10088
10488
  cost: {
10089
- input: 0.2145,
10090
- output: 0.32175,
10091
- cacheRead: 0.02145,
10489
+ input: 0.26899999999999996,
10490
+ output: 0.39999999999999997,
10491
+ cacheRead: 0.13449999999999998,
10092
10492
  cacheWrite: 0,
10093
10493
  },
10094
- contextWindow: 131072,
10095
- maxTokens: 64000,
10494
+ contextWindow: 163840,
10495
+ maxTokens: 65536,
10096
10496
  },
10097
10497
  "deepseek/deepseek-v3.2-exp": {
10098
10498
  id: "deepseek/deepseek-v3.2-exp",
@@ -10122,13 +10522,13 @@ export const MODELS = {
10122
10522
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
10123
10523
  input: ["text"],
10124
10524
  cost: {
10125
- input: 0.077,
10126
- output: 0.154,
10127
- cacheRead: 0.015399999999999999,
10525
+ input: 0.14,
10526
+ output: 0.28,
10527
+ cacheRead: 0.028,
10128
10528
  cacheWrite: 0,
10129
10529
  },
10130
10530
  contextWindow: 1048576,
10131
- maxTokens: 384000,
10531
+ maxTokens: 393216,
10132
10532
  },
10133
10533
  "deepseek/deepseek-v4-pro": {
10134
10534
  id: "deepseek/deepseek-v4-pro",
@@ -10265,7 +10665,7 @@ export const MODELS = {
10265
10665
  cacheRead: 0.19999999999999998,
10266
10666
  cacheWrite: 0.375,
10267
10667
  },
10268
- contextWindow: 65536,
10668
+ contextWindow: 131072,
10269
10669
  maxTokens: 32768,
10270
10670
  },
10271
10671
  "google/gemini-3.1-flash-lite": {
@@ -10333,7 +10733,7 @@ export const MODELS = {
10333
10733
  cacheRead: 0.19999999999999998,
10334
10734
  cacheWrite: 0.375,
10335
10735
  },
10336
- contextWindow: 1048756,
10736
+ contextWindow: 1048576,
10337
10737
  maxTokens: 65536,
10338
10738
  },
10339
10739
  "google/gemini-3.5-flash": {
@@ -10353,6 +10753,40 @@ export const MODELS = {
10353
10753
  contextWindow: 1048576,
10354
10754
  maxTokens: 65536,
10355
10755
  },
10756
+ "google/gemini-3.5-flash-lite": {
10757
+ id: "google/gemini-3.5-flash-lite",
10758
+ name: "Google: Gemini 3.5 Flash Lite",
10759
+ api: "openai-completions",
10760
+ provider: "openrouter",
10761
+ baseUrl: "https://openrouter.ai/api/v1",
10762
+ reasoning: true,
10763
+ input: ["text", "image"],
10764
+ cost: {
10765
+ input: 0.3,
10766
+ output: 2.5,
10767
+ cacheRead: 0.03,
10768
+ cacheWrite: 0.08333333333333334,
10769
+ },
10770
+ contextWindow: 1048576,
10771
+ maxTokens: 65536,
10772
+ },
10773
+ "google/gemini-3.6-flash": {
10774
+ id: "google/gemini-3.6-flash",
10775
+ name: "Google: Gemini 3.6 Flash",
10776
+ api: "openai-completions",
10777
+ provider: "openrouter",
10778
+ baseUrl: "https://openrouter.ai/api/v1",
10779
+ reasoning: true,
10780
+ input: ["text", "image"],
10781
+ cost: {
10782
+ input: 1.5,
10783
+ output: 7.5,
10784
+ cacheRead: 0.15,
10785
+ cacheWrite: 0.08333333333333334,
10786
+ },
10787
+ contextWindow: 1048576,
10788
+ maxTokens: 65536,
10789
+ },
10356
10790
  "google/gemma-3-12b-it": {
10357
10791
  id: "google/gemma-3-12b-it",
10358
10792
  name: "Google: Gemma 3 12B",
@@ -10380,12 +10814,12 @@ export const MODELS = {
10380
10814
  input: ["text", "image"],
10381
10815
  cost: {
10382
10816
  input: 0.08,
10383
- output: 0.16,
10384
- cacheRead: 0,
10817
+ output: 0.44999999999999996,
10818
+ cacheRead: 0.04,
10385
10819
  cacheWrite: 0,
10386
10820
  },
10387
- contextWindow: 131072,
10388
- maxTokens: 16384,
10821
+ contextWindow: 262144,
10822
+ maxTokens: 131072,
10389
10823
  },
10390
10824
  "google/gemma-4-26b-a4b-it": {
10391
10825
  id: "google/gemma-4-26b-a4b-it",
@@ -10396,13 +10830,13 @@ export const MODELS = {
10396
10830
  reasoning: true,
10397
10831
  input: ["text", "image"],
10398
10832
  cost: {
10399
- input: 0.06,
10400
- output: 0.33,
10401
- cacheRead: 0,
10833
+ input: 0.12,
10834
+ output: 0.35,
10835
+ cacheRead: 0.049999999999999996,
10402
10836
  cacheWrite: 0,
10403
10837
  },
10404
10838
  contextWindow: 262144,
10405
- maxTokens: 4096,
10839
+ maxTokens: 262144,
10406
10840
  },
10407
10841
  "google/gemma-4-26b-a4b-it:free": {
10408
10842
  id: "google/gemma-4-26b-a4b-it:free",
@@ -10430,9 +10864,9 @@ export const MODELS = {
10430
10864
  reasoning: true,
10431
10865
  input: ["text", "image"],
10432
10866
  cost: {
10433
- input: 0.12,
10434
- output: 0.35,
10435
- cacheRead: 0.09,
10867
+ input: 0.14,
10868
+ output: 0.39999999999999997,
10869
+ cacheRead: 0,
10436
10870
  cacheWrite: 0,
10437
10871
  },
10438
10872
  contextWindow: 262144,
@@ -10453,7 +10887,7 @@ export const MODELS = {
10453
10887
  cacheWrite: 0,
10454
10888
  },
10455
10889
  contextWindow: 262144,
10456
- maxTokens: 8192,
10890
+ maxTokens: 32768,
10457
10891
  },
10458
10892
  "ibm-granite/granite-4.1-8b": {
10459
10893
  id: "ibm-granite/granite-4.1-8b",
@@ -10524,6 +10958,23 @@ export const MODELS = {
10524
10958
  contextWindow: 262144,
10525
10959
  maxTokens: 32768,
10526
10960
  },
10961
+ "inclusionai/ling-3.0-flash:free": {
10962
+ id: "inclusionai/ling-3.0-flash:free",
10963
+ name: "Ling-3.0-flash (free)",
10964
+ api: "openai-completions",
10965
+ provider: "openrouter",
10966
+ baseUrl: "https://openrouter.ai/api/v1",
10967
+ reasoning: true,
10968
+ input: ["text"],
10969
+ cost: {
10970
+ input: 0,
10971
+ output: 0,
10972
+ cacheRead: 0,
10973
+ cacheWrite: 0,
10974
+ },
10975
+ contextWindow: 262144,
10976
+ maxTokens: 32768,
10977
+ },
10527
10978
  "inclusionai/ring-2.6-1t": {
10528
10979
  id: "inclusionai/ring-2.6-1t",
10529
10980
  name: "inclusionAI: Ring-2.6-1T",
@@ -10541,6 +10992,23 @@ export const MODELS = {
10541
10992
  contextWindow: 262144,
10542
10993
  maxTokens: 65536,
10543
10994
  },
10995
+ "kwaipilot/kat-coder-air-v2.5": {
10996
+ id: "kwaipilot/kat-coder-air-v2.5",
10997
+ name: "Kwaipilot: KAT-Coder-Air V2.5",
10998
+ api: "openai-completions",
10999
+ provider: "openrouter",
11000
+ baseUrl: "https://openrouter.ai/api/v1",
11001
+ reasoning: false,
11002
+ input: ["text"],
11003
+ cost: {
11004
+ input: 0.15,
11005
+ output: 0.6,
11006
+ cacheRead: 0.03,
11007
+ cacheWrite: 0,
11008
+ },
11009
+ contextWindow: 256000,
11010
+ maxTokens: 80000,
11011
+ },
10544
11012
  "kwaipilot/kat-coder-pro-v2": {
10545
11013
  id: "kwaipilot/kat-coder-pro-v2",
10546
11014
  name: "Kwaipilot: KAT-Coder-Pro V2",
@@ -10555,25 +11023,42 @@ export const MODELS = {
10555
11023
  cacheRead: 0.06,
10556
11024
  cacheWrite: 0,
10557
11025
  },
11026
+ contextWindow: 262144,
11027
+ maxTokens: 80000,
11028
+ },
11029
+ "kwaipilot/kat-coder-pro-v2.5": {
11030
+ id: "kwaipilot/kat-coder-pro-v2.5",
11031
+ name: "Kwaipilot: KAT-Coder-Pro V2.5",
11032
+ api: "openai-completions",
11033
+ provider: "openrouter",
11034
+ baseUrl: "https://openrouter.ai/api/v1",
11035
+ reasoning: false,
11036
+ input: ["text"],
11037
+ cost: {
11038
+ input: 0.74,
11039
+ output: 2.96,
11040
+ cacheRead: 0.15,
11041
+ cacheWrite: 0,
11042
+ },
10558
11043
  contextWindow: 256000,
10559
11044
  maxTokens: 80000,
10560
11045
  },
10561
- "liquid/lfm-2.5-1.2b-thinking:free": {
10562
- id: "liquid/lfm-2.5-1.2b-thinking:free",
10563
- name: "LiquidAI: LFM2.5-1.2B-Thinking (free)",
11046
+ "meituan/longcat-2.0": {
11047
+ id: "meituan/longcat-2.0",
11048
+ name: "Meituan: LongCat 2.0",
10564
11049
  api: "openai-completions",
10565
11050
  provider: "openrouter",
10566
11051
  baseUrl: "https://openrouter.ai/api/v1",
10567
11052
  reasoning: true,
10568
11053
  input: ["text"],
10569
11054
  cost: {
10570
- input: 0,
10571
- output: 0,
10572
- cacheRead: 0,
11055
+ input: 0.3,
11056
+ output: 1.2,
11057
+ cacheRead: 0.006,
10573
11058
  cacheWrite: 0,
10574
11059
  },
10575
- contextWindow: 32768,
10576
- maxTokens: 4096,
11060
+ contextWindow: 1048756,
11061
+ maxTokens: 262144,
10577
11062
  },
10578
11063
  "meta-llama/llama-3.1-70b-instruct": {
10579
11064
  id: "meta-llama/llama-3.1-70b-instruct",
@@ -10601,13 +11086,13 @@ export const MODELS = {
10601
11086
  reasoning: false,
10602
11087
  input: ["text"],
10603
11088
  cost: {
10604
- input: 0.02,
10605
- output: 0.03,
10606
- cacheRead: 0,
11089
+ input: 0.049999999999999996,
11090
+ output: 0.08,
11091
+ cacheRead: 0.024999999999999998,
10607
11092
  cacheWrite: 0,
10608
11093
  },
10609
11094
  contextWindow: 131072,
10610
- maxTokens: 16384,
11095
+ maxTokens: 131072,
10611
11096
  },
10612
11097
  "meta-llama/llama-3.3-70b-instruct": {
10613
11098
  id: "meta-llama/llama-3.3-70b-instruct",
@@ -10618,30 +11103,13 @@ export const MODELS = {
10618
11103
  reasoning: false,
10619
11104
  input: ["text"],
10620
11105
  cost: {
10621
- input: 0.09999999999999999,
10622
- output: 0.32,
10623
- cacheRead: 0,
10624
- cacheWrite: 0,
10625
- },
10626
- contextWindow: 131072,
10627
- maxTokens: 16384,
10628
- },
10629
- "meta-llama/llama-3.3-70b-instruct:free": {
10630
- id: "meta-llama/llama-3.3-70b-instruct:free",
10631
- name: "Meta: Llama 3.3 70B Instruct (free)",
10632
- api: "openai-completions",
10633
- provider: "openrouter",
10634
- baseUrl: "https://openrouter.ai/api/v1",
10635
- reasoning: false,
10636
- input: ["text"],
10637
- cost: {
10638
- input: 0,
10639
- output: 0,
11106
+ input: 0.13,
11107
+ output: 0.39999999999999997,
10640
11108
  cacheRead: 0,
10641
11109
  cacheWrite: 0,
10642
11110
  },
10643
11111
  contextWindow: 131072,
10644
- maxTokens: 4096,
11112
+ maxTokens: 128000,
10645
11113
  },
10646
11114
  "meta-llama/llama-4-maverick": {
10647
11115
  id: "meta-llama/llama-4-maverick",
@@ -10674,9 +11142,26 @@ export const MODELS = {
10674
11142
  cacheRead: 0,
10675
11143
  cacheWrite: 0,
10676
11144
  },
10677
- contextWindow: 10000000,
11145
+ contextWindow: 1310720,
10678
11146
  maxTokens: 16384,
10679
11147
  },
11148
+ "meta/muse-spark-1.1": {
11149
+ id: "meta/muse-spark-1.1",
11150
+ name: "Meta: Muse Spark 1.1",
11151
+ api: "openai-completions",
11152
+ provider: "openrouter",
11153
+ baseUrl: "https://openrouter.ai/api/v1",
11154
+ reasoning: true,
11155
+ input: ["text", "image"],
11156
+ cost: {
11157
+ input: 1.25,
11158
+ output: 4.25,
11159
+ cacheRead: 0.15,
11160
+ cacheWrite: 0,
11161
+ },
11162
+ contextWindow: 1048576,
11163
+ maxTokens: 4096,
11164
+ },
10680
11165
  "minimax/minimax-m1": {
10681
11166
  id: "minimax/minimax-m1",
10682
11167
  name: "MiniMax: MiniMax M1",
@@ -10686,7 +11171,7 @@ export const MODELS = {
10686
11171
  reasoning: true,
10687
11172
  input: ["text"],
10688
11173
  cost: {
10689
- input: 0.39999999999999997,
11174
+ input: 0.55,
10690
11175
  output: 2.2,
10691
11176
  cacheRead: 0,
10692
11177
  cacheWrite: 0,
@@ -10754,13 +11239,13 @@ export const MODELS = {
10754
11239
  reasoning: true,
10755
11240
  input: ["text"],
10756
11241
  cost: {
10757
- input: 0.24,
10758
- output: 0.96,
10759
- cacheRead: 0,
11242
+ input: 0.25,
11243
+ output: 1,
11244
+ cacheRead: 0.049999999999999996,
10760
11245
  cacheWrite: 0,
10761
11246
  },
10762
11247
  contextWindow: 204800,
10763
- maxTokens: 196608,
11248
+ maxTokens: 131072,
10764
11249
  },
10765
11250
  "minimax/minimax-m3": {
10766
11251
  id: "minimax/minimax-m3",
@@ -10777,7 +11262,7 @@ export const MODELS = {
10777
11262
  cacheWrite: 0,
10778
11263
  },
10779
11264
  contextWindow: 1048576,
10780
- maxTokens: 131072,
11265
+ maxTokens: 512000,
10781
11266
  },
10782
11267
  "mistralai/codestral-2508": {
10783
11268
  id: "mistralai/codestral-2508",
@@ -10975,13 +11460,13 @@ export const MODELS = {
10975
11460
  reasoning: false,
10976
11461
  input: ["text"],
10977
11462
  cost: {
10978
- input: 0.02,
11463
+ input: 0.019000000000000003,
10979
11464
  output: 0.03,
10980
11465
  cacheRead: 0,
10981
11466
  cacheWrite: 0,
10982
11467
  },
10983
11468
  contextWindow: 131072,
10984
- maxTokens: 4096,
11469
+ maxTokens: 16384,
10985
11470
  },
10986
11471
  "mistralai/mistral-saba": {
10987
11472
  id: "mistralai/mistral-saba",
@@ -11026,13 +11511,13 @@ export const MODELS = {
11026
11511
  reasoning: false,
11027
11512
  input: ["text", "image"],
11028
11513
  cost: {
11029
- input: 0.075,
11030
- output: 0.19999999999999998,
11031
- cacheRead: 0,
11514
+ input: 0.09999999999999999,
11515
+ output: 0.3,
11516
+ cacheRead: 0.01,
11032
11517
  cacheWrite: 0,
11033
11518
  },
11034
- contextWindow: 128000,
11035
- maxTokens: 16384,
11519
+ contextWindow: 256000,
11520
+ maxTokens: 4096,
11036
11521
  },
11037
11522
  "mistralai/mixtral-8x22b-instruct": {
11038
11523
  id: "mistralai/mixtral-8x22b-instruct",
@@ -11146,9 +11631,9 @@ export const MODELS = {
11146
11631
  reasoning: true,
11147
11632
  input: ["text", "image"],
11148
11633
  cost: {
11149
- input: 0.66,
11150
- output: 3.41,
11151
- cacheRead: 0.15,
11634
+ input: 0.646,
11635
+ output: 2.7199999999999998,
11636
+ cacheRead: 0.1088,
11152
11637
  cacheWrite: 0,
11153
11638
  },
11154
11639
  contextWindow: 262144,
@@ -11163,14 +11648,31 @@ export const MODELS = {
11163
11648
  reasoning: true,
11164
11649
  input: ["text", "image"],
11165
11650
  cost: {
11166
- input: 0.719,
11167
- output: 3.49,
11168
- cacheRead: 0.149,
11651
+ input: 0.78,
11652
+ output: 3.5,
11653
+ cacheRead: 0.15,
11169
11654
  cacheWrite: 0,
11170
11655
  },
11171
11656
  contextWindow: 262144,
11172
11657
  maxTokens: 262144,
11173
11658
  },
11659
+ "moonshotai/kimi-k3": {
11660
+ id: "moonshotai/kimi-k3",
11661
+ name: "MoonshotAI: Kimi K3",
11662
+ api: "openai-completions",
11663
+ provider: "openrouter",
11664
+ baseUrl: "https://openrouter.ai/api/v1",
11665
+ reasoning: true,
11666
+ input: ["text", "image"],
11667
+ cost: {
11668
+ input: 3,
11669
+ output: 15,
11670
+ cacheRead: 0.3,
11671
+ cacheWrite: 0,
11672
+ },
11673
+ contextWindow: 1048576,
11674
+ maxTokens: 4096,
11675
+ },
11174
11676
  "nex-agi/nex-n2-mini": {
11175
11677
  id: "nex-agi/nex-n2-mini",
11176
11678
  name: "Nex AGI: Nex-N2-Mini",
@@ -11205,23 +11707,6 @@ export const MODELS = {
11205
11707
  contextWindow: 262144,
11206
11708
  maxTokens: 262144,
11207
11709
  },
11208
- "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
11209
- id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
11210
- name: "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5",
11211
- api: "openai-completions",
11212
- provider: "openrouter",
11213
- baseUrl: "https://openrouter.ai/api/v1",
11214
- reasoning: true,
11215
- input: ["text"],
11216
- cost: {
11217
- input: 0.39999999999999997,
11218
- output: 0.39999999999999997,
11219
- cacheRead: 0,
11220
- cacheWrite: 0,
11221
- },
11222
- contextWindow: 131072,
11223
- maxTokens: 16384,
11224
- },
11225
11710
  "nvidia/nemotron-3-nano-30b-a3b": {
11226
11711
  id: "nvidia/nemotron-3-nano-30b-a3b",
11227
11712
  name: "NVIDIA: Nemotron 3 Nano 30B A3B",
@@ -11282,13 +11767,13 @@ export const MODELS = {
11282
11767
  reasoning: true,
11283
11768
  input: ["text"],
11284
11769
  cost: {
11285
- input: 0.08,
11286
- output: 0.44999999999999996,
11770
+ input: 0.08499999999999999,
11771
+ output: 0.39999999999999997,
11287
11772
  cacheRead: 0,
11288
11773
  cacheWrite: 0,
11289
11774
  },
11290
11775
  contextWindow: 1000000,
11291
- maxTokens: 4096,
11776
+ maxTokens: 16384,
11292
11777
  },
11293
11778
  "nvidia/nemotron-3-super-120b-a12b:free": {
11294
11779
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -11304,7 +11789,7 @@ export const MODELS = {
11304
11789
  cacheRead: 0,
11305
11790
  cacheWrite: 0,
11306
11791
  },
11307
- contextWindow: 1000000,
11792
+ contextWindow: 262144,
11308
11793
  maxTokens: 262144,
11309
11794
  },
11310
11795
  "nvidia/nemotron-3-ultra-550b-a55b": {
@@ -11321,7 +11806,7 @@ export const MODELS = {
11321
11806
  cacheRead: 0.09999999999999999,
11322
11807
  cacheWrite: 0,
11323
11808
  },
11324
- contextWindow: 1000000,
11809
+ contextWindow: 512288,
11325
11810
  maxTokens: 16384,
11326
11811
  },
11327
11812
  "nvidia/nemotron-3-ultra-550b-a55b:free": {
@@ -11492,7 +11977,7 @@ export const MODELS = {
11492
11977
  cacheWrite: 0,
11493
11978
  },
11494
11979
  contextWindow: 1047576,
11495
- maxTokens: 4096,
11980
+ maxTokens: 32768,
11496
11981
  },
11497
11982
  "openai/gpt-4.1-mini": {
11498
11983
  id: "openai/gpt-4.1-mini",
@@ -11539,7 +12024,7 @@ export const MODELS = {
11539
12024
  cost: {
11540
12025
  input: 2.5,
11541
12026
  output: 10,
11542
- cacheRead: 0,
12027
+ cacheRead: 1.25,
11543
12028
  cacheWrite: 0,
11544
12029
  },
11545
12030
  contextWindow: 128000,
@@ -11692,11 +12177,11 @@ export const MODELS = {
11692
12177
  cost: {
11693
12178
  input: 0.049999999999999996,
11694
12179
  output: 0.39999999999999997,
11695
- cacheRead: 0.01,
12180
+ cacheRead: 0.005,
11696
12181
  cacheWrite: 0,
11697
12182
  },
11698
12183
  contextWindow: 400000,
11699
- maxTokens: 4096,
12184
+ maxTokens: 128000,
11700
12185
  },
11701
12186
  "openai/gpt-5-pro": {
11702
12187
  id: "openai/gpt-5-pro",
@@ -11726,7 +12211,7 @@ export const MODELS = {
11726
12211
  cost: {
11727
12212
  input: 1.25,
11728
12213
  output: 10,
11729
- cacheRead: 0.13,
12214
+ cacheRead: 0.125,
11730
12215
  cacheWrite: 0,
11731
12216
  },
11732
12217
  contextWindow: 400000,
@@ -11743,11 +12228,11 @@ export const MODELS = {
11743
12228
  cost: {
11744
12229
  input: 1.25,
11745
12230
  output: 10,
11746
- cacheRead: 0.13,
12231
+ cacheRead: 0.125,
11747
12232
  cacheWrite: 0,
11748
12233
  },
11749
12234
  contextWindow: 128000,
11750
- maxTokens: 32000,
12235
+ maxTokens: 16384,
11751
12236
  },
11752
12237
  "openai/gpt-5.1-codex": {
11753
12238
  id: "openai/gpt-5.1-codex",
@@ -11794,11 +12279,11 @@ export const MODELS = {
11794
12279
  cost: {
11795
12280
  input: 0.25,
11796
12281
  output: 2,
11797
- cacheRead: 0.024999999999999998,
12282
+ cacheRead: 0.03,
11798
12283
  cacheWrite: 0,
11799
12284
  },
11800
12285
  contextWindow: 400000,
11801
- maxTokens: 100000,
12286
+ maxTokens: 128000,
11802
12287
  },
11803
12288
  "openai/gpt-5.2": {
11804
12289
  id: "openai/gpt-5.2",
@@ -12184,25 +12669,8 @@ export const MODELS = {
12184
12669
  reasoning: true,
12185
12670
  input: ["text"],
12186
12671
  cost: {
12187
- input: 0.036,
12188
- output: 0.18,
12189
- cacheRead: 0,
12190
- cacheWrite: 0,
12191
- },
12192
- contextWindow: 131072,
12193
- maxTokens: 4096,
12194
- },
12195
- "openai/gpt-oss-120b:free": {
12196
- id: "openai/gpt-oss-120b:free",
12197
- name: "OpenAI: gpt-oss-120b (free)",
12198
- api: "openai-completions",
12199
- provider: "openrouter",
12200
- baseUrl: "https://openrouter.ai/api/v1",
12201
- reasoning: true,
12202
- input: ["text"],
12203
- cost: {
12204
- input: 0,
12205
- output: 0,
12672
+ input: 0.037,
12673
+ output: 0.16999999999999998,
12206
12674
  cacheRead: 0,
12207
12675
  cacheWrite: 0,
12208
12676
  },
@@ -12218,13 +12686,13 @@ export const MODELS = {
12218
12686
  reasoning: true,
12219
12687
  input: ["text"],
12220
12688
  cost: {
12221
- input: 0.029,
12689
+ input: 0.03,
12222
12690
  output: 0.14,
12223
12691
  cacheRead: 0,
12224
12692
  cacheWrite: 0,
12225
12693
  },
12226
12694
  contextWindow: 131072,
12227
- maxTokens: 4096,
12695
+ maxTokens: 131072,
12228
12696
  },
12229
12697
  "openai/gpt-oss-20b:free": {
12230
12698
  id: "openai/gpt-oss-20b:free",
@@ -12430,6 +12898,23 @@ export const MODELS = {
12430
12898
  contextWindow: 2000000,
12431
12899
  maxTokens: 4096,
12432
12900
  },
12901
+ "openrouter/auto-beta": {
12902
+ id: "openrouter/auto-beta",
12903
+ name: "Auto Router (Beta)",
12904
+ api: "openai-completions",
12905
+ provider: "openrouter",
12906
+ baseUrl: "https://openrouter.ai/api/v1",
12907
+ reasoning: true,
12908
+ input: ["text", "image"],
12909
+ cost: {
12910
+ input: -1000000,
12911
+ output: -1000000,
12912
+ cacheRead: 0,
12913
+ cacheWrite: 0,
12914
+ },
12915
+ contextWindow: 2000000,
12916
+ maxTokens: 4096,
12917
+ },
12433
12918
  "openrouter/free": {
12434
12919
  id: "openrouter/free",
12435
12920
  name: "Free Models Router",
@@ -12481,6 +12966,40 @@ export const MODELS = {
12481
12966
  contextWindow: 262144,
12482
12967
  maxTokens: 32768,
12483
12968
  },
12969
+ "poolside/laguna-s-2.1": {
12970
+ id: "poolside/laguna-s-2.1",
12971
+ name: "Poolside: Laguna S 2.1",
12972
+ api: "openai-completions",
12973
+ provider: "openrouter",
12974
+ baseUrl: "https://openrouter.ai/api/v1",
12975
+ reasoning: true,
12976
+ input: ["text"],
12977
+ cost: {
12978
+ input: 0.09999999999999999,
12979
+ output: 0.19999999999999998,
12980
+ cacheRead: 0.01,
12981
+ cacheWrite: 0,
12982
+ },
12983
+ contextWindow: 1048576,
12984
+ maxTokens: 131072,
12985
+ },
12986
+ "poolside/laguna-s-2.1:free": {
12987
+ id: "poolside/laguna-s-2.1:free",
12988
+ name: "Poolside: Laguna S 2.1 (free)",
12989
+ api: "openai-completions",
12990
+ provider: "openrouter",
12991
+ baseUrl: "https://openrouter.ai/api/v1",
12992
+ reasoning: true,
12993
+ input: ["text"],
12994
+ cost: {
12995
+ input: 0,
12996
+ output: 0,
12997
+ cacheRead: 0,
12998
+ cacheWrite: 0,
12999
+ },
13000
+ contextWindow: 262144,
13001
+ maxTokens: 32768,
13002
+ },
12484
13003
  "poolside/laguna-xs-2.1": {
12485
13004
  id: "poolside/laguna-xs-2.1",
12486
13005
  name: "Poolside: Laguna XS 2.1",
@@ -12529,7 +13048,7 @@ export const MODELS = {
12529
13048
  cacheRead: 0,
12530
13049
  cacheWrite: 0,
12531
13050
  },
12532
- contextWindow: 131072,
13051
+ contextWindow: 32768,
12533
13052
  maxTokens: 16384,
12534
13053
  },
12535
13054
  "qwen/qwen-2.5-7b-instruct": {
@@ -12546,7 +13065,7 @@ export const MODELS = {
12546
13065
  cacheRead: 0,
12547
13066
  cacheWrite: 0,
12548
13067
  },
12549
- contextWindow: 131072,
13068
+ contextWindow: 32768,
12550
13069
  maxTokens: 32768,
12551
13070
  },
12552
13071
  "qwen/qwen-plus": {
@@ -12609,13 +13128,13 @@ export const MODELS = {
12609
13128
  reasoning: true,
12610
13129
  input: ["text"],
12611
13130
  cost: {
12612
- input: 0.09999999999999999,
12613
- output: 0.24,
13131
+ input: 0.22749999999999998,
13132
+ output: 0.9099999999999999,
12614
13133
  cacheRead: 0,
12615
13134
  cacheWrite: 0,
12616
13135
  },
12617
- contextWindow: 131702,
12618
- maxTokens: 40960,
13136
+ contextWindow: 131072,
13137
+ maxTokens: 8192,
12619
13138
  },
12620
13139
  "qwen/qwen3-235b-a22b": {
12621
13140
  id: "qwen/qwen3-235b-a22b",
@@ -12660,13 +13179,13 @@ export const MODELS = {
12660
13179
  reasoning: true,
12661
13180
  input: ["text"],
12662
13181
  cost: {
12663
- input: 0.14950000000000002,
12664
- output: 1.495,
13182
+ input: 0.3,
13183
+ output: 3,
12665
13184
  cacheRead: 0,
12666
13185
  cacheWrite: 0,
12667
13186
  },
12668
13187
  contextWindow: 262144,
12669
- maxTokens: 4096,
13188
+ maxTokens: 32768,
12670
13189
  },
12671
13190
  "qwen/qwen3-30b-a3b": {
12672
13191
  id: "qwen/qwen3-30b-a3b",
@@ -12699,7 +13218,7 @@ export const MODELS = {
12699
13218
  cacheRead: 0,
12700
13219
  cacheWrite: 0,
12701
13220
  },
12702
- contextWindow: 131072,
13221
+ contextWindow: 262144,
12703
13222
  maxTokens: 32000,
12704
13223
  },
12705
13224
  "qwen/qwen3-30b-a3b-thinking-2507": {
@@ -12716,7 +13235,7 @@ export const MODELS = {
12716
13235
  cacheRead: 0,
12717
13236
  cacheWrite: 0,
12718
13237
  },
12719
- contextWindow: 131072,
13238
+ contextWindow: 81920,
12720
13239
  maxTokens: 32768,
12721
13240
  },
12722
13241
  "qwen/qwen3-32b": {
@@ -12762,12 +13281,12 @@ export const MODELS = {
12762
13281
  reasoning: false,
12763
13282
  input: ["text"],
12764
13283
  cost: {
12765
- input: 0.22,
12766
- output: 1.7999999999999998,
12767
- cacheRead: 0,
13284
+ input: 0.3,
13285
+ output: 1,
13286
+ cacheRead: 0.09999999999999999,
12768
13287
  cacheWrite: 0,
12769
13288
  },
12770
- contextWindow: 1048576,
13289
+ contextWindow: 262144,
12771
13290
  maxTokens: 65536,
12772
13291
  },
12773
13292
  "qwen/qwen3-coder-30b-a3b-instruct": {
@@ -12784,7 +13303,7 @@ export const MODELS = {
12784
13303
  cacheRead: 0,
12785
13304
  cacheWrite: 0,
12786
13305
  },
12787
- contextWindow: 160000,
13306
+ contextWindow: 262144,
12788
13307
  maxTokens: 32768,
12789
13308
  },
12790
13309
  "qwen/qwen3-coder-flash": {
@@ -12838,23 +13357,6 @@ export const MODELS = {
12838
13357
  contextWindow: 1000000,
12839
13358
  maxTokens: 65536,
12840
13359
  },
12841
- "qwen/qwen3-coder:free": {
12842
- id: "qwen/qwen3-coder:free",
12843
- name: "Qwen: Qwen3 Coder 480B A35B (free)",
12844
- api: "openai-completions",
12845
- provider: "openrouter",
12846
- baseUrl: "https://openrouter.ai/api/v1",
12847
- reasoning: false,
12848
- input: ["text"],
12849
- cost: {
12850
- input: 0,
12851
- output: 0,
12852
- cacheRead: 0,
12853
- cacheWrite: 0,
12854
- },
12855
- contextWindow: 1048576,
12856
- maxTokens: 262000,
12857
- },
12858
13360
  "qwen/qwen3-max": {
12859
13361
  id: "qwen/qwen3-max",
12860
13362
  name: "Qwen: Qwen3 Max",
@@ -12898,30 +13400,13 @@ export const MODELS = {
12898
13400
  reasoning: false,
12899
13401
  input: ["text"],
12900
13402
  cost: {
12901
- input: 0.09,
13403
+ input: 0.09999999999999999,
12902
13404
  output: 1.1,
12903
- cacheRead: 0,
12904
- cacheWrite: 0,
12905
- },
12906
- contextWindow: 262144,
12907
- maxTokens: 16384,
12908
- },
12909
- "qwen/qwen3-next-80b-a3b-instruct:free": {
12910
- id: "qwen/qwen3-next-80b-a3b-instruct:free",
12911
- name: "Qwen: Qwen3 Next 80B A3B Instruct (free)",
12912
- api: "openai-completions",
12913
- provider: "openrouter",
12914
- baseUrl: "https://openrouter.ai/api/v1",
12915
- reasoning: false,
12916
- input: ["text"],
12917
- cost: {
12918
- input: 0,
12919
- output: 0,
12920
- cacheRead: 0,
13405
+ cacheRead: 0.07,
12921
13406
  cacheWrite: 0,
12922
13407
  },
12923
13408
  contextWindow: 262144,
12924
- maxTokens: 4096,
13409
+ maxTokens: 262144,
12925
13410
  },
12926
13411
  "qwen/qwen3-next-80b-a3b-thinking": {
12927
13412
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -12949,13 +13434,13 @@ export const MODELS = {
12949
13434
  reasoning: false,
12950
13435
  input: ["text", "image"],
12951
13436
  cost: {
12952
- input: 0.19999999999999998,
12953
- output: 0.88,
12954
- cacheRead: 0.11,
13437
+ input: 0.21,
13438
+ output: 1.9,
13439
+ cacheRead: 0.09999999999999999,
12955
13440
  cacheWrite: 0,
12956
13441
  },
12957
13442
  contextWindow: 262144,
12958
- maxTokens: 16384,
13443
+ maxTokens: 32768,
12959
13444
  },
12960
13445
  "qwen/qwen3-vl-235b-a22b-thinking": {
12961
13446
  id: "qwen/qwen3-vl-235b-a22b-thinking",
@@ -12983,13 +13468,13 @@ export const MODELS = {
12983
13468
  reasoning: false,
12984
13469
  input: ["text", "image"],
12985
13470
  cost: {
12986
- input: 0.13,
12987
- output: 0.52,
13471
+ input: 0.15,
13472
+ output: 0.6,
12988
13473
  cacheRead: 0,
12989
13474
  cacheWrite: 0,
12990
13475
  },
12991
13476
  contextWindow: 262144,
12992
- maxTokens: 32768,
13477
+ maxTokens: 16384,
12993
13478
  },
12994
13479
  "qwen/qwen3-vl-30b-a3b-thinking": {
12995
13480
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -13005,7 +13490,7 @@ export const MODELS = {
13005
13490
  cacheRead: 0,
13006
13491
  cacheWrite: 0,
13007
13492
  },
13008
- contextWindow: 131072,
13493
+ contextWindow: 262144,
13009
13494
  maxTokens: 32768,
13010
13495
  },
13011
13496
  "qwen/qwen3-vl-32b-instruct": {
@@ -13022,7 +13507,7 @@ export const MODELS = {
13022
13507
  cacheRead: 0,
13023
13508
  cacheWrite: 0,
13024
13509
  },
13025
- contextWindow: 262144,
13510
+ contextWindow: 131072,
13026
13511
  maxTokens: 32768,
13027
13512
  },
13028
13513
  "qwen/qwen3-vl-8b-instruct": {
@@ -13039,7 +13524,7 @@ export const MODELS = {
13039
13524
  cacheRead: 0,
13040
13525
  cacheWrite: 0,
13041
13526
  },
13042
- contextWindow: 256000,
13527
+ contextWindow: 262144,
13043
13528
  maxTokens: 32768,
13044
13529
  },
13045
13530
  "qwen/qwen3-vl-8b-thinking": {
@@ -13056,7 +13541,7 @@ export const MODELS = {
13056
13541
  cacheRead: 0,
13057
13542
  cacheWrite: 0,
13058
13543
  },
13059
- contextWindow: 256000,
13544
+ contextWindow: 131072,
13060
13545
  maxTokens: 32768,
13061
13546
  },
13062
13547
  "qwen/qwen3.5-122b-a10b": {
@@ -13074,7 +13559,7 @@ export const MODELS = {
13074
13559
  cacheWrite: 0,
13075
13560
  },
13076
13561
  contextWindow: 262144,
13077
- maxTokens: 262144,
13562
+ maxTokens: 65536,
13078
13563
  },
13079
13564
  "qwen/qwen3.5-27b": {
13080
13565
  id: "qwen/qwen3.5-27b",
@@ -13104,11 +13589,11 @@ export const MODELS = {
13104
13589
  cost: {
13105
13590
  input: 0.14,
13106
13591
  output: 1,
13107
- cacheRead: 0.049999999999999996,
13592
+ cacheRead: 0,
13108
13593
  cacheWrite: 0,
13109
13594
  },
13110
13595
  contextWindow: 262144,
13111
- maxTokens: 81920,
13596
+ maxTokens: 262144,
13112
13597
  },
13113
13598
  "qwen/qwen3.5-397b-a17b": {
13114
13599
  id: "qwen/qwen3.5-397b-a17b",
@@ -13119,13 +13604,13 @@ export const MODELS = {
13119
13604
  reasoning: true,
13120
13605
  input: ["text", "image"],
13121
13606
  cost: {
13122
- input: 0.385,
13123
- output: 2.4499999999999997,
13124
- cacheRead: 0.111,
13607
+ input: 0.39,
13608
+ output: 2.34,
13609
+ cacheRead: 0,
13125
13610
  cacheWrite: 0,
13126
13611
  },
13127
- contextWindow: 256000,
13128
- maxTokens: 4096,
13612
+ contextWindow: 262144,
13613
+ maxTokens: 65536,
13129
13614
  },
13130
13615
  "qwen/qwen3.5-9b": {
13131
13616
  id: "qwen/qwen3.5-9b",
@@ -13204,13 +13689,13 @@ export const MODELS = {
13204
13689
  reasoning: true,
13205
13690
  input: ["text", "image"],
13206
13691
  cost: {
13207
- input: 0.28500000000000003,
13208
- output: 2.4,
13692
+ input: 0.3,
13693
+ output: 2,
13209
13694
  cacheRead: 0.15,
13210
13695
  cacheWrite: 0,
13211
13696
  },
13212
13697
  contextWindow: 262144,
13213
- maxTokens: 262140,
13698
+ maxTokens: 65536,
13214
13699
  },
13215
13700
  "qwen/qwen3.6-35b-a3b": {
13216
13701
  id: "qwen/qwen3.6-35b-a3b",
@@ -13289,10 +13774,10 @@ export const MODELS = {
13289
13774
  reasoning: true,
13290
13775
  input: ["text"],
13291
13776
  cost: {
13292
- input: 1.25,
13293
- output: 3.75,
13294
- cacheRead: 0.25,
13295
- cacheWrite: 1.5625,
13777
+ input: 1.475,
13778
+ output: 4.425,
13779
+ cacheRead: 0.295,
13780
+ cacheWrite: 1.84375,
13296
13781
  },
13297
13782
  contextWindow: 1000000,
13298
13783
  maxTokens: 65536,
@@ -13413,7 +13898,7 @@ export const MODELS = {
13413
13898
  cacheRead: 0.04,
13414
13899
  cacheWrite: 0,
13415
13900
  },
13416
- contextWindow: 256000,
13901
+ contextWindow: 262144,
13417
13902
  maxTokens: 256000,
13418
13903
  },
13419
13904
  "tencent/hy3": {
@@ -13425,13 +13910,13 @@ export const MODELS = {
13425
13910
  reasoning: true,
13426
13911
  input: ["text"],
13427
13912
  cost: {
13428
- input: 0.14,
13429
- output: 0.58,
13430
- cacheRead: 0.035,
13913
+ input: 0.13199999999999998,
13914
+ output: 0.5279999999999999,
13915
+ cacheRead: 0.032999999999999995,
13431
13916
  cacheWrite: 0,
13432
13917
  },
13433
13918
  contextWindow: 262144,
13434
- maxTokens: 4096,
13919
+ maxTokens: 128000,
13435
13920
  },
13436
13921
  "tencent/hy3-preview": {
13437
13922
  id: "tencent/hy3-preview",
@@ -13450,23 +13935,6 @@ export const MODELS = {
13450
13935
  contextWindow: 262144,
13451
13936
  maxTokens: 4096,
13452
13937
  },
13453
- "tencent/hy3:free": {
13454
- id: "tencent/hy3:free",
13455
- name: "Tencent: Hy3 (free)",
13456
- api: "openai-completions",
13457
- provider: "openrouter",
13458
- baseUrl: "https://openrouter.ai/api/v1",
13459
- reasoning: true,
13460
- input: ["text"],
13461
- cost: {
13462
- input: 0,
13463
- output: 0,
13464
- cacheRead: 0,
13465
- cacheWrite: 0,
13466
- },
13467
- contextWindow: 262144,
13468
- maxTokens: 262144,
13469
- },
13470
13938
  "thedrummer/unslopnemo-12b": {
13471
13939
  id: "thedrummer/unslopnemo-12b",
13472
13940
  name: "TheDrummer: UnslopNemo 12B",
@@ -13484,6 +13952,23 @@ export const MODELS = {
13484
13952
  contextWindow: 32768,
13485
13953
  maxTokens: 32768,
13486
13954
  },
13955
+ "thinkingmachines/inkling": {
13956
+ id: "thinkingmachines/inkling",
13957
+ name: "Thinking Machines: Inkling",
13958
+ api: "openai-completions",
13959
+ provider: "openrouter",
13960
+ baseUrl: "https://openrouter.ai/api/v1",
13961
+ reasoning: true,
13962
+ input: ["text", "image"],
13963
+ cost: {
13964
+ input: 1,
13965
+ output: 4.05,
13966
+ cacheRead: 0.16999999999999998,
13967
+ cacheWrite: 0,
13968
+ },
13969
+ contextWindow: 1048576,
13970
+ maxTokens: 4096,
13971
+ },
13487
13972
  "upstage/solar-pro-3": {
13488
13973
  id: "upstage/solar-pro-3",
13489
13974
  name: "Upstage: Solar Pro 3",
@@ -13546,7 +14031,7 @@ export const MODELS = {
13546
14031
  cost: {
13547
14032
  input: 2,
13548
14033
  output: 6,
13549
- cacheRead: 0.5,
14034
+ cacheRead: 0.3,
13550
14035
  cacheWrite: 0,
13551
14036
  },
13552
14037
  contextWindow: 500000,
@@ -13578,13 +14063,13 @@ export const MODELS = {
13578
14063
  reasoning: true,
13579
14064
  input: ["text", "image"],
13580
14065
  cost: {
13581
- input: 0.105,
14066
+ input: 0.14,
13582
14067
  output: 0.28,
13583
- cacheRead: 0.028,
14068
+ cacheRead: 0.0028,
13584
14069
  cacheWrite: 0,
13585
14070
  },
13586
- contextWindow: 1048576,
13587
- maxTokens: 4096,
14071
+ contextWindow: 1050000,
14072
+ maxTokens: 131072,
13588
14073
  },
13589
14074
  "xiaomi/mimo-v2.5-pro": {
13590
14075
  id: "xiaomi/mimo-v2.5-pro",
@@ -13600,7 +14085,7 @@ export const MODELS = {
13600
14085
  cacheRead: 0.0036,
13601
14086
  cacheWrite: 0,
13602
14087
  },
13603
- contextWindow: 1048576,
14088
+ contextWindow: 1050000,
13604
14089
  maxTokens: 131072,
13605
14090
  },
13606
14091
  "z-ai/glm-4.5": {
@@ -13663,13 +14148,13 @@ export const MODELS = {
13663
14148
  reasoning: true,
13664
14149
  input: ["text"],
13665
14150
  cost: {
13666
- input: 0.43,
13667
- output: 1.75,
13668
- cacheRead: 0.08,
14151
+ input: 0.5,
14152
+ output: 2,
14153
+ cacheRead: 0.09999999999999999,
13669
14154
  cacheWrite: 0,
13670
14155
  },
13671
- contextWindow: 200000,
13672
- maxTokens: 16384,
14156
+ contextWindow: 204800,
14157
+ maxTokens: 131072,
13673
14158
  },
13674
14159
  "z-ai/glm-4.6v": {
13675
14160
  id: "z-ai/glm-4.6v",
@@ -13702,7 +14187,7 @@ export const MODELS = {
13702
14187
  cacheRead: 0.08,
13703
14188
  cacheWrite: 0,
13704
14189
  },
13705
- contextWindow: 202752,
14190
+ contextWindow: 204800,
13706
14191
  maxTokens: 131072,
13707
14192
  },
13708
14193
  "z-ai/glm-4.7-flash": {
@@ -13736,8 +14221,8 @@ export const MODELS = {
13736
14221
  cacheRead: 0.119,
13737
14222
  cacheWrite: 0,
13738
14223
  },
13739
- contextWindow: 202752,
13740
- maxTokens: 4096,
14224
+ contextWindow: 204800,
14225
+ maxTokens: 131072,
13741
14226
  },
13742
14227
  "z-ai/glm-5-turbo": {
13743
14228
  id: "z-ai/glm-5-turbo",
@@ -13753,7 +14238,7 @@ export const MODELS = {
13753
14238
  cacheRead: 0.24,
13754
14239
  cacheWrite: 0,
13755
14240
  },
13756
- contextWindow: 262144,
14241
+ contextWindow: 202752,
13757
14242
  maxTokens: 131072,
13758
14243
  },
13759
14244
  "z-ai/glm-5.1": {
@@ -13770,7 +14255,7 @@ export const MODELS = {
13770
14255
  cacheRead: 0.1794,
13771
14256
  cacheWrite: 0,
13772
14257
  },
13773
- contextWindow: 202752,
14258
+ contextWindow: 204800,
13774
14259
  maxTokens: 128000,
13775
14260
  },
13776
14261
  "z-ai/glm-5.2": {
@@ -13782,13 +14267,13 @@ export const MODELS = {
13782
14267
  reasoning: true,
13783
14268
  input: ["text"],
13784
14269
  cost: {
13785
- input: 0.9299999999999999,
13786
- output: 3,
13787
- cacheRead: 0.18,
14270
+ input: 0.7616,
14271
+ output: 2.3935999999999997,
14272
+ cacheRead: 0.14144,
13788
14273
  cacheWrite: 0,
13789
14274
  },
13790
14275
  contextWindow: 1048576,
13791
- maxTokens: 32768,
14276
+ maxTokens: 131072,
13792
14277
  },
13793
14278
  "z-ai/glm-5v-turbo": {
13794
14279
  id: "z-ai/glm-5v-turbo",
@@ -13885,7 +14370,7 @@ export const MODELS = {
13885
14370
  input: ["text", "image"],
13886
14371
  cost: {
13887
14372
  input: 1.5,
13888
- output: 9,
14373
+ output: 7.5,
13889
14374
  cacheRead: 0.15,
13890
14375
  cacheWrite: 0.08333333333333334,
13891
14376
  },
@@ -13918,13 +14403,13 @@ export const MODELS = {
13918
14403
  reasoning: true,
13919
14404
  input: ["text", "image"],
13920
14405
  cost: {
13921
- input: 0.66,
13922
- output: 3.41,
13923
- cacheRead: 0.15,
14406
+ input: 3,
14407
+ output: 15,
14408
+ cacheRead: 0.3,
13924
14409
  cacheWrite: 0,
13925
14410
  },
13926
- contextWindow: 262144,
13927
- maxTokens: 262144,
14411
+ contextWindow: 1048576,
14412
+ maxTokens: 4096,
13928
14413
  },
13929
14414
  "~openai/gpt-latest": {
13930
14415
  id: "~openai/gpt-latest",
@@ -13971,7 +14456,7 @@ export const MODELS = {
13971
14456
  cost: {
13972
14457
  input: 2,
13973
14458
  output: 6,
13974
- cacheRead: 0.5,
14459
+ cacheRead: 0.3,
13975
14460
  cacheWrite: 0,
13976
14461
  },
13977
14462
  contextWindow: 500000,
@@ -14035,43 +14520,6 @@ export const MODELS = {
14035
14520
  contextWindow: 32768,
14036
14521
  maxTokens: 32768,
14037
14522
  },
14038
- "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
14039
- id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
14040
- name: "Qwen3 235B A22B Instruct 2507 FP8",
14041
- api: "openai-completions",
14042
- provider: "together",
14043
- baseUrl: "https://api.together.ai/v1",
14044
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
14045
- reasoning: false,
14046
- input: ["text"],
14047
- cost: {
14048
- input: 0.2,
14049
- output: 0.6,
14050
- cacheRead: 0,
14051
- cacheWrite: 0,
14052
- },
14053
- contextWindow: 262144,
14054
- maxTokens: 262144,
14055
- },
14056
- "Qwen/Qwen3.5-397B-A17B": {
14057
- id: "Qwen/Qwen3.5-397B-A17B",
14058
- name: "Qwen3.5 397B A17B",
14059
- api: "openai-completions",
14060
- provider: "together",
14061
- baseUrl: "https://api.together.ai/v1",
14062
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
14063
- reasoning: true,
14064
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
14065
- input: ["text", "image"],
14066
- cost: {
14067
- input: 0.6,
14068
- output: 3.6,
14069
- cacheRead: 0,
14070
- cacheWrite: 0,
14071
- },
14072
- contextWindow: 262144,
14073
- maxTokens: 130000,
14074
- },
14075
14523
  "Qwen/Qwen3.5-9B": {
14076
14524
  id: "Qwen/Qwen3.5-9B",
14077
14525
  name: "Qwen3.5 9B",
@@ -14147,24 +14595,6 @@ export const MODELS = {
14147
14595
  contextWindow: 512000,
14148
14596
  maxTokens: 384000,
14149
14597
  },
14150
- "essentialai/Rnj-1-Instruct": {
14151
- id: "essentialai/Rnj-1-Instruct",
14152
- name: "Rnj-1 Instruct",
14153
- api: "openai-completions",
14154
- provider: "together",
14155
- baseUrl: "https://api.together.ai/v1",
14156
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
14157
- reasoning: false,
14158
- input: ["text"],
14159
- cost: {
14160
- input: 0.15,
14161
- output: 0.15,
14162
- cacheRead: 0,
14163
- cacheWrite: 0,
14164
- },
14165
- contextWindow: 32768,
14166
- maxTokens: 32768,
14167
- },
14168
14598
  "google/gemma-4-31B-it": {
14169
14599
  id: "google/gemma-4-31B-it",
14170
14600
  name: "Gemma 4 31B Instruct",
@@ -14297,42 +14727,23 @@ export const MODELS = {
14297
14727
  contextWindow: 131072,
14298
14728
  maxTokens: 131072,
14299
14729
  },
14300
- "zai-org/GLM-5": {
14301
- id: "zai-org/GLM-5",
14302
- name: "GLM-5",
14730
+ "thinkingmachines/Inkling": {
14731
+ id: "thinkingmachines/Inkling",
14732
+ name: "Inkling",
14303
14733
  api: "openai-completions",
14304
14734
  provider: "together",
14305
14735
  baseUrl: "https://api.together.ai/v1",
14306
14736
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
14307
14737
  reasoning: true,
14308
14738
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
14309
- input: ["text"],
14739
+ input: ["text", "image"],
14310
14740
  cost: {
14311
14741
  input: 1,
14312
- output: 3.2,
14313
- cacheRead: 0,
14314
- cacheWrite: 0,
14315
- },
14316
- contextWindow: 202752,
14317
- maxTokens: 131072,
14318
- },
14319
- "zai-org/GLM-5.1": {
14320
- id: "zai-org/GLM-5.1",
14321
- name: "GLM-5.1",
14322
- api: "openai-completions",
14323
- provider: "together",
14324
- baseUrl: "https://api.together.ai/v1",
14325
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
14326
- reasoning: true,
14327
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
14328
- input: ["text"],
14329
- cost: {
14330
- input: 1.4,
14331
- output: 4.4,
14332
- cacheRead: 0,
14742
+ output: 4.05,
14743
+ cacheRead: 0.17,
14333
14744
  cacheWrite: 0,
14334
14745
  },
14335
- contextWindow: 202752,
14746
+ contextWindow: 524288,
14336
14747
  maxTokens: 131072,
14337
14748
  },
14338
14749
  "zai-org/GLM-5.2": {
@@ -14739,10 +15150,10 @@ export const MODELS = {
14739
15150
  reasoning: true,
14740
15151
  input: ["text"],
14741
15152
  cost: {
14742
- input: 1.25,
14743
- output: 3.75,
14744
- cacheRead: 0.25,
14745
- cacheWrite: 1.5625,
15153
+ input: 2.5,
15154
+ output: 7.5,
15155
+ cacheRead: 0.5,
15156
+ cacheWrite: 3.125,
14746
15157
  },
14747
15158
  contextWindow: 991000,
14748
15159
  maxTokens: 64000,
@@ -14849,23 +15260,6 @@ export const MODELS = {
14849
15260
  contextWindow: 200000,
14850
15261
  maxTokens: 4096,
14851
15262
  },
14852
- "anthropic/claude-3.5-haiku": {
14853
- id: "anthropic/claude-3.5-haiku",
14854
- name: "Claude 3.5 Haiku",
14855
- api: "anthropic-messages",
14856
- provider: "vercel-ai-gateway",
14857
- baseUrl: "https://ai-gateway.vercel.sh",
14858
- reasoning: false,
14859
- input: ["text", "image"],
14860
- cost: {
14861
- input: 0.7999999999999999,
14862
- output: 4,
14863
- cacheRead: 0.08,
14864
- cacheWrite: 1,
14865
- },
14866
- contextWindow: 200000,
14867
- maxTokens: 8192,
14868
- },
14869
15263
  "anthropic/claude-fable-5": {
14870
15264
  id: "anthropic/claude-fable-5",
14871
15265
  name: "Claude Fable 5",
@@ -15007,6 +15401,58 @@ export const MODELS = {
15007
15401
  contextWindow: 1000000,
15008
15402
  maxTokens: 128000,
15009
15403
  },
15404
+ "anthropic/claude-opus-4.8-fast": {
15405
+ id: "anthropic/claude-opus-4.8-fast",
15406
+ name: "Claude Opus 4.8 (Fast)",
15407
+ api: "anthropic-messages",
15408
+ provider: "vercel-ai-gateway",
15409
+ baseUrl: "https://ai-gateway.vercel.sh",
15410
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
15411
+ reasoning: true,
15412
+ input: ["text", "image"],
15413
+ cost: {
15414
+ input: 10,
15415
+ output: 50,
15416
+ cacheRead: 1,
15417
+ cacheWrite: 12.5,
15418
+ },
15419
+ contextWindow: 1000000,
15420
+ maxTokens: 128000,
15421
+ },
15422
+ "anthropic/claude-opus-5": {
15423
+ id: "anthropic/claude-opus-5",
15424
+ name: "Claude Opus 5",
15425
+ api: "anthropic-messages",
15426
+ provider: "vercel-ai-gateway",
15427
+ baseUrl: "https://ai-gateway.vercel.sh",
15428
+ reasoning: true,
15429
+ input: ["text", "image"],
15430
+ cost: {
15431
+ input: 5,
15432
+ output: 25,
15433
+ cacheRead: 0.5,
15434
+ cacheWrite: 6.25,
15435
+ },
15436
+ contextWindow: 1000000,
15437
+ maxTokens: 128000,
15438
+ },
15439
+ "anthropic/claude-opus-5-fast": {
15440
+ id: "anthropic/claude-opus-5-fast",
15441
+ name: "Claude Opus 5 (Fast)",
15442
+ api: "anthropic-messages",
15443
+ provider: "vercel-ai-gateway",
15444
+ baseUrl: "https://ai-gateway.vercel.sh",
15445
+ reasoning: true,
15446
+ input: ["text", "image"],
15447
+ cost: {
15448
+ input: 10,
15449
+ output: 50,
15450
+ cacheRead: 1,
15451
+ cacheWrite: 12.5,
15452
+ },
15453
+ contextWindow: 1000000,
15454
+ maxTokens: 128000,
15455
+ },
15010
15456
  "anthropic/claude-sonnet-4": {
15011
15457
  id: "anthropic/claude-sonnet-4",
15012
15458
  name: "Claude Sonnet 4",
@@ -15076,23 +15522,6 @@ export const MODELS = {
15076
15522
  contextWindow: 1000000,
15077
15523
  maxTokens: 128000,
15078
15524
  },
15079
- "arcee-ai/trinity-large-preview": {
15080
- id: "arcee-ai/trinity-large-preview",
15081
- name: "Trinity Large Preview",
15082
- api: "anthropic-messages",
15083
- provider: "vercel-ai-gateway",
15084
- baseUrl: "https://ai-gateway.vercel.sh",
15085
- reasoning: false,
15086
- input: ["text"],
15087
- cost: {
15088
- input: 0.25,
15089
- output: 1,
15090
- cacheRead: 0,
15091
- cacheWrite: 0,
15092
- },
15093
- contextWindow: 131000,
15094
- maxTokens: 131000,
15095
- },
15096
15525
  "arcee-ai/trinity-large-thinking": {
15097
15526
  id: "arcee-ai/trinity-large-thinking",
15098
15527
  name: "Trinity Large Thinking",
@@ -15221,8 +15650,8 @@ export const MODELS = {
15221
15650
  reasoning: true,
15222
15651
  input: ["text"],
15223
15652
  cost: {
15224
- input: 0.21,
15225
- output: 0.7899999999999999,
15653
+ input: 0.25,
15654
+ output: 0.95,
15226
15655
  cacheRead: 0.13,
15227
15656
  cacheWrite: 0,
15228
15657
  },
@@ -15416,43 +15845,60 @@ export const MODELS = {
15416
15845
  contextWindow: 1000000,
15417
15846
  maxTokens: 65000,
15418
15847
  },
15419
- "google/gemini-3.1-flash-lite-preview": {
15420
- id: "google/gemini-3.1-flash-lite-preview",
15421
- name: "Gemini 3.1 Flash Lite Preview",
15848
+ "google/gemini-3.1-pro-preview": {
15849
+ id: "google/gemini-3.1-pro-preview",
15850
+ name: "Gemini 3.1 Pro Preview",
15851
+ api: "anthropic-messages",
15852
+ provider: "vercel-ai-gateway",
15853
+ baseUrl: "https://ai-gateway.vercel.sh",
15854
+ reasoning: true,
15855
+ input: ["text", "image"],
15856
+ cost: {
15857
+ input: 2,
15858
+ output: 12,
15859
+ cacheRead: 0.19999999999999998,
15860
+ cacheWrite: 0,
15861
+ },
15862
+ contextWindow: 1000000,
15863
+ maxTokens: 64000,
15864
+ },
15865
+ "google/gemini-3.5-flash": {
15866
+ id: "google/gemini-3.5-flash",
15867
+ name: "Gemini 3.5 Flash",
15422
15868
  api: "anthropic-messages",
15423
15869
  provider: "vercel-ai-gateway",
15424
15870
  baseUrl: "https://ai-gateway.vercel.sh",
15425
15871
  reasoning: true,
15426
15872
  input: ["text", "image"],
15427
15873
  cost: {
15428
- input: 0.25,
15429
- output: 1.5,
15430
- cacheRead: 0.03,
15874
+ input: 1.5,
15875
+ output: 9,
15876
+ cacheRead: 0.15,
15431
15877
  cacheWrite: 0,
15432
15878
  },
15433
15879
  contextWindow: 1000000,
15434
- maxTokens: 65000,
15880
+ maxTokens: 64000,
15435
15881
  },
15436
- "google/gemini-3.1-pro-preview": {
15437
- id: "google/gemini-3.1-pro-preview",
15438
- name: "Gemini 3.1 Pro Preview",
15882
+ "google/gemini-3.5-flash-lite": {
15883
+ id: "google/gemini-3.5-flash-lite",
15884
+ name: "Gemini 3.5 Flash Lite",
15439
15885
  api: "anthropic-messages",
15440
15886
  provider: "vercel-ai-gateway",
15441
15887
  baseUrl: "https://ai-gateway.vercel.sh",
15442
15888
  reasoning: true,
15443
15889
  input: ["text", "image"],
15444
15890
  cost: {
15445
- input: 2,
15446
- output: 12,
15447
- cacheRead: 0.19999999999999998,
15891
+ input: 0.3,
15892
+ output: 2.5,
15893
+ cacheRead: 0.03,
15448
15894
  cacheWrite: 0,
15449
15895
  },
15450
15896
  contextWindow: 1000000,
15451
- maxTokens: 64000,
15897
+ maxTokens: 65000,
15452
15898
  },
15453
- "google/gemini-3.5-flash": {
15454
- id: "google/gemini-3.5-flash",
15455
- name: "Gemini 3.5 Flash",
15899
+ "google/gemini-3.6-flash": {
15900
+ id: "google/gemini-3.6-flash",
15901
+ name: "Gemini 3.6 Flash",
15456
15902
  api: "anthropic-messages",
15457
15903
  provider: "vercel-ai-gateway",
15458
15904
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -15460,7 +15906,7 @@ export const MODELS = {
15460
15906
  input: ["text", "image"],
15461
15907
  cost: {
15462
15908
  input: 1.5,
15463
- output: 9,
15909
+ output: 7.5,
15464
15910
  cacheRead: 0.15,
15465
15911
  cacheWrite: 0,
15466
15912
  },
@@ -15535,6 +15981,23 @@ export const MODELS = {
15535
15981
  contextWindow: 32000,
15536
15982
  maxTokens: 16384,
15537
15983
  },
15984
+ "inclusionai/ling-3.0-flash-free": {
15985
+ id: "inclusionai/ling-3.0-flash-free",
15986
+ name: "Ling 3.0 Flash",
15987
+ api: "anthropic-messages",
15988
+ provider: "vercel-ai-gateway",
15989
+ baseUrl: "https://ai-gateway.vercel.sh",
15990
+ reasoning: true,
15991
+ input: ["text"],
15992
+ cost: {
15993
+ input: 0,
15994
+ output: 0,
15995
+ cacheRead: 0,
15996
+ cacheWrite: 0,
15997
+ },
15998
+ contextWindow: 256000,
15999
+ maxTokens: 256000,
16000
+ },
15538
16001
  "interfaze/interfaze-beta": {
15539
16002
  id: "interfaze/interfaze-beta",
15540
16003
  name: "Interfaze Beta",
@@ -15620,40 +16083,6 @@ export const MODELS = {
15620
16083
  contextWindow: 256000,
15621
16084
  maxTokens: 80000,
15622
16085
  },
15623
- "meituan/longcat-flash-chat": {
15624
- id: "meituan/longcat-flash-chat",
15625
- name: "LongCat Flash Chat",
15626
- api: "anthropic-messages",
15627
- provider: "vercel-ai-gateway",
15628
- baseUrl: "https://ai-gateway.vercel.sh",
15629
- reasoning: false,
15630
- input: ["text"],
15631
- cost: {
15632
- input: 0,
15633
- output: 0,
15634
- cacheRead: 0,
15635
- cacheWrite: 0,
15636
- },
15637
- contextWindow: 128000,
15638
- maxTokens: 100000,
15639
- },
15640
- "meituan/longcat-flash-thinking-2601": {
15641
- id: "meituan/longcat-flash-thinking-2601",
15642
- name: "LongCat Flash Thinking 2601",
15643
- api: "anthropic-messages",
15644
- provider: "vercel-ai-gateway",
15645
- baseUrl: "https://ai-gateway.vercel.sh",
15646
- reasoning: true,
15647
- input: ["text"],
15648
- cost: {
15649
- input: 0,
15650
- output: 0,
15651
- cacheRead: 0,
15652
- cacheWrite: 0,
15653
- },
15654
- contextWindow: 32768,
15655
- maxTokens: 32768,
15656
- },
15657
16086
  "meta/llama-3.1-70b": {
15658
16087
  id: "meta/llama-3.1-70b",
15659
16088
  name: "Llama 3.1 70B Instruct",
@@ -15688,40 +16117,6 @@ export const MODELS = {
15688
16117
  contextWindow: 128000,
15689
16118
  maxTokens: 8192,
15690
16119
  },
15691
- "meta/llama-3.2-11b": {
15692
- id: "meta/llama-3.2-11b",
15693
- name: "Llama 3.2 11B Vision Instruct",
15694
- api: "anthropic-messages",
15695
- provider: "vercel-ai-gateway",
15696
- baseUrl: "https://ai-gateway.vercel.sh",
15697
- reasoning: false,
15698
- input: ["text", "image"],
15699
- cost: {
15700
- input: 0.16,
15701
- output: 0.16,
15702
- cacheRead: 0,
15703
- cacheWrite: 0,
15704
- },
15705
- contextWindow: 128000,
15706
- maxTokens: 8192,
15707
- },
15708
- "meta/llama-3.2-90b": {
15709
- id: "meta/llama-3.2-90b",
15710
- name: "Llama 3.2 90B Vision Instruct",
15711
- api: "anthropic-messages",
15712
- provider: "vercel-ai-gateway",
15713
- baseUrl: "https://ai-gateway.vercel.sh",
15714
- reasoning: false,
15715
- input: ["text", "image"],
15716
- cost: {
15717
- input: 0.72,
15718
- output: 0.72,
15719
- cacheRead: 0,
15720
- cacheWrite: 0,
15721
- },
15722
- contextWindow: 128000,
15723
- maxTokens: 8192,
15724
- },
15725
16120
  "meta/llama-3.3-70b": {
15726
16121
  id: "meta/llama-3.3-70b",
15727
16122
  name: "Llama 3.3 70B Instruct",
@@ -15960,23 +16355,6 @@ export const MODELS = {
15960
16355
  contextWindow: 256000,
15961
16356
  maxTokens: 256000,
15962
16357
  },
15963
- "mistral/devstral-small": {
15964
- id: "mistral/devstral-small",
15965
- name: "Devstral Small 1.1",
15966
- api: "anthropic-messages",
15967
- provider: "vercel-ai-gateway",
15968
- baseUrl: "https://ai-gateway.vercel.sh",
15969
- reasoning: false,
15970
- input: ["text"],
15971
- cost: {
15972
- input: 0.09999999999999999,
15973
- output: 0.3,
15974
- cacheRead: 0,
15975
- cacheWrite: 0,
15976
- },
15977
- contextWindow: 128000,
15978
- maxTokens: 64000,
15979
- },
15980
16358
  "mistral/devstral-small-2": {
15981
16359
  id: "mistral/devstral-small-2",
15982
16360
  name: "Devstral Small 2",
@@ -16181,23 +16559,6 @@ export const MODELS = {
16181
16559
  contextWindow: 128000,
16182
16560
  maxTokens: 4000,
16183
16561
  },
16184
- "mistral/pixtral-large": {
16185
- id: "mistral/pixtral-large",
16186
- name: "Pixtral Large",
16187
- api: "anthropic-messages",
16188
- provider: "vercel-ai-gateway",
16189
- baseUrl: "https://ai-gateway.vercel.sh",
16190
- reasoning: false,
16191
- input: ["text", "image"],
16192
- cost: {
16193
- input: 2,
16194
- output: 6,
16195
- cacheRead: 0,
16196
- cacheWrite: 0,
16197
- },
16198
- contextWindow: 128000,
16199
- maxTokens: 4000,
16200
- },
16201
16562
  "moonshotai/kimi-k2": {
16202
16563
  id: "moonshotai/kimi-k2",
16203
16564
  name: "Kimi K2 Instruct",
@@ -16300,6 +16661,23 @@ export const MODELS = {
16300
16661
  contextWindow: 262144,
16301
16662
  maxTokens: 32768,
16302
16663
  },
16664
+ "moonshotai/kimi-k3": {
16665
+ id: "moonshotai/kimi-k3",
16666
+ name: "Kimi K3",
16667
+ api: "anthropic-messages",
16668
+ provider: "vercel-ai-gateway",
16669
+ baseUrl: "https://ai-gateway.vercel.sh",
16670
+ reasoning: true,
16671
+ input: ["text", "image"],
16672
+ cost: {
16673
+ input: 3,
16674
+ output: 15,
16675
+ cacheRead: 0.3,
16676
+ cacheWrite: 0,
16677
+ },
16678
+ contextWindow: 1000000,
16679
+ maxTokens: 131072,
16680
+ },
16303
16681
  "nvidia/nemotron-3-nano-30b-a3b": {
16304
16682
  id: "nvidia/nemotron-3-nano-30b-a3b",
16305
16683
  name: "Nemotron 3 Nano 30B A3B",
@@ -16521,23 +16899,6 @@ export const MODELS = {
16521
16899
  contextWindow: 400000,
16522
16900
  maxTokens: 128000,
16523
16901
  },
16524
- "openai/gpt-5-chat": {
16525
- id: "openai/gpt-5-chat",
16526
- name: "GPT 5 Chat",
16527
- api: "anthropic-messages",
16528
- provider: "vercel-ai-gateway",
16529
- baseUrl: "https://ai-gateway.vercel.sh",
16530
- reasoning: false,
16531
- input: ["text", "image"],
16532
- cost: {
16533
- input: 1.25,
16534
- output: 10,
16535
- cacheRead: 0.125,
16536
- cacheWrite: 0,
16537
- },
16538
- contextWindow: 128000,
16539
- maxTokens: 16384,
16540
- },
16541
16902
  "openai/gpt-5-codex": {
16542
16903
  id: "openai/gpt-5-codex",
16543
16904
  name: "GPT-5-Codex",
@@ -16709,24 +17070,6 @@ export const MODELS = {
16709
17070
  contextWindow: 400000,
16710
17071
  maxTokens: 128000,
16711
17072
  },
16712
- "openai/gpt-5.2-chat": {
16713
- id: "openai/gpt-5.2-chat",
16714
- name: "GPT 5.2 Chat",
16715
- api: "anthropic-messages",
16716
- provider: "vercel-ai-gateway",
16717
- baseUrl: "https://ai-gateway.vercel.sh",
16718
- reasoning: false,
16719
- thinkingLevelMap: { "xhigh": "xhigh" },
16720
- input: ["text", "image"],
16721
- cost: {
16722
- input: 1.75,
16723
- output: 14,
16724
- cacheRead: 0.175,
16725
- cacheWrite: 0,
16726
- },
16727
- contextWindow: 128000,
16728
- maxTokens: 16384,
16729
- },
16730
17073
  "openai/gpt-5.2-codex": {
16731
17074
  id: "openai/gpt-5.2-codex",
16732
17075
  name: "GPT 5.2 Codex",
@@ -17114,6 +17457,40 @@ export const MODELS = {
17114
17457
  contextWindow: 200000,
17115
17458
  maxTokens: 100000,
17116
17459
  },
17460
+ "poolside/laguna-s-2.1": {
17461
+ id: "poolside/laguna-s-2.1",
17462
+ name: "Laguna S 2.1",
17463
+ api: "anthropic-messages",
17464
+ provider: "vercel-ai-gateway",
17465
+ baseUrl: "https://ai-gateway.vercel.sh",
17466
+ reasoning: true,
17467
+ input: ["text"],
17468
+ cost: {
17469
+ input: 0.09999999999999999,
17470
+ output: 0.19999999999999998,
17471
+ cacheRead: 0.01,
17472
+ cacheWrite: 0,
17473
+ },
17474
+ contextWindow: 1000000,
17475
+ maxTokens: 131072,
17476
+ },
17477
+ "poolside/laguna-s-2.1-free": {
17478
+ id: "poolside/laguna-s-2.1-free",
17479
+ name: "Laguna S 2.1 Free",
17480
+ api: "anthropic-messages",
17481
+ provider: "vercel-ai-gateway",
17482
+ baseUrl: "https://ai-gateway.vercel.sh",
17483
+ reasoning: true,
17484
+ input: ["text"],
17485
+ cost: {
17486
+ input: 0,
17487
+ output: 0,
17488
+ cacheRead: 0,
17489
+ cacheWrite: 0,
17490
+ },
17491
+ contextWindow: 256000,
17492
+ maxTokens: 32768,
17493
+ },
17117
17494
  "sakana/fugu-ultra": {
17118
17495
  id: "sakana/fugu-ultra",
17119
17496
  name: "Fugu Ultra",
@@ -17165,6 +17542,40 @@ export const MODELS = {
17165
17542
  contextWindow: 256000,
17166
17543
  maxTokens: 256000,
17167
17544
  },
17545
+ "tencent/hy3": {
17546
+ id: "tencent/hy3",
17547
+ name: "Hy3",
17548
+ api: "anthropic-messages",
17549
+ provider: "vercel-ai-gateway",
17550
+ baseUrl: "https://ai-gateway.vercel.sh",
17551
+ reasoning: true,
17552
+ input: ["text"],
17553
+ cost: {
17554
+ input: 0.14,
17555
+ output: 0.58,
17556
+ cacheRead: 0.035,
17557
+ cacheWrite: 0,
17558
+ },
17559
+ contextWindow: 262144,
17560
+ maxTokens: 262144,
17561
+ },
17562
+ "thinkingmachines/inkling": {
17563
+ id: "thinkingmachines/inkling",
17564
+ name: "Inkling",
17565
+ api: "anthropic-messages",
17566
+ provider: "vercel-ai-gateway",
17567
+ baseUrl: "https://ai-gateway.vercel.sh",
17568
+ reasoning: true,
17569
+ input: ["text", "image"],
17570
+ cost: {
17571
+ input: 1,
17572
+ output: 4.05,
17573
+ cacheRead: 0.16999999999999998,
17574
+ cacheWrite: 0,
17575
+ },
17576
+ contextWindow: 256000,
17577
+ maxTokens: 256000,
17578
+ },
17168
17579
  "xai/grok-4.1-fast-non-reasoning": {
17169
17580
  id: "xai/grok-4.1-fast-non-reasoning",
17170
17581
  name: "Grok 4.1 Fast Non-Reasoning",
@@ -17329,7 +17740,7 @@ export const MODELS = {
17329
17740
  cost: {
17330
17741
  input: 2,
17331
17742
  output: 6,
17332
- cacheRead: 0.5,
17743
+ cacheRead: 0.3,
17333
17744
  cacheWrite: 0,
17334
17745
  },
17335
17746
  contextWindow: 500000,
@@ -17352,40 +17763,6 @@ export const MODELS = {
17352
17763
  contextWindow: 256000,
17353
17764
  maxTokens: 256000,
17354
17765
  },
17355
- "xiaomi/mimo-v2-flash": {
17356
- id: "xiaomi/mimo-v2-flash",
17357
- name: "MiMo V2 Flash",
17358
- api: "anthropic-messages",
17359
- provider: "vercel-ai-gateway",
17360
- baseUrl: "https://ai-gateway.vercel.sh",
17361
- reasoning: true,
17362
- input: ["text"],
17363
- cost: {
17364
- input: 0.09999999999999999,
17365
- output: 0.3,
17366
- cacheRead: 0.01,
17367
- cacheWrite: 0,
17368
- },
17369
- contextWindow: 262144,
17370
- maxTokens: 32000,
17371
- },
17372
- "xiaomi/mimo-v2-pro": {
17373
- id: "xiaomi/mimo-v2-pro",
17374
- name: "MiMo V2 Pro",
17375
- api: "anthropic-messages",
17376
- provider: "vercel-ai-gateway",
17377
- baseUrl: "https://ai-gateway.vercel.sh",
17378
- reasoning: true,
17379
- input: ["text"],
17380
- cost: {
17381
- input: 1,
17382
- output: 3,
17383
- cacheRead: 0.19999999999999998,
17384
- cacheWrite: 0,
17385
- },
17386
- contextWindow: 1000000,
17387
- maxTokens: 128000,
17388
- },
17389
17766
  "xiaomi/mimo-v2.5": {
17390
17767
  id: "xiaomi/mimo-v2.5",
17391
17768
  name: "MiMo M2.5",
@@ -17773,7 +18150,7 @@ export const MODELS = {
17773
18150
  cost: {
17774
18151
  input: 2,
17775
18152
  output: 6,
17776
- cacheRead: 0.5,
18153
+ cacheRead: 0.3,
17777
18154
  cacheWrite: 0,
17778
18155
  },
17779
18156
  contextWindow: 500000,