@elyracode/ai 0.8.4 → 0.9.1

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.
@@ -391,8 +391,8 @@ export const MODELS = {
391
391
  thinkingLevelMap: { "xhigh": "max" },
392
392
  input: ["text", "image"],
393
393
  cost: {
394
- input: 5,
395
- output: 25,
394
+ input: 5.5,
395
+ output: 27.5,
396
396
  cacheRead: 0.5,
397
397
  cacheWrite: 6.25,
398
398
  },
@@ -409,10 +409,10 @@ export const MODELS = {
409
409
  thinkingLevelMap: { "xhigh": "xhigh" },
410
410
  input: ["text", "image"],
411
411
  cost: {
412
- input: 5,
413
- output: 25,
414
- cacheRead: 0.5,
415
- cacheWrite: 6.25,
412
+ input: 5.5,
413
+ output: 27.5,
414
+ cacheRead: 0.55,
415
+ cacheWrite: 6.875,
416
416
  },
417
417
  contextWindow: 1000000,
418
418
  maxTokens: 128000,
@@ -427,10 +427,10 @@ export const MODELS = {
427
427
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
428
428
  input: ["text", "image"],
429
429
  cost: {
430
- input: 5,
431
- output: 25,
432
- cacheRead: 0.5,
433
- cacheWrite: 6.25,
430
+ input: 5.5,
431
+ output: 27.5,
432
+ cacheRead: 0.55,
433
+ cacheWrite: 6.875,
434
434
  },
435
435
  contextWindow: 1000000,
436
436
  maxTokens: 128000,
@@ -444,10 +444,10 @@ export const MODELS = {
444
444
  reasoning: true,
445
445
  input: ["text", "image"],
446
446
  cost: {
447
- input: 3,
448
- output: 15,
449
- cacheRead: 0.3,
450
- cacheWrite: 3.75,
447
+ input: 3.3,
448
+ output: 16.5,
449
+ cacheRead: 0.33,
450
+ cacheWrite: 4.125,
451
451
  },
452
452
  contextWindow: 200000,
453
453
  maxTokens: 64000,
@@ -461,10 +461,10 @@ export const MODELS = {
461
461
  reasoning: true,
462
462
  input: ["text", "image"],
463
463
  cost: {
464
- input: 3,
465
- output: 15,
466
- cacheRead: 0.3,
467
- cacheWrite: 3.75,
464
+ input: 3.3,
465
+ output: 16.5,
466
+ cacheRead: 0.33,
467
+ cacheWrite: 4.125,
468
468
  },
469
469
  contextWindow: 1000000,
470
470
  maxTokens: 64000,
@@ -2826,6 +2826,24 @@ export const MODELS = {
2826
2826
  contextWindow: 1000000,
2827
2827
  maxTokens: 128000,
2828
2828
  },
2829
+ "claude-opus-4-8": {
2830
+ id: "claude-opus-4-8",
2831
+ name: "Claude Opus 4.8",
2832
+ api: "anthropic-messages",
2833
+ provider: "cloudflare-ai-gateway",
2834
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
2835
+ reasoning: true,
2836
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
2837
+ input: ["text", "image"],
2838
+ cost: {
2839
+ input: 5,
2840
+ output: 25,
2841
+ cacheRead: 0.5,
2842
+ cacheWrite: 6.25,
2843
+ },
2844
+ contextWindow: 1000000,
2845
+ maxTokens: 128000,
2846
+ },
2829
2847
  "claude-sonnet-4": {
2830
2848
  id: "claude-sonnet-4",
2831
2849
  name: "Claude Sonnet 4 (latest)",
@@ -3303,24 +3321,6 @@ export const MODELS = {
3303
3321
  contextWindow: 128000,
3304
3322
  maxTokens: 128000,
3305
3323
  },
3306
- "@cf/moonshotai/kimi-k2.5": {
3307
- id: "@cf/moonshotai/kimi-k2.5",
3308
- name: "Kimi K2.5",
3309
- api: "openai-completions",
3310
- provider: "cloudflare-workers-ai",
3311
- baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
3312
- compat: { "sendSessionAffinityHeaders": true },
3313
- reasoning: true,
3314
- input: ["text", "image"],
3315
- cost: {
3316
- input: 0.6,
3317
- output: 3,
3318
- cacheRead: 0.1,
3319
- cacheWrite: 0,
3320
- },
3321
- contextWindow: 256000,
3322
- maxTokens: 256000,
3323
- },
3324
3324
  "@cf/moonshotai/kimi-k2.6": {
3325
3325
  id: "@cf/moonshotai/kimi-k2.6",
3326
3326
  name: "Kimi K2.6",
@@ -3693,8 +3693,8 @@ export const MODELS = {
3693
3693
  cacheRead: 0,
3694
3694
  cacheWrite: 0,
3695
3695
  },
3696
- contextWindow: 144000,
3697
- maxTokens: 32000,
3696
+ contextWindow: 200000,
3697
+ maxTokens: 64000,
3698
3698
  },
3699
3699
  "claude-opus-4.5": {
3700
3700
  id: "claude-opus-4.5",
@@ -3711,7 +3711,7 @@ export const MODELS = {
3711
3711
  cacheRead: 0,
3712
3712
  cacheWrite: 0,
3713
3713
  },
3714
- contextWindow: 160000,
3714
+ contextWindow: 200000,
3715
3715
  maxTokens: 32000,
3716
3716
  },
3717
3717
  "claude-opus-4.6": {
@@ -3731,7 +3731,7 @@ export const MODELS = {
3731
3731
  cacheWrite: 0,
3732
3732
  },
3733
3733
  contextWindow: 1000000,
3734
- maxTokens: 64000,
3734
+ maxTokens: 32000,
3735
3735
  },
3736
3736
  "claude-opus-4.7": {
3737
3737
  id: "claude-opus-4.7",
@@ -3749,7 +3749,26 @@ export const MODELS = {
3749
3749
  cacheRead: 0,
3750
3750
  cacheWrite: 0,
3751
3751
  },
3752
- contextWindow: 144000,
3752
+ contextWindow: 200000,
3753
+ maxTokens: 32000,
3754
+ },
3755
+ "claude-opus-4.8": {
3756
+ id: "claude-opus-4.8",
3757
+ name: "Claude Opus 4.8",
3758
+ api: "anthropic-messages",
3759
+ provider: "github-copilot",
3760
+ baseUrl: "https://api.individual.githubcopilot.com",
3761
+ 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" },
3762
+ reasoning: true,
3763
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
3764
+ input: ["text", "image"],
3765
+ cost: {
3766
+ input: 0,
3767
+ output: 0,
3768
+ cacheRead: 0,
3769
+ cacheWrite: 0,
3770
+ },
3771
+ contextWindow: 200000,
3753
3772
  maxTokens: 64000,
3754
3773
  },
3755
3774
  "claude-sonnet-4.5": {
@@ -3768,7 +3787,7 @@ export const MODELS = {
3768
3787
  cacheRead: 0,
3769
3788
  cacheWrite: 0,
3770
3789
  },
3771
- contextWindow: 144000,
3790
+ contextWindow: 200000,
3772
3791
  maxTokens: 32000,
3773
3792
  },
3774
3793
  "claude-sonnet-4.6": {
@@ -3843,7 +3862,7 @@ export const MODELS = {
3843
3862
  cacheRead: 0,
3844
3863
  cacheWrite: 0,
3845
3864
  },
3846
- contextWindow: 128000,
3865
+ contextWindow: 200000,
3847
3866
  maxTokens: 64000,
3848
3867
  },
3849
3868
  "gemini-3.5-flash": {
@@ -3862,7 +3881,7 @@ export const MODELS = {
3862
3881
  cacheRead: 0,
3863
3882
  cacheWrite: 0,
3864
3883
  },
3865
- contextWindow: 128000,
3884
+ contextWindow: 200000,
3866
3885
  maxTokens: 64000,
3867
3886
  },
3868
3887
  "gpt-4.1": {
@@ -3938,8 +3957,8 @@ export const MODELS = {
3938
3957
  cacheRead: 0,
3939
3958
  cacheWrite: 0,
3940
3959
  },
3941
- contextWindow: 264000,
3942
- maxTokens: 64000,
3960
+ contextWindow: 400000,
3961
+ maxTokens: 128000,
3943
3962
  },
3944
3963
  "gpt-5.2-codex": {
3945
3964
  id: "gpt-5.2-codex",
@@ -7125,6 +7144,25 @@ export const MODELS = {
7125
7144
  contextWindow: 1000000,
7126
7145
  maxTokens: 64000,
7127
7146
  },
7147
+ "deepseek-v4-flash": {
7148
+ id: "deepseek-v4-flash",
7149
+ name: "DeepSeek V4 Flash",
7150
+ api: "openai-completions",
7151
+ provider: "opencode",
7152
+ baseUrl: "https://opencode.ai/zen/v1",
7153
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7154
+ reasoning: true,
7155
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7156
+ input: ["text"],
7157
+ cost: {
7158
+ input: 0.14,
7159
+ output: 0.28,
7160
+ cacheRead: 0.03,
7161
+ cacheWrite: 0,
7162
+ },
7163
+ contextWindow: 1000000,
7164
+ maxTokens: 384000,
7165
+ },
7128
7166
  "deepseek-v4-flash-free": {
7129
7167
  id: "deepseek-v4-flash-free",
7130
7168
  name: "DeepSeek V4 Flash Free",
@@ -7585,8 +7623,8 @@ export const MODELS = {
7585
7623
  cacheRead: 0,
7586
7624
  cacheWrite: 0,
7587
7625
  },
7588
- contextWindow: 1000000,
7589
- maxTokens: 128000,
7626
+ contextWindow: 200000,
7627
+ maxTokens: 32000,
7590
7628
  },
7591
7629
  "minimax-m2.5": {
7592
7630
  id: "minimax-m2.5",
@@ -7622,6 +7660,23 @@ export const MODELS = {
7622
7660
  contextWindow: 204800,
7623
7661
  maxTokens: 131072,
7624
7662
  },
7663
+ "minimax-m3-free": {
7664
+ id: "minimax-m3-free",
7665
+ name: "MiniMax M3 Free",
7666
+ api: "anthropic-messages",
7667
+ provider: "opencode",
7668
+ baseUrl: "https://opencode.ai/zen",
7669
+ reasoning: true,
7670
+ input: ["text", "image"],
7671
+ cost: {
7672
+ input: 0,
7673
+ output: 0,
7674
+ cacheRead: 0,
7675
+ cacheWrite: 0,
7676
+ },
7677
+ contextWindow: 200000,
7678
+ maxTokens: 32000,
7679
+ },
7625
7680
  "nemotron-3-super-free": {
7626
7681
  id: "nemotron-3-super-free",
7627
7682
  name: "Nemotron 3 Super Free",
@@ -7849,23 +7904,22 @@ export const MODELS = {
7849
7904
  contextWindow: 204800,
7850
7905
  maxTokens: 131072,
7851
7906
  },
7852
- "qwen3.5-plus": {
7853
- id: "qwen3.5-plus",
7854
- name: "Qwen3.5 Plus",
7855
- api: "openai-completions",
7907
+ "minimax-m3": {
7908
+ id: "minimax-m3",
7909
+ name: "MiniMax M3",
7910
+ api: "anthropic-messages",
7856
7911
  provider: "opencode-go",
7857
- baseUrl: "https://opencode.ai/zen/go/v1",
7858
- compat: { "thinkingFormat": "qwen" },
7912
+ baseUrl: "https://opencode.ai/zen/go",
7859
7913
  reasoning: true,
7860
7914
  input: ["text", "image"],
7861
7915
  cost: {
7862
- input: 0.2,
7863
- output: 1.2,
7864
- cacheRead: 0.02,
7865
- cacheWrite: 0.25,
7916
+ input: 0.6,
7917
+ output: 2.4,
7918
+ cacheRead: 0.12,
7919
+ cacheWrite: 0,
7866
7920
  },
7867
- contextWindow: 262144,
7868
- maxTokens: 65536,
7921
+ contextWindow: 512000,
7922
+ maxTokens: 131072,
7869
7923
  },
7870
7924
  "qwen3.6-plus": {
7871
7925
  id: "qwen3.6-plus",
@@ -8318,23 +8372,6 @@ export const MODELS = {
8318
8372
  contextWindow: 2000000,
8319
8373
  maxTokens: 30000,
8320
8374
  },
8321
- "baidu/ernie-4.5-21b-a3b": {
8322
- id: "baidu/ernie-4.5-21b-a3b",
8323
- name: "Baidu: ERNIE 4.5 21B A3B",
8324
- api: "openai-completions",
8325
- provider: "openrouter",
8326
- baseUrl: "https://openrouter.ai/api/v1",
8327
- reasoning: false,
8328
- input: ["text"],
8329
- cost: {
8330
- input: 0.07,
8331
- output: 0.28,
8332
- cacheRead: 0,
8333
- cacheWrite: 0,
8334
- },
8335
- contextWindow: 131072,
8336
- maxTokens: 8000,
8337
- },
8338
8375
  "baidu/ernie-4.5-vl-28b-a3b": {
8339
8376
  id: "baidu/ernie-4.5-vl-28b-a3b",
8340
8377
  name: "Baidu: ERNIE 4.5 VL 28B A3B",
@@ -8601,32 +8638,13 @@ export const MODELS = {
8601
8638
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8602
8639
  input: ["text"],
8603
8640
  cost: {
8604
- input: 0.09999999999999999,
8605
- output: 0.19999999999999998,
8606
- cacheRead: 0.02,
8607
- cacheWrite: 0,
8608
- },
8609
- contextWindow: 1048576,
8610
- maxTokens: 16384,
8611
- },
8612
- "deepseek/deepseek-v4-flash:free": {
8613
- id: "deepseek/deepseek-v4-flash:free",
8614
- name: "DeepSeek: DeepSeek V4 Flash (free)",
8615
- api: "openai-completions",
8616
- provider: "openrouter",
8617
- baseUrl: "https://openrouter.ai/api/v1",
8618
- compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
8619
- reasoning: true,
8620
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8621
- input: ["text"],
8622
- cost: {
8623
- input: 0,
8624
- output: 0,
8625
- cacheRead: 0,
8641
+ input: 0.0983,
8642
+ output: 0.1966,
8643
+ cacheRead: 0.019700000000000002,
8626
8644
  cacheWrite: 0,
8627
8645
  },
8628
8646
  contextWindow: 1048576,
8629
- maxTokens: 384000,
8647
+ maxTokens: 131072,
8630
8648
  },
8631
8649
  "deepseek/deepseek-v4-pro": {
8632
8650
  id: "deepseek/deepseek-v4-pro",
@@ -8678,7 +8696,7 @@ export const MODELS = {
8678
8696
  cacheRead: 0.024999999999999998,
8679
8697
  cacheWrite: 0.08333333333333334,
8680
8698
  },
8681
- contextWindow: 1000000,
8699
+ contextWindow: 1048576,
8682
8700
  maxTokens: 8192,
8683
8701
  },
8684
8702
  "google/gemini-2.0-flash-lite-001": {
@@ -9081,9 +9099,9 @@ export const MODELS = {
9081
9099
  reasoning: true,
9082
9100
  input: ["text"],
9083
9101
  cost: {
9084
- input: 0.075,
9085
- output: 0.625,
9086
- cacheRead: 0.015,
9102
+ input: 0.3,
9103
+ output: 2.5,
9104
+ cacheRead: 0.06,
9087
9105
  cacheWrite: 0,
9088
9106
  },
9089
9107
  contextWindow: 262144,
@@ -9174,6 +9192,23 @@ export const MODELS = {
9174
9192
  contextWindow: 131072,
9175
9193
  maxTokens: 4096,
9176
9194
  },
9195
+ "meta-llama/llama-4-maverick": {
9196
+ id: "meta-llama/llama-4-maverick",
9197
+ name: "Meta: Llama 4 Maverick",
9198
+ api: "openai-completions",
9199
+ provider: "openrouter",
9200
+ baseUrl: "https://openrouter.ai/api/v1",
9201
+ reasoning: false,
9202
+ input: ["text", "image"],
9203
+ cost: {
9204
+ input: 0.15,
9205
+ output: 0.6,
9206
+ cacheRead: 0,
9207
+ cacheWrite: 0,
9208
+ },
9209
+ contextWindow: 1048576,
9210
+ maxTokens: 16384,
9211
+ },
9177
9212
  "meta-llama/llama-4-scout": {
9178
9213
  id: "meta-llama/llama-4-scout",
9179
9214
  name: "Meta: Llama 4 Scout",
@@ -9259,39 +9294,39 @@ export const MODELS = {
9259
9294
  contextWindow: 204800,
9260
9295
  maxTokens: 196608,
9261
9296
  },
9262
- "minimax/minimax-m2.5:free": {
9263
- id: "minimax/minimax-m2.5:free",
9264
- name: "MiniMax: MiniMax M2.5 (free)",
9297
+ "minimax/minimax-m2.7": {
9298
+ id: "minimax/minimax-m2.7",
9299
+ name: "MiniMax: MiniMax M2.7",
9265
9300
  api: "openai-completions",
9266
9301
  provider: "openrouter",
9267
9302
  baseUrl: "https://openrouter.ai/api/v1",
9268
9303
  reasoning: true,
9269
9304
  input: ["text"],
9270
9305
  cost: {
9271
- input: 0,
9272
- output: 0,
9306
+ input: 0.26,
9307
+ output: 1.2,
9273
9308
  cacheRead: 0,
9274
9309
  cacheWrite: 0,
9275
9310
  },
9276
- contextWindow: 262144,
9277
- maxTokens: 8192,
9311
+ contextWindow: 204800,
9312
+ maxTokens: 4096,
9278
9313
  },
9279
- "minimax/minimax-m2.7": {
9280
- id: "minimax/minimax-m2.7",
9281
- name: "MiniMax: MiniMax M2.7",
9314
+ "minimax/minimax-m3": {
9315
+ id: "minimax/minimax-m3",
9316
+ name: "MiniMax: MiniMax M3",
9282
9317
  api: "openai-completions",
9283
9318
  provider: "openrouter",
9284
9319
  baseUrl: "https://openrouter.ai/api/v1",
9285
9320
  reasoning: true,
9286
- input: ["text"],
9321
+ input: ["text", "image"],
9287
9322
  cost: {
9288
- input: 0.27899999999999997,
9323
+ input: 0.3,
9289
9324
  output: 1.2,
9290
- cacheRead: 0,
9325
+ cacheRead: 0.06,
9291
9326
  cacheWrite: 0,
9292
9327
  },
9293
- contextWindow: 204800,
9294
- maxTokens: 131072,
9328
+ contextWindow: 1048576,
9329
+ maxTokens: 512000,
9295
9330
  },
9296
9331
  "mistralai/codestral-2508": {
9297
9332
  id: "mistralai/codestral-2508",
@@ -9327,68 +9362,34 @@ export const MODELS = {
9327
9362
  contextWindow: 262144,
9328
9363
  maxTokens: 4096,
9329
9364
  },
9330
- "mistralai/devstral-medium": {
9331
- id: "mistralai/devstral-medium",
9332
- name: "Mistral: Devstral Medium",
9365
+ "mistralai/ministral-14b-2512": {
9366
+ id: "mistralai/ministral-14b-2512",
9367
+ name: "Mistral: Ministral 3 14B 2512",
9333
9368
  api: "openai-completions",
9334
9369
  provider: "openrouter",
9335
9370
  baseUrl: "https://openrouter.ai/api/v1",
9336
9371
  reasoning: false,
9337
- input: ["text"],
9372
+ input: ["text", "image"],
9338
9373
  cost: {
9339
- input: 0.39999999999999997,
9340
- output: 2,
9341
- cacheRead: 0.04,
9374
+ input: 0.19999999999999998,
9375
+ output: 0.19999999999999998,
9376
+ cacheRead: 0.02,
9342
9377
  cacheWrite: 0,
9343
9378
  },
9344
- contextWindow: 131072,
9379
+ contextWindow: 262144,
9345
9380
  maxTokens: 4096,
9346
9381
  },
9347
- "mistralai/devstral-small": {
9348
- id: "mistralai/devstral-small",
9349
- name: "Mistral: Devstral Small 1.1",
9382
+ "mistralai/ministral-3b-2512": {
9383
+ id: "mistralai/ministral-3b-2512",
9384
+ name: "Mistral: Ministral 3 3B 2512",
9350
9385
  api: "openai-completions",
9351
9386
  provider: "openrouter",
9352
9387
  baseUrl: "https://openrouter.ai/api/v1",
9353
9388
  reasoning: false,
9354
- input: ["text"],
9389
+ input: ["text", "image"],
9355
9390
  cost: {
9356
9391
  input: 0.09999999999999999,
9357
- output: 0.3,
9358
- cacheRead: 0.01,
9359
- cacheWrite: 0,
9360
- },
9361
- contextWindow: 131072,
9362
- maxTokens: 4096,
9363
- },
9364
- "mistralai/ministral-14b-2512": {
9365
- id: "mistralai/ministral-14b-2512",
9366
- name: "Mistral: Ministral 3 14B 2512",
9367
- api: "openai-completions",
9368
- provider: "openrouter",
9369
- baseUrl: "https://openrouter.ai/api/v1",
9370
- reasoning: false,
9371
- input: ["text", "image"],
9372
- cost: {
9373
- input: 0.19999999999999998,
9374
- output: 0.19999999999999998,
9375
- cacheRead: 0.02,
9376
- cacheWrite: 0,
9377
- },
9378
- contextWindow: 262144,
9379
- maxTokens: 4096,
9380
- },
9381
- "mistralai/ministral-3b-2512": {
9382
- id: "mistralai/ministral-3b-2512",
9383
- name: "Mistral: Ministral 3 3B 2512",
9384
- api: "openai-completions",
9385
- provider: "openrouter",
9386
- baseUrl: "https://openrouter.ai/api/v1",
9387
- reasoning: false,
9388
- input: ["text", "image"],
9389
- cost: {
9390
- input: 0.09999999999999999,
9391
- output: 0.09999999999999999,
9392
+ output: 0.09999999999999999,
9392
9393
  cacheRead: 0.01,
9393
9394
  cacheWrite: 0,
9394
9395
  },
@@ -9446,23 +9447,6 @@ export const MODELS = {
9446
9447
  contextWindow: 131072,
9447
9448
  maxTokens: 4096,
9448
9449
  },
9449
- "mistralai/mistral-large-2411": {
9450
- id: "mistralai/mistral-large-2411",
9451
- name: "Mistral Large 2411",
9452
- api: "openai-completions",
9453
- provider: "openrouter",
9454
- baseUrl: "https://openrouter.ai/api/v1",
9455
- reasoning: false,
9456
- input: ["text"],
9457
- cost: {
9458
- input: 2,
9459
- output: 6,
9460
- cacheRead: 0.19999999999999998,
9461
- cacheWrite: 0,
9462
- },
9463
- contextWindow: 131072,
9464
- maxTokens: 4096,
9465
- },
9466
9450
  "mistralai/mistral-large-2512": {
9467
9451
  id: "mistralai/mistral-large-2512",
9468
9452
  name: "Mistral: Mistral Large 3 2512",
@@ -9616,23 +9600,6 @@ export const MODELS = {
9616
9600
  contextWindow: 65536,
9617
9601
  maxTokens: 4096,
9618
9602
  },
9619
- "mistralai/pixtral-large-2411": {
9620
- id: "mistralai/pixtral-large-2411",
9621
- name: "Mistral: Pixtral Large 2411",
9622
- api: "openai-completions",
9623
- provider: "openrouter",
9624
- baseUrl: "https://openrouter.ai/api/v1",
9625
- reasoning: false,
9626
- input: ["text", "image"],
9627
- cost: {
9628
- input: 2,
9629
- output: 6,
9630
- cacheRead: 0.19999999999999998,
9631
- cacheWrite: 0,
9632
- },
9633
- contextWindow: 131072,
9634
- maxTokens: 4096,
9635
- },
9636
9603
  "mistralai/voxtral-small-24b-2507": {
9637
9604
  id: "mistralai/voxtral-small-24b-2507",
9638
9605
  name: "Mistral: Voxtral Small 24B 2507",
@@ -9727,13 +9694,13 @@ export const MODELS = {
9727
9694
  reasoning: true,
9728
9695
  input: ["text", "image"],
9729
9696
  cost: {
9730
- input: 0.73,
9731
- output: 3.49,
9732
- cacheRead: 0.25,
9697
+ input: 0.684,
9698
+ output: 3.42,
9699
+ cacheRead: 0.144,
9733
9700
  cacheWrite: 0,
9734
9701
  },
9735
9702
  contextWindow: 262144,
9736
- maxTokens: 262142,
9703
+ maxTokens: 262144,
9737
9704
  },
9738
9705
  "moonshotai/kimi-k2.6:free": {
9739
9706
  id: "moonshotai/kimi-k2.6:free",
@@ -10092,23 +10059,6 @@ export const MODELS = {
10092
10059
  contextWindow: 128000,
10093
10060
  maxTokens: 16384,
10094
10061
  },
10095
- "openai/gpt-4o-audio-preview": {
10096
- id: "openai/gpt-4o-audio-preview",
10097
- name: "OpenAI: GPT-4o Audio",
10098
- api: "openai-completions",
10099
- provider: "openrouter",
10100
- baseUrl: "https://openrouter.ai/api/v1",
10101
- reasoning: false,
10102
- input: ["text"],
10103
- cost: {
10104
- input: 2.5,
10105
- output: 10,
10106
- cacheRead: 0,
10107
- cacheWrite: 0,
10108
- },
10109
- contextWindow: 128000,
10110
- maxTokens: 16384,
10111
- },
10112
10062
  "openai/gpt-4o-mini": {
10113
10063
  id: "openai/gpt-4o-mini",
10114
10064
  name: "OpenAI: GPT-4o-mini",
@@ -10606,13 +10556,13 @@ export const MODELS = {
10606
10556
  reasoning: true,
10607
10557
  input: ["text"],
10608
10558
  cost: {
10609
- input: 0.03,
10559
+ input: 0.029,
10610
10560
  output: 0.14,
10611
10561
  cacheRead: 0,
10612
10562
  cacheWrite: 0,
10613
10563
  },
10614
10564
  contextWindow: 131072,
10615
- maxTokens: 131072,
10565
+ maxTokens: 4096,
10616
10566
  },
10617
10567
  "openai/gpt-oss-20b:free": {
10618
10568
  id: "openai/gpt-oss-20b:free",
@@ -10920,23 +10870,6 @@ export const MODELS = {
10920
10870
  contextWindow: 131072,
10921
10871
  maxTokens: 16384,
10922
10872
  },
10923
- "qwen/qwen-2.5-7b-instruct": {
10924
- id: "qwen/qwen-2.5-7b-instruct",
10925
- name: "Qwen: Qwen2.5 7B Instruct",
10926
- api: "openai-completions",
10927
- provider: "openrouter",
10928
- baseUrl: "https://openrouter.ai/api/v1",
10929
- reasoning: false,
10930
- input: ["text"],
10931
- cost: {
10932
- input: 0.04,
10933
- output: 0.09999999999999999,
10934
- cacheRead: 0,
10935
- cacheWrite: 0,
10936
- },
10937
- contextWindow: 131072,
10938
- maxTokens: 32768,
10939
- },
10940
10873
  "qwen/qwen-plus": {
10941
10874
  id: "qwen/qwen-plus",
10942
10875
  name: "Qwen: Qwen-Plus",
@@ -11048,13 +10981,13 @@ export const MODELS = {
11048
10981
  reasoning: true,
11049
10982
  input: ["text"],
11050
10983
  cost: {
11051
- input: 0.14950000000000002,
11052
- output: 1.495,
11053
- cacheRead: 0,
10984
+ input: 0.09999999999999999,
10985
+ output: 0.09999999999999999,
10986
+ cacheRead: 0.09999999999999999,
11054
10987
  cacheWrite: 0,
11055
10988
  },
11056
10989
  contextWindow: 262144,
11057
- maxTokens: 4096,
10990
+ maxTokens: 262144,
11058
10991
  },
11059
10992
  "qwen/qwen3-30b-a3b": {
11060
10993
  id: "qwen/qwen3-30b-a3b",
@@ -11490,13 +11423,13 @@ export const MODELS = {
11490
11423
  reasoning: true,
11491
11424
  input: ["text", "image"],
11492
11425
  cost: {
11493
- input: 0.13899999999999998,
11426
+ input: 0.14,
11494
11427
  output: 1,
11495
- cacheRead: 0,
11428
+ cacheRead: 0.049999999999999996,
11496
11429
  cacheWrite: 0,
11497
11430
  },
11498
11431
  contextWindow: 262144,
11499
- maxTokens: 4096,
11432
+ maxTokens: 262144,
11500
11433
  },
11501
11434
  "qwen/qwen3.5-397b-a17b": {
11502
11435
  id: "qwen/qwen3.5-397b-a17b",
@@ -11770,6 +11703,23 @@ export const MODELS = {
11770
11703
  contextWindow: 262144,
11771
11704
  maxTokens: 16384,
11772
11705
  },
11706
+ "stepfun/step-3.7-flash": {
11707
+ id: "stepfun/step-3.7-flash",
11708
+ name: "StepFun: Step 3.7 Flash",
11709
+ api: "openai-completions",
11710
+ provider: "openrouter",
11711
+ baseUrl: "https://openrouter.ai/api/v1",
11712
+ reasoning: true,
11713
+ input: ["text", "image"],
11714
+ cost: {
11715
+ input: 0.19999999999999998,
11716
+ output: 1.15,
11717
+ cacheRead: 0.04,
11718
+ cacheWrite: 0,
11719
+ },
11720
+ contextWindow: 256000,
11721
+ maxTokens: 256000,
11722
+ },
11773
11723
  "thedrummer/rocinante-12b": {
11774
11724
  id: "thedrummer/rocinante-12b",
11775
11725
  name: "TheDrummer: Rocinante 12B",
@@ -12074,7 +12024,7 @@ export const MODELS = {
12074
12024
  cacheWrite: 0,
12075
12025
  },
12076
12026
  contextWindow: 202752,
12077
- maxTokens: 4096,
12027
+ maxTokens: 16384,
12078
12028
  },
12079
12029
  "z-ai/glm-5-turbo": {
12080
12030
  id: "z-ai/glm-5-turbo",
@@ -12108,7 +12058,7 @@ export const MODELS = {
12108
12058
  cacheWrite: 0,
12109
12059
  },
12110
12060
  contextWindow: 202752,
12111
- maxTokens: 4096,
12061
+ maxTokens: 131072,
12112
12062
  },
12113
12063
  "z-ai/glm-5v-turbo": {
12114
12064
  id: "z-ai/glm-5v-turbo",
@@ -12221,13 +12171,13 @@ export const MODELS = {
12221
12171
  reasoning: true,
12222
12172
  input: ["text", "image"],
12223
12173
  cost: {
12224
- input: 0.73,
12225
- output: 3.49,
12226
- cacheRead: 0.25,
12174
+ input: 0.684,
12175
+ output: 3.42,
12176
+ cacheRead: 0.144,
12227
12177
  cacheWrite: 0,
12228
12178
  },
12229
12179
  contextWindow: 262144,
12230
- maxTokens: 262142,
12180
+ maxTokens: 262144,
12231
12181
  },
12232
12182
  "~openai/gpt-latest": {
12233
12183
  id: "~openai/gpt-latest",
@@ -12625,20 +12575,20 @@ export const MODELS = {
12625
12575
  },
12626
12576
  "alibaba/qwen-3-235b": {
12627
12577
  id: "alibaba/qwen-3-235b",
12628
- name: "Qwen3 235B A22b Instruct 2507",
12578
+ name: "Qwen3 235B A22B",
12629
12579
  api: "anthropic-messages",
12630
12580
  provider: "vercel-ai-gateway",
12631
12581
  baseUrl: "https://ai-gateway.vercel.sh",
12632
- reasoning: false,
12582
+ reasoning: true,
12633
12583
  input: ["text"],
12634
12584
  cost: {
12635
- input: 0.6,
12636
- output: 1.2,
12637
- cacheRead: 0.6,
12585
+ input: 0.22,
12586
+ output: 0.88,
12587
+ cacheRead: 0,
12638
12588
  cacheWrite: 0,
12639
12589
  },
12640
- contextWindow: 131000,
12641
- maxTokens: 40000,
12590
+ contextWindow: 262144,
12591
+ maxTokens: 16384,
12642
12592
  },
12643
12593
  "alibaba/qwen-3-30b": {
12644
12594
  id: "alibaba/qwen-3-30b",
@@ -12714,7 +12664,7 @@ export const MODELS = {
12714
12664
  api: "anthropic-messages",
12715
12665
  provider: "vercel-ai-gateway",
12716
12666
  baseUrl: "https://ai-gateway.vercel.sh",
12717
- reasoning: false,
12667
+ reasoning: true,
12718
12668
  input: ["text"],
12719
12669
  cost: {
12720
12670
  input: 1.5,
@@ -12748,7 +12698,7 @@ export const MODELS = {
12748
12698
  api: "anthropic-messages",
12749
12699
  provider: "vercel-ai-gateway",
12750
12700
  baseUrl: "https://ai-gateway.vercel.sh",
12751
- reasoning: false,
12701
+ reasoning: true,
12752
12702
  input: ["text"],
12753
12703
  cost: {
12754
12704
  input: 0.5,
@@ -12827,6 +12777,40 @@ export const MODELS = {
12827
12777
  contextWindow: 256000,
12828
12778
  maxTokens: 65536,
12829
12779
  },
12780
+ "alibaba/qwen3-next-80b-a3b-instruct": {
12781
+ id: "alibaba/qwen3-next-80b-a3b-instruct",
12782
+ name: "Qwen3 Next 80B A3B Instruct",
12783
+ api: "anthropic-messages",
12784
+ provider: "vercel-ai-gateway",
12785
+ baseUrl: "https://ai-gateway.vercel.sh",
12786
+ reasoning: false,
12787
+ input: ["text"],
12788
+ cost: {
12789
+ input: 0.15,
12790
+ output: 1.2,
12791
+ cacheRead: 0,
12792
+ cacheWrite: 0,
12793
+ },
12794
+ contextWindow: 131072,
12795
+ maxTokens: 32768,
12796
+ },
12797
+ "alibaba/qwen3-next-80b-a3b-thinking": {
12798
+ id: "alibaba/qwen3-next-80b-a3b-thinking",
12799
+ name: "Qwen3 Next 80B A3B Thinking",
12800
+ api: "anthropic-messages",
12801
+ provider: "vercel-ai-gateway",
12802
+ baseUrl: "https://ai-gateway.vercel.sh",
12803
+ reasoning: true,
12804
+ input: ["text"],
12805
+ cost: {
12806
+ input: 0.15,
12807
+ output: 1.2,
12808
+ cacheRead: 0,
12809
+ cacheWrite: 0,
12810
+ },
12811
+ contextWindow: 131072,
12812
+ maxTokens: 32768,
12813
+ },
12830
12814
  "alibaba/qwen3-vl-thinking": {
12831
12815
  id: "alibaba/qwen3-vl-thinking",
12832
12816
  name: "Qwen3 VL 235B A22B Thinking",
@@ -13213,17 +13197,17 @@ export const MODELS = {
13213
13197
  reasoning: false,
13214
13198
  input: ["text"],
13215
13199
  cost: {
13216
- input: 0.77,
13217
- output: 0.77,
13218
- cacheRead: 0,
13200
+ input: 0.27,
13201
+ output: 1.12,
13202
+ cacheRead: 0.135,
13219
13203
  cacheWrite: 0,
13220
13204
  },
13221
13205
  contextWindow: 163840,
13222
- maxTokens: 16384,
13206
+ maxTokens: 163840,
13223
13207
  },
13224
13208
  "deepseek/deepseek-v3.1": {
13225
13209
  id: "deepseek/deepseek-v3.1",
13226
- name: "DeepSeek-V3.1",
13210
+ name: "DeepSeek V3.1",
13227
13211
  api: "anthropic-messages",
13228
13212
  provider: "vercel-ai-gateway",
13229
13213
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -13261,8 +13245,8 @@ export const MODELS = {
13261
13245
  api: "anthropic-messages",
13262
13246
  provider: "vercel-ai-gateway",
13263
13247
  baseUrl: "https://ai-gateway.vercel.sh",
13264
- reasoning: false,
13265
- input: ["text"],
13248
+ reasoning: true,
13249
+ input: ["text", "image"],
13266
13250
  cost: {
13267
13251
  input: 0.28,
13268
13252
  output: 0.42,
@@ -13278,8 +13262,8 @@ export const MODELS = {
13278
13262
  api: "anthropic-messages",
13279
13263
  provider: "vercel-ai-gateway",
13280
13264
  baseUrl: "https://ai-gateway.vercel.sh",
13281
- reasoning: false,
13282
- input: ["text"],
13265
+ reasoning: true,
13266
+ input: ["text", "image"],
13283
13267
  cost: {
13284
13268
  input: 0.62,
13285
13269
  output: 1.85,
@@ -13296,7 +13280,7 @@ export const MODELS = {
13296
13280
  provider: "vercel-ai-gateway",
13297
13281
  baseUrl: "https://ai-gateway.vercel.sh",
13298
13282
  reasoning: true,
13299
- input: ["text"],
13283
+ input: ["text", "image"],
13300
13284
  cost: {
13301
13285
  input: 0.14,
13302
13286
  output: 0.28,
@@ -13313,7 +13297,7 @@ export const MODELS = {
13313
13297
  provider: "vercel-ai-gateway",
13314
13298
  baseUrl: "https://ai-gateway.vercel.sh",
13315
13299
  reasoning: true,
13316
- input: ["text"],
13300
+ input: ["text", "image"],
13317
13301
  cost: {
13318
13302
  input: 0.435,
13319
13303
  output: 0.87,
@@ -13806,7 +13790,7 @@ export const MODELS = {
13806
13790
  provider: "vercel-ai-gateway",
13807
13791
  baseUrl: "https://ai-gateway.vercel.sh",
13808
13792
  reasoning: true,
13809
- input: ["text"],
13793
+ input: ["text", "image"],
13810
13794
  cost: {
13811
13795
  input: 0.6,
13812
13796
  output: 2.4,
@@ -13850,6 +13834,23 @@ export const MODELS = {
13850
13834
  contextWindow: 204800,
13851
13835
  maxTokens: 131100,
13852
13836
  },
13837
+ "minimax/minimax-m3": {
13838
+ id: "minimax/minimax-m3",
13839
+ name: "MiniMax M3",
13840
+ api: "anthropic-messages",
13841
+ provider: "vercel-ai-gateway",
13842
+ baseUrl: "https://ai-gateway.vercel.sh",
13843
+ reasoning: true,
13844
+ input: ["text", "image"],
13845
+ cost: {
13846
+ input: 0.3,
13847
+ output: 1.2,
13848
+ cacheRead: 0.06,
13849
+ cacheWrite: 0,
13850
+ },
13851
+ contextWindow: 1000000,
13852
+ maxTokens: 1000000,
13853
+ },
13853
13854
  "mistral/codestral": {
13854
13855
  id: "mistral/codestral",
13855
13856
  name: "Mistral Codestral",
@@ -13986,6 +13987,23 @@ export const MODELS = {
13986
13987
  contextWindow: 256000,
13987
13988
  maxTokens: 256000,
13988
13989
  },
13990
+ "mistral/mistral-nemo": {
13991
+ id: "mistral/mistral-nemo",
13992
+ name: "Mistral Nemo 12B",
13993
+ api: "anthropic-messages",
13994
+ provider: "vercel-ai-gateway",
13995
+ baseUrl: "https://ai-gateway.vercel.sh",
13996
+ reasoning: false,
13997
+ input: ["text"],
13998
+ cost: {
13999
+ input: 0.02,
14000
+ output: 0.04,
14001
+ cacheRead: 0,
14002
+ cacheWrite: 0,
14003
+ },
14004
+ contextWindow: 131072,
14005
+ maxTokens: 131072,
14006
+ },
13989
14007
  "mistral/mistral-small": {
13990
14008
  id: "mistral/mistral-small",
13991
14009
  name: "Mistral Small",
@@ -14139,6 +14157,23 @@ export const MODELS = {
14139
14157
  contextWindow: 262000,
14140
14158
  maxTokens: 262000,
14141
14159
  },
14160
+ "nvidia/nemotron-3-super-120b-a12b": {
14161
+ id: "nvidia/nemotron-3-super-120b-a12b",
14162
+ name: "NVIDIA Nemotron 3 Super 120B A12B",
14163
+ api: "anthropic-messages",
14164
+ provider: "vercel-ai-gateway",
14165
+ baseUrl: "https://ai-gateway.vercel.sh",
14166
+ reasoning: true,
14167
+ input: ["text"],
14168
+ cost: {
14169
+ input: 0.15,
14170
+ output: 0.65,
14171
+ cacheRead: 0,
14172
+ cacheWrite: 0,
14173
+ },
14174
+ contextWindow: 256000,
14175
+ maxTokens: 32000,
14176
+ },
14142
14177
  "nvidia/nemotron-nano-12b-v2-vl": {
14143
14178
  id: "nvidia/nemotron-nano-12b-v2-vl",
14144
14179
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -14299,7 +14334,7 @@ export const MODELS = {
14299
14334
  provider: "vercel-ai-gateway",
14300
14335
  baseUrl: "https://ai-gateway.vercel.sh",
14301
14336
  reasoning: true,
14302
- input: ["text"],
14337
+ input: ["text", "image"],
14303
14338
  cost: {
14304
14339
  input: 1.25,
14305
14340
  output: 10,
@@ -14661,6 +14696,23 @@ export const MODELS = {
14661
14696
  contextWindow: 1000000,
14662
14697
  maxTokens: 128000,
14663
14698
  },
14699
+ "openai/gpt-oss-120b": {
14700
+ id: "openai/gpt-oss-120b",
14701
+ name: "GPT OSS 120B",
14702
+ api: "anthropic-messages",
14703
+ provider: "vercel-ai-gateway",
14704
+ baseUrl: "https://ai-gateway.vercel.sh",
14705
+ reasoning: true,
14706
+ input: ["text"],
14707
+ cost: {
14708
+ input: 0.35,
14709
+ output: 0.75,
14710
+ cacheRead: 0.25,
14711
+ cacheWrite: 0,
14712
+ },
14713
+ contextWindow: 131072,
14714
+ maxTokens: 131000,
14715
+ },
14664
14716
  "openai/gpt-oss-20b": {
14665
14717
  id: "openai/gpt-oss-20b",
14666
14718
  name: "GPT OSS 20B",
@@ -14831,6 +14883,40 @@ export const MODELS = {
14831
14883
  contextWindow: 200000,
14832
14884
  maxTokens: 8000,
14833
14885
  },
14886
+ "stepfun/step-3.5-flash": {
14887
+ id: "stepfun/step-3.5-flash",
14888
+ name: "StepFun 3.5 Flash",
14889
+ api: "anthropic-messages",
14890
+ provider: "vercel-ai-gateway",
14891
+ baseUrl: "https://ai-gateway.vercel.sh",
14892
+ reasoning: true,
14893
+ input: ["text"],
14894
+ cost: {
14895
+ input: 0.09,
14896
+ output: 0.3,
14897
+ cacheRead: 0,
14898
+ cacheWrite: 0.02,
14899
+ },
14900
+ contextWindow: 262114,
14901
+ maxTokens: 262114,
14902
+ },
14903
+ "stepfun/step-3.7-flash": {
14904
+ id: "stepfun/step-3.7-flash",
14905
+ name: "Step 3.7 Flash",
14906
+ api: "anthropic-messages",
14907
+ provider: "vercel-ai-gateway",
14908
+ baseUrl: "https://ai-gateway.vercel.sh",
14909
+ reasoning: true,
14910
+ input: ["text", "image"],
14911
+ cost: {
14912
+ input: 0.19999999999999998,
14913
+ output: 1.15,
14914
+ cacheRead: 0.04,
14915
+ cacheWrite: 0,
14916
+ },
14917
+ contextWindow: 256000,
14918
+ maxTokens: 256000,
14919
+ },
14834
14920
  "xai/grok-4.1-fast-non-reasoning": {
14835
14921
  id: "xai/grok-4.1-fast-non-reasoning",
14836
14922
  name: "Grok 4.1 Fast Non-Reasoning",
@@ -15075,7 +15161,7 @@ export const MODELS = {
15075
15161
  api: "anthropic-messages",
15076
15162
  provider: "vercel-ai-gateway",
15077
15163
  baseUrl: "https://ai-gateway.vercel.sh",
15078
- reasoning: false,
15164
+ reasoning: true,
15079
15165
  input: ["text", "image"],
15080
15166
  cost: {
15081
15167
  input: 0.6,
@@ -15229,7 +15315,7 @@ export const MODELS = {
15229
15315
  provider: "vercel-ai-gateway",
15230
15316
  baseUrl: "https://ai-gateway.vercel.sh",
15231
15317
  reasoning: true,
15232
- input: ["text"],
15318
+ input: ["text", "image"],
15233
15319
  cost: {
15234
15320
  input: 1.4,
15235
15321
  output: 4.4,