@fleetagent/pi-ai 0.2.0 → 0.2.2

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.
@@ -862,6 +862,23 @@ export const MODELS = {
862
862
  contextWindow: 1050000,
863
863
  maxTokens: 128000,
864
864
  },
865
+ "global.xai.grok-4.6": {
866
+ id: "global.xai.grok-4.6",
867
+ name: "Grok 4.6 (Global)",
868
+ api: "bedrock-converse-stream",
869
+ provider: "amazon-bedrock",
870
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
871
+ reasoning: true,
872
+ input: ["text", "image"],
873
+ cost: {
874
+ input: 2,
875
+ output: 6,
876
+ cacheRead: 0.5,
877
+ cacheWrite: 0,
878
+ },
879
+ contextWindow: 500000,
880
+ maxTokens: 500000,
881
+ },
865
882
  "google.gemma-3-27b-it": {
866
883
  id: "google.gemma-3-27b-it",
867
884
  name: "Google Gemma 3 27B Instruct",
@@ -1975,6 +1992,23 @@ export const MODELS = {
1975
1992
  contextWindow: 3500000,
1976
1993
  maxTokens: 16384,
1977
1994
  },
1995
+ "us.xai.grok-4.6": {
1996
+ id: "us.xai.grok-4.6",
1997
+ name: "Grok 4.6 (US)",
1998
+ api: "bedrock-converse-stream",
1999
+ provider: "amazon-bedrock",
2000
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2001
+ reasoning: true,
2002
+ input: ["text", "image"],
2003
+ cost: {
2004
+ input: 2.2,
2005
+ output: 6.6,
2006
+ cacheRead: 0.55,
2007
+ cacheWrite: 0,
2008
+ },
2009
+ contextWindow: 500000,
2010
+ maxTokens: 500000,
2011
+ },
1978
2012
  "writer.palmyra-x4-v1:0": {
1979
2013
  id: "writer.palmyra-x4-v1:0",
1980
2014
  name: "Palmyra X4",
@@ -2944,6 +2978,24 @@ export const MODELS = {
2944
2978
  contextWindow: 1050000,
2945
2979
  maxTokens: 128000,
2946
2980
  },
2981
+ "gpt-6-astra": {
2982
+ id: "gpt-6-astra",
2983
+ name: "GPT-6 Astra",
2984
+ api: "azure-openai-responses",
2985
+ provider: "azure-openai-responses",
2986
+ baseUrl: "",
2987
+ reasoning: true,
2988
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
2989
+ input: ["text", "image"],
2990
+ cost: {
2991
+ input: 10,
2992
+ output: 50,
2993
+ cacheRead: 1,
2994
+ cacheWrite: 12.5,
2995
+ },
2996
+ contextWindow: 1050000,
2997
+ maxTokens: 128000,
2998
+ },
2947
2999
  "gpt-realtime-2.1": {
2948
3000
  id: "gpt-realtime-2.1",
2949
3001
  name: "GPT-Realtime-2.1",
@@ -3065,39 +3117,39 @@ export const MODELS = {
3065
3117
  },
3066
3118
  },
3067
3119
  "cerebras": {
3068
- "gemma-4-31b": {
3069
- id: "gemma-4-31b",
3070
- name: "Gemma 4 31B IT",
3120
+ "gpt-oss-120b": {
3121
+ id: "gpt-oss-120b",
3122
+ name: "GPT OSS 120B",
3071
3123
  api: "openai-completions",
3072
3124
  provider: "cerebras",
3073
3125
  baseUrl: "https://api.cerebras.ai/v1",
3074
3126
  reasoning: true,
3075
- input: ["text", "image"],
3127
+ input: ["text"],
3076
3128
  cost: {
3077
- input: 0.99,
3078
- output: 1.49,
3129
+ input: 0.35,
3130
+ output: 0.75,
3079
3131
  cacheRead: 0,
3080
3132
  cacheWrite: 0,
3081
3133
  },
3082
3134
  contextWindow: 131072,
3083
3135
  maxTokens: 40960,
3084
3136
  },
3085
- "gpt-oss-120b": {
3086
- id: "gpt-oss-120b",
3087
- name: "GPT OSS 120B",
3137
+ "qwen-3.8-27b": {
3138
+ id: "qwen-3.8-27b",
3139
+ name: "Qwen3.8 27B",
3088
3140
  api: "openai-completions",
3089
3141
  provider: "cerebras",
3090
3142
  baseUrl: "https://api.cerebras.ai/v1",
3091
3143
  reasoning: true,
3092
- input: ["text"],
3144
+ input: ["text", "image"],
3093
3145
  cost: {
3094
- input: 0.35,
3095
- output: 0.75,
3146
+ input: 0.99,
3147
+ output: 1.49,
3096
3148
  cacheRead: 0,
3097
3149
  cacheWrite: 0,
3098
3150
  },
3099
- contextWindow: 131072,
3100
- maxTokens: 40960,
3151
+ contextWindow: 65536,
3152
+ maxTokens: 32768,
3101
3153
  },
3102
3154
  },
3103
3155
  "cloudflare-ai-gateway": {
@@ -3118,6 +3170,23 @@ export const MODELS = {
3118
3170
  contextWindow: 1000000,
3119
3171
  maxTokens: 128000,
3120
3172
  },
3173
+ "claude-fable-5.1": {
3174
+ id: "claude-fable-5.1",
3175
+ name: "Claude Fable 5.1",
3176
+ api: "anthropic-messages",
3177
+ provider: "cloudflare-ai-gateway",
3178
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3179
+ reasoning: true,
3180
+ input: ["text", "image"],
3181
+ cost: {
3182
+ input: 10,
3183
+ output: 50,
3184
+ cacheRead: 0.25,
3185
+ cacheWrite: 12.5,
3186
+ },
3187
+ contextWindow: 1000000,
3188
+ maxTokens: 128000,
3189
+ },
3121
3190
  "claude-haiku-4.5": {
3122
3191
  id: "claude-haiku-4.5",
3123
3192
  name: "Claude Haiku 4.5 (latest)",
@@ -4053,6 +4122,24 @@ export const MODELS = {
4053
4122
  contextWindow: 1000000,
4054
4123
  maxTokens: 384000,
4055
4124
  },
4125
+ "accounts/fireworks/models/deepseek-v4-flash-vision-exp": {
4126
+ id: "accounts/fireworks/models/deepseek-v4-flash-vision-exp",
4127
+ name: "DeepSeek V4 Flash Vision Exp",
4128
+ api: "anthropic-messages",
4129
+ provider: "fireworks",
4130
+ baseUrl: "https://api.fireworks.ai/inference",
4131
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4132
+ reasoning: true,
4133
+ input: ["text", "image"],
4134
+ cost: {
4135
+ input: 0.22,
4136
+ output: 0.66,
4137
+ cacheRead: 0.007,
4138
+ cacheWrite: 0,
4139
+ },
4140
+ contextWindow: 1000000,
4141
+ maxTokens: 384000,
4142
+ },
4056
4143
  "accounts/fireworks/models/deepseek-v4-pro-0813": {
4057
4144
  id: "accounts/fireworks/models/deepseek-v4-pro-0813",
4058
4145
  name: "DeepSeek V4 Pro 0813",
@@ -4104,8 +4191,8 @@ export const MODELS = {
4104
4191
  cacheRead: 0.26,
4105
4192
  cacheWrite: 0,
4106
4193
  },
4107
- contextWindow: 1000000,
4108
- maxTokens: 131072,
4194
+ contextWindow: 1048573,
4195
+ maxTokens: 262144,
4109
4196
  },
4110
4197
  "accounts/fireworks/models/glm-5p3-flash": {
4111
4198
  id: "accounts/fireworks/models/glm-5p3-flash",
@@ -4119,10 +4206,10 @@ export const MODELS = {
4119
4206
  cost: {
4120
4207
  input: 0.15,
4121
4208
  output: 0.5,
4122
- cacheRead: 0.029,
4209
+ cacheRead: 0.03,
4123
4210
  cacheWrite: 0,
4124
4211
  },
4125
- contextWindow: 1000000,
4212
+ contextWindow: 1048573,
4126
4213
  maxTokens: 131072,
4127
4214
  },
4128
4215
  "accounts/fireworks/models/gpt-oss-120b": {
@@ -4305,6 +4392,24 @@ export const MODELS = {
4305
4392
  contextWindow: 262144,
4306
4393
  maxTokens: 65536,
4307
4394
  },
4395
+ "accounts/fireworks/models/qwen3p8-2p4t-a95b": {
4396
+ id: "accounts/fireworks/models/qwen3p8-2p4t-a95b",
4397
+ name: "Qwen3.8 2.4T A95B",
4398
+ api: "anthropic-messages",
4399
+ provider: "fireworks",
4400
+ baseUrl: "https://api.fireworks.ai/inference",
4401
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4402
+ reasoning: true,
4403
+ input: ["text"],
4404
+ cost: {
4405
+ input: 2,
4406
+ output: 6,
4407
+ cacheRead: 0.25,
4408
+ cacheWrite: 0,
4409
+ },
4410
+ contextWindow: 262144,
4411
+ maxTokens: 131072,
4412
+ },
4308
4413
  "accounts/fireworks/models/qwen3p8-max": {
4309
4414
  id: "accounts/fireworks/models/qwen3p8-max",
4310
4415
  name: "Qwen3.8 Max",
@@ -4341,6 +4446,24 @@ export const MODELS = {
4341
4446
  contextWindow: 1048575,
4342
4447
  maxTokens: 131072,
4343
4448
  },
4449
+ "accounts/fireworks/routers/glm-5p3-fast": {
4450
+ id: "accounts/fireworks/routers/glm-5p3-fast",
4451
+ name: "GLM 5.3 Fast",
4452
+ api: "anthropic-messages",
4453
+ provider: "fireworks",
4454
+ baseUrl: "https://api.fireworks.ai/inference",
4455
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4456
+ reasoning: true,
4457
+ input: ["text"],
4458
+ cost: {
4459
+ input: 2.1,
4460
+ output: 6.6,
4461
+ cacheRead: 0.39,
4462
+ cacheWrite: 0,
4463
+ },
4464
+ contextWindow: 1048572,
4465
+ maxTokens: 262144,
4466
+ },
4344
4467
  "accounts/fireworks/routers/kimi-k2p6-turbo": {
4345
4468
  id: "accounts/fireworks/routers/kimi-k2p6-turbo",
4346
4469
  name: "Kimi K2.6 Turbo",
@@ -4398,6 +4521,25 @@ export const MODELS = {
4398
4521
  contextWindow: 1000000,
4399
4522
  maxTokens: 128000,
4400
4523
  },
4524
+ "claude-fable-5.1": {
4525
+ id: "claude-fable-5.1",
4526
+ name: "Claude Fable 5.1",
4527
+ api: "openai-completions",
4528
+ provider: "github-copilot",
4529
+ baseUrl: "https://api.individual.githubcopilot.com",
4530
+ 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" },
4531
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4532
+ reasoning: true,
4533
+ input: ["text", "image"],
4534
+ cost: {
4535
+ input: 10,
4536
+ output: 50,
4537
+ cacheRead: 0.25,
4538
+ cacheWrite: 12.5,
4539
+ },
4540
+ contextWindow: 1000000,
4541
+ maxTokens: 128000,
4542
+ },
4401
4543
  "claude-haiku-4.5": {
4402
4544
  id: "claude-haiku-4.5",
4403
4545
  name: "Claude Haiku 4.5 (latest)",
@@ -4417,44 +4559,6 @@ export const MODELS = {
4417
4559
  contextWindow: 200000,
4418
4560
  maxTokens: 64000,
4419
4561
  },
4420
- "claude-opus-4.5": {
4421
- id: "claude-opus-4.5",
4422
- name: "Claude Opus 4.5 (latest)",
4423
- api: "anthropic-messages",
4424
- provider: "github-copilot",
4425
- baseUrl: "https://api.individual.githubcopilot.com",
4426
- 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" },
4427
- reasoning: true,
4428
- input: ["text", "image"],
4429
- cost: {
4430
- input: 5,
4431
- output: 25,
4432
- cacheRead: 0.5,
4433
- cacheWrite: 6.25,
4434
- },
4435
- contextWindow: 200000,
4436
- maxTokens: 32000,
4437
- },
4438
- "claude-opus-4.6": {
4439
- id: "claude-opus-4.6",
4440
- name: "Claude Opus 4.6",
4441
- api: "anthropic-messages",
4442
- provider: "github-copilot",
4443
- baseUrl: "https://api.individual.githubcopilot.com",
4444
- 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" },
4445
- compat: { "forceAdaptiveThinking": true },
4446
- reasoning: true,
4447
- thinkingLevelMap: { "xhigh": "max" },
4448
- input: ["text", "image"],
4449
- cost: {
4450
- input: 5,
4451
- output: 25,
4452
- cacheRead: 0.5,
4453
- cacheWrite: 6.25,
4454
- },
4455
- contextWindow: 1000000,
4456
- maxTokens: 32000,
4457
- },
4458
4562
  "claude-opus-4.7": {
4459
4563
  id: "claude-opus-4.7",
4460
4564
  name: "Claude Opus 4.7",
@@ -4513,44 +4617,6 @@ export const MODELS = {
4513
4617
  contextWindow: 1000000,
4514
4618
  maxTokens: 64000,
4515
4619
  },
4516
- "claude-sonnet-4": {
4517
- id: "claude-sonnet-4",
4518
- name: "Claude Sonnet 4 (latest)",
4519
- api: "anthropic-messages",
4520
- provider: "github-copilot",
4521
- baseUrl: "https://api.individual.githubcopilot.com",
4522
- 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" },
4523
- compat: { "supportsEagerToolInputStreaming": false },
4524
- reasoning: true,
4525
- input: ["text", "image"],
4526
- cost: {
4527
- input: 3,
4528
- output: 15,
4529
- cacheRead: 0.3,
4530
- cacheWrite: 3.75,
4531
- },
4532
- contextWindow: 216000,
4533
- maxTokens: 16000,
4534
- },
4535
- "claude-sonnet-4.5": {
4536
- id: "claude-sonnet-4.5",
4537
- name: "Claude Sonnet 4.5 (latest)",
4538
- api: "anthropic-messages",
4539
- provider: "github-copilot",
4540
- baseUrl: "https://api.individual.githubcopilot.com",
4541
- 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" },
4542
- compat: { "supportsEagerToolInputStreaming": false },
4543
- reasoning: true,
4544
- input: ["text", "image"],
4545
- cost: {
4546
- input: 3,
4547
- output: 15,
4548
- cacheRead: 0.3,
4549
- cacheWrite: 3.75,
4550
- },
4551
- contextWindow: 200000,
4552
- maxTokens: 32000,
4553
- },
4554
4620
  "claude-sonnet-4.6": {
4555
4621
  id: "claude-sonnet-4.6",
4556
4622
  name: "Claude Sonnet 4.6",
@@ -4589,25 +4655,6 @@ export const MODELS = {
4589
4655
  contextWindow: 1000000,
4590
4656
  maxTokens: 128000,
4591
4657
  },
4592
- "gemini-3.1-pro-preview": {
4593
- id: "gemini-3.1-pro-preview",
4594
- name: "Gemini 3.1 Pro Preview",
4595
- api: "openai-completions",
4596
- provider: "github-copilot",
4597
- baseUrl: "https://api.individual.githubcopilot.com",
4598
- 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" },
4599
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4600
- reasoning: true,
4601
- input: ["text", "image"],
4602
- cost: {
4603
- input: 2,
4604
- output: 12,
4605
- cacheRead: 0.2,
4606
- cacheWrite: 0,
4607
- },
4608
- contextWindow: 1000000,
4609
- maxTokens: 64000,
4610
- },
4611
4658
  "gemini-3.5-flash": {
4612
4659
  id: "gemini-3.5-flash",
4613
4660
  name: "Gemini 3.5 Flash",
@@ -4665,25 +4712,25 @@ export const MODELS = {
4665
4712
  contextWindow: 1000000,
4666
4713
  maxTokens: 64000,
4667
4714
  },
4668
- "gpt-4.1": {
4669
- id: "gpt-4.1",
4670
- name: "GPT-4.1",
4715
+ "gemini-3.8-flash": {
4716
+ id: "gemini-3.8-flash",
4717
+ name: "Gemini 3.8 Flash",
4671
4718
  api: "openai-completions",
4672
4719
  provider: "github-copilot",
4673
4720
  baseUrl: "https://api.individual.githubcopilot.com",
4674
4721
  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" },
4675
4722
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4676
- reasoning: false,
4723
+ reasoning: true,
4677
4724
  input: ["text", "image"],
4678
4725
  cost: {
4679
- input: 2,
4680
- output: 8,
4681
- cacheRead: 0.5,
4726
+ input: 0.75,
4727
+ output: 3.75,
4728
+ cacheRead: 0.075,
4682
4729
  cacheWrite: 0,
4683
4730
  },
4684
- contextWindow: 128000,
4685
- maxTokens: 16384,
4686
- },
4731
+ contextWindow: 1000000,
4732
+ maxTokens: 64000,
4733
+ },
4687
4734
  "gpt-5-mini": {
4688
4735
  id: "gpt-5-mini",
4689
4736
  name: "GPT-5 Mini",
@@ -4703,44 +4750,6 @@ export const MODELS = {
4703
4750
  contextWindow: 264000,
4704
4751
  maxTokens: 64000,
4705
4752
  },
4706
- "gpt-5.2": {
4707
- id: "gpt-5.2",
4708
- name: "GPT-5.2",
4709
- api: "openai-responses",
4710
- provider: "github-copilot",
4711
- baseUrl: "https://api.individual.githubcopilot.com",
4712
- 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" },
4713
- reasoning: true,
4714
- thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4715
- input: ["text", "image"],
4716
- cost: {
4717
- input: 1.75,
4718
- output: 14,
4719
- cacheRead: 0.175,
4720
- cacheWrite: 0,
4721
- },
4722
- contextWindow: 400000,
4723
- maxTokens: 128000,
4724
- },
4725
- "gpt-5.2-codex": {
4726
- id: "gpt-5.2-codex",
4727
- name: "GPT-5.2 Codex",
4728
- api: "openai-responses",
4729
- provider: "github-copilot",
4730
- baseUrl: "https://api.individual.githubcopilot.com",
4731
- 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" },
4732
- reasoning: true,
4733
- thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4734
- input: ["text", "image"],
4735
- cost: {
4736
- input: 1.75,
4737
- output: 14,
4738
- cacheRead: 0.175,
4739
- cacheWrite: 0,
4740
- },
4741
- contextWindow: 400000,
4742
- maxTokens: 128000,
4743
- },
4744
4753
  "gpt-5.3-codex": {
4745
4754
  id: "gpt-5.3-codex",
4746
4755
  name: "GPT-5.3 Codex",
@@ -4866,10 +4875,10 @@ export const MODELS = {
4866
4875
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4867
4876
  input: ["text", "image"],
4868
4877
  cost: {
4869
- input: 2,
4870
- output: 10,
4871
- cacheRead: 0.2,
4872
- cacheWrite: 2.5,
4878
+ input: 4,
4879
+ output: 20,
4880
+ cacheRead: 0.4,
4881
+ cacheWrite: 5,
4873
4882
  },
4874
4883
  contextWindow: 1050000,
4875
4884
  maxTokens: 128000,
@@ -4893,6 +4902,25 @@ export const MODELS = {
4893
4902
  contextWindow: 1050000,
4894
4903
  maxTokens: 128000,
4895
4904
  },
4905
+ "gpt-6-astra": {
4906
+ id: "gpt-6-astra",
4907
+ name: "GPT-6 Astra",
4908
+ api: "openai-responses",
4909
+ provider: "github-copilot",
4910
+ baseUrl: "https://api.individual.githubcopilot.com",
4911
+ 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" },
4912
+ reasoning: true,
4913
+ thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4914
+ input: ["text", "image"],
4915
+ cost: {
4916
+ input: 10,
4917
+ output: 50,
4918
+ cacheRead: 1,
4919
+ cacheWrite: 12.5,
4920
+ },
4921
+ contextWindow: 1050000,
4922
+ maxTokens: 128000,
4923
+ },
4896
4924
  "grok-4.5": {
4897
4925
  id: "grok-4.5",
4898
4926
  name: "Grok 4.5",
@@ -5309,6 +5337,24 @@ export const MODELS = {
5309
5337
  contextWindow: 1048576,
5310
5338
  maxTokens: 65536,
5311
5339
  },
5340
+ "gemini-3.8-flash": {
5341
+ id: "gemini-3.8-flash",
5342
+ name: "Gemini 3.8 Flash",
5343
+ api: "google-generative-ai",
5344
+ provider: "google",
5345
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5346
+ reasoning: true,
5347
+ thinkingLevelMap: { "off": null },
5348
+ input: ["text", "image"],
5349
+ cost: {
5350
+ input: 0.75,
5351
+ output: 3.75,
5352
+ cacheRead: 0.075,
5353
+ cacheWrite: 0,
5354
+ },
5355
+ contextWindow: 1048576,
5356
+ maxTokens: 65536,
5357
+ },
5312
5358
  "gemini-flash-latest": {
5313
5359
  id: "gemini-flash-latest",
5314
5360
  name: "Gemini Flash Latest",
@@ -6414,6 +6460,24 @@ export const MODELS = {
6414
6460
  contextWindow: 1048576,
6415
6461
  maxTokens: 384000,
6416
6462
  },
6463
+ "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp": {
6464
+ id: "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
6465
+ name: "DeepSeek V4 Flash Vision Exp",
6466
+ api: "openai-completions",
6467
+ provider: "huggingface",
6468
+ baseUrl: "https://router.huggingface.co/v1",
6469
+ compat: { "supportsDeveloperRole": false },
6470
+ reasoning: true,
6471
+ input: ["text", "image"],
6472
+ cost: {
6473
+ input: 0.44,
6474
+ output: 1.32,
6475
+ cacheRead: 0,
6476
+ cacheWrite: 0,
6477
+ },
6478
+ contextWindow: 1048576,
6479
+ maxTokens: 384000,
6480
+ },
6417
6481
  "deepseek-ai/DeepSeek-V4-Pro": {
6418
6482
  id: "deepseek-ai/DeepSeek-V4-Pro",
6419
6483
  name: "DeepSeek V4 Pro",
@@ -6736,7 +6800,7 @@ export const MODELS = {
6736
6800
  cacheWrite: 0,
6737
6801
  },
6738
6802
  contextWindow: 262144,
6739
- maxTokens: 64000,
6803
+ maxTokens: 128000,
6740
6804
  },
6741
6805
  "thinkingmachines/Inkling": {
6742
6806
  id: "thinkingmachines/Inkling",
@@ -7684,114 +7748,6 @@ export const MODELS = {
7684
7748
  },
7685
7749
  },
7686
7750
  "moonshotai": {
7687
- "kimi-k2-0711-preview": {
7688
- id: "kimi-k2-0711-preview",
7689
- name: "Kimi K2 0711",
7690
- api: "openai-completions",
7691
- provider: "moonshotai",
7692
- baseUrl: "https://api.moonshot.ai/v1",
7693
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7694
- reasoning: false,
7695
- input: ["text"],
7696
- cost: {
7697
- input: 0.6,
7698
- output: 2.5,
7699
- cacheRead: 0.15,
7700
- cacheWrite: 0,
7701
- },
7702
- contextWindow: 131072,
7703
- maxTokens: 16384,
7704
- },
7705
- "kimi-k2-0905-preview": {
7706
- id: "kimi-k2-0905-preview",
7707
- name: "Kimi K2 0905",
7708
- api: "openai-completions",
7709
- provider: "moonshotai",
7710
- baseUrl: "https://api.moonshot.ai/v1",
7711
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7712
- reasoning: false,
7713
- input: ["text"],
7714
- cost: {
7715
- input: 0.6,
7716
- output: 2.5,
7717
- cacheRead: 0.15,
7718
- cacheWrite: 0,
7719
- },
7720
- contextWindow: 262144,
7721
- maxTokens: 262144,
7722
- },
7723
- "kimi-k2-thinking": {
7724
- id: "kimi-k2-thinking",
7725
- name: "Kimi K2 Thinking",
7726
- api: "openai-completions",
7727
- provider: "moonshotai",
7728
- baseUrl: "https://api.moonshot.ai/v1",
7729
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7730
- reasoning: true,
7731
- input: ["text"],
7732
- cost: {
7733
- input: 0.6,
7734
- output: 2.5,
7735
- cacheRead: 0.15,
7736
- cacheWrite: 0,
7737
- },
7738
- contextWindow: 262144,
7739
- maxTokens: 262144,
7740
- },
7741
- "kimi-k2-thinking-turbo": {
7742
- id: "kimi-k2-thinking-turbo",
7743
- name: "Kimi K2 Thinking Turbo",
7744
- api: "openai-completions",
7745
- provider: "moonshotai",
7746
- baseUrl: "https://api.moonshot.ai/v1",
7747
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7748
- reasoning: true,
7749
- input: ["text"],
7750
- cost: {
7751
- input: 1.15,
7752
- output: 8,
7753
- cacheRead: 0.15,
7754
- cacheWrite: 0,
7755
- },
7756
- contextWindow: 262144,
7757
- maxTokens: 262144,
7758
- },
7759
- "kimi-k2-turbo-preview": {
7760
- id: "kimi-k2-turbo-preview",
7761
- name: "Kimi K2 Turbo",
7762
- api: "openai-completions",
7763
- provider: "moonshotai",
7764
- baseUrl: "https://api.moonshot.ai/v1",
7765
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7766
- reasoning: false,
7767
- input: ["text"],
7768
- cost: {
7769
- input: 2.4,
7770
- output: 10,
7771
- cacheRead: 0.6,
7772
- cacheWrite: 0,
7773
- },
7774
- contextWindow: 262144,
7775
- maxTokens: 262144,
7776
- },
7777
- "kimi-k2.5": {
7778
- id: "kimi-k2.5",
7779
- name: "Kimi K2.5",
7780
- api: "openai-completions",
7781
- provider: "moonshotai",
7782
- baseUrl: "https://api.moonshot.ai/v1",
7783
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7784
- reasoning: true,
7785
- input: ["text", "image"],
7786
- cost: {
7787
- input: 0.6,
7788
- output: 3,
7789
- cacheRead: 0.1,
7790
- cacheWrite: 0,
7791
- },
7792
- contextWindow: 262144,
7793
- maxTokens: 262144,
7794
- },
7795
7751
  "kimi-k2.6": {
7796
7752
  id: "kimi-k2.6",
7797
7753
  name: "Kimi K2.6",
@@ -7866,114 +7822,6 @@ export const MODELS = {
7866
7822
  },
7867
7823
  },
7868
7824
  "moonshotai-cn": {
7869
- "kimi-k2-0711-preview": {
7870
- id: "kimi-k2-0711-preview",
7871
- name: "Kimi K2 0711",
7872
- api: "openai-completions",
7873
- provider: "moonshotai-cn",
7874
- baseUrl: "https://api.moonshot.cn/v1",
7875
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7876
- reasoning: false,
7877
- input: ["text"],
7878
- cost: {
7879
- input: 0.6,
7880
- output: 2.5,
7881
- cacheRead: 0.15,
7882
- cacheWrite: 0,
7883
- },
7884
- contextWindow: 131072,
7885
- maxTokens: 16384,
7886
- },
7887
- "kimi-k2-0905-preview": {
7888
- id: "kimi-k2-0905-preview",
7889
- name: "Kimi K2 0905",
7890
- api: "openai-completions",
7891
- provider: "moonshotai-cn",
7892
- baseUrl: "https://api.moonshot.cn/v1",
7893
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7894
- reasoning: false,
7895
- input: ["text"],
7896
- cost: {
7897
- input: 0.6,
7898
- output: 2.5,
7899
- cacheRead: 0.15,
7900
- cacheWrite: 0,
7901
- },
7902
- contextWindow: 262144,
7903
- maxTokens: 262144,
7904
- },
7905
- "kimi-k2-thinking": {
7906
- id: "kimi-k2-thinking",
7907
- name: "Kimi K2 Thinking",
7908
- api: "openai-completions",
7909
- provider: "moonshotai-cn",
7910
- baseUrl: "https://api.moonshot.cn/v1",
7911
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7912
- reasoning: true,
7913
- input: ["text"],
7914
- cost: {
7915
- input: 0.6,
7916
- output: 2.5,
7917
- cacheRead: 0.15,
7918
- cacheWrite: 0,
7919
- },
7920
- contextWindow: 262144,
7921
- maxTokens: 262144,
7922
- },
7923
- "kimi-k2-thinking-turbo": {
7924
- id: "kimi-k2-thinking-turbo",
7925
- name: "Kimi K2 Thinking Turbo",
7926
- api: "openai-completions",
7927
- provider: "moonshotai-cn",
7928
- baseUrl: "https://api.moonshot.cn/v1",
7929
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7930
- reasoning: true,
7931
- input: ["text"],
7932
- cost: {
7933
- input: 1.15,
7934
- output: 8,
7935
- cacheRead: 0.15,
7936
- cacheWrite: 0,
7937
- },
7938
- contextWindow: 262144,
7939
- maxTokens: 262144,
7940
- },
7941
- "kimi-k2-turbo-preview": {
7942
- id: "kimi-k2-turbo-preview",
7943
- name: "Kimi K2 Turbo",
7944
- api: "openai-completions",
7945
- provider: "moonshotai-cn",
7946
- baseUrl: "https://api.moonshot.cn/v1",
7947
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7948
- reasoning: false,
7949
- input: ["text"],
7950
- cost: {
7951
- input: 2.4,
7952
- output: 10,
7953
- cacheRead: 0.6,
7954
- cacheWrite: 0,
7955
- },
7956
- contextWindow: 262144,
7957
- maxTokens: 262144,
7958
- },
7959
- "kimi-k2.5": {
7960
- id: "kimi-k2.5",
7961
- name: "Kimi K2.5",
7962
- api: "openai-completions",
7963
- provider: "moonshotai-cn",
7964
- baseUrl: "https://api.moonshot.cn/v1",
7965
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7966
- reasoning: true,
7967
- input: ["text", "image"],
7968
- cost: {
7969
- input: 0.6,
7970
- output: 3,
7971
- cacheRead: 0.1,
7972
- cacheWrite: 0,
7973
- },
7974
- contextWindow: 262144,
7975
- maxTokens: 262144,
7976
- },
7977
7825
  "kimi-k2.6": {
7978
7826
  id: "kimi-k2.6",
7979
7827
  name: "Kimi K2.6",
@@ -8650,6 +8498,24 @@ export const MODELS = {
8650
8498
  contextWindow: 1050000,
8651
8499
  maxTokens: 128000,
8652
8500
  },
8501
+ "gpt-6-astra": {
8502
+ id: "gpt-6-astra",
8503
+ name: "GPT-6 Astra",
8504
+ api: "openai-responses",
8505
+ provider: "openai",
8506
+ baseUrl: "https://api.openai.com/v1",
8507
+ reasoning: true,
8508
+ thinkingLevelMap: { "off": "none", "xhigh": "xhigh", "minimal": null },
8509
+ input: ["text", "image"],
8510
+ cost: {
8511
+ input: 10,
8512
+ output: 50,
8513
+ cacheRead: 1,
8514
+ cacheWrite: 12.5,
8515
+ },
8516
+ contextWindow: 1050000,
8517
+ maxTokens: 128000,
8518
+ },
8653
8519
  "gpt-realtime-2.1": {
8654
8520
  id: "gpt-realtime-2.1",
8655
8521
  name: "GPT-Realtime-2.1",
@@ -8951,6 +8817,24 @@ export const MODELS = {
8951
8817
  contextWindow: 272000,
8952
8818
  maxTokens: 128000,
8953
8819
  },
8820
+ "gpt-6-astra": {
8821
+ id: "gpt-6-astra",
8822
+ name: "GPT-6 Astra",
8823
+ api: "openai-codex-responses",
8824
+ provider: "openai-codex",
8825
+ baseUrl: "https://chatgpt.com/backend-api",
8826
+ reasoning: true,
8827
+ thinkingLevelMap: { "xhigh": "xhigh", "minimal": "low" },
8828
+ input: ["text", "image"],
8829
+ cost: {
8830
+ input: 10,
8831
+ output: 50,
8832
+ cacheRead: 1,
8833
+ cacheWrite: 12.5,
8834
+ },
8835
+ contextWindow: 272000,
8836
+ maxTokens: 128000,
8837
+ },
8954
8838
  },
8955
8839
  "opencode": {
8956
8840
  "big-pickle": {
@@ -9199,6 +9083,25 @@ export const MODELS = {
9199
9083
  contextWindow: 1000000,
9200
9084
  maxTokens: 384000,
9201
9085
  },
9086
+ "deepseek-v4-flash-vision-exp": {
9087
+ id: "deepseek-v4-flash-vision-exp",
9088
+ name: "DeepSeek V4 Flash Vision Exp",
9089
+ api: "openai-completions",
9090
+ provider: "opencode",
9091
+ baseUrl: "https://opencode.ai/zen/v1",
9092
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
9093
+ reasoning: true,
9094
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9095
+ input: ["text", "image"],
9096
+ cost: {
9097
+ input: 0.14,
9098
+ output: 0.28,
9099
+ cacheRead: 0.028,
9100
+ cacheWrite: 0,
9101
+ },
9102
+ contextWindow: 1000000,
9103
+ maxTokens: 384000,
9104
+ },
9202
9105
  "deepseek-v4-pro": {
9203
9106
  id: "deepseek-v4-pro",
9204
9107
  name: "DeepSeek V4 Pro",
@@ -9326,6 +9229,24 @@ export const MODELS = {
9326
9229
  contextWindow: 1048576,
9327
9230
  maxTokens: 65536,
9328
9231
  },
9232
+ "gemini-3.8-flash": {
9233
+ id: "gemini-3.8-flash",
9234
+ name: "Gemini 3.8 Flash",
9235
+ api: "google-generative-ai",
9236
+ provider: "opencode",
9237
+ baseUrl: "https://opencode.ai/zen/v1",
9238
+ reasoning: true,
9239
+ thinkingLevelMap: { "off": null },
9240
+ input: ["text", "image"],
9241
+ cost: {
9242
+ input: 1.5,
9243
+ output: 7.5,
9244
+ cacheRead: 0.15,
9245
+ cacheWrite: 0,
9246
+ },
9247
+ contextWindow: 1048576,
9248
+ maxTokens: 65536,
9249
+ },
9329
9250
  "glm-5": {
9330
9251
  id: "glm-5",
9331
9252
  name: "GLM-5",
@@ -9377,6 +9298,40 @@ export const MODELS = {
9377
9298
  contextWindow: 1000000,
9378
9299
  maxTokens: 131072,
9379
9300
  },
9301
+ "glm-5.3": {
9302
+ id: "glm-5.3",
9303
+ name: "GLM-5.3",
9304
+ api: "openai-completions",
9305
+ provider: "opencode",
9306
+ baseUrl: "https://opencode.ai/zen/v1",
9307
+ reasoning: true,
9308
+ input: ["text"],
9309
+ cost: {
9310
+ input: 1.4,
9311
+ output: 4.4,
9312
+ cacheRead: 0.26,
9313
+ cacheWrite: 0,
9314
+ },
9315
+ contextWindow: 1000000,
9316
+ maxTokens: 131072,
9317
+ },
9318
+ "glm-5.3-flash": {
9319
+ id: "glm-5.3-flash",
9320
+ name: "GLM-5.3-Flash",
9321
+ api: "openai-completions",
9322
+ provider: "opencode",
9323
+ baseUrl: "https://opencode.ai/zen/v1",
9324
+ reasoning: true,
9325
+ input: ["text", "image"],
9326
+ cost: {
9327
+ input: 0.15,
9328
+ output: 0.5,
9329
+ cacheRead: 0.03,
9330
+ cacheWrite: 0,
9331
+ },
9332
+ contextWindow: 1000000,
9333
+ maxTokens: 131072,
9334
+ },
9380
9335
  "gpt-5": {
9381
9336
  id: "gpt-5",
9382
9337
  name: "GPT-5",
@@ -9719,6 +9674,24 @@ export const MODELS = {
9719
9674
  contextWindow: 1050000,
9720
9675
  maxTokens: 128000,
9721
9676
  },
9677
+ "gpt-6-astra": {
9678
+ id: "gpt-6-astra",
9679
+ name: "GPT-6 Astra",
9680
+ api: "openai-responses",
9681
+ provider: "opencode",
9682
+ baseUrl: "https://opencode.ai/zen/v1",
9683
+ reasoning: true,
9684
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
9685
+ input: ["text", "image"],
9686
+ cost: {
9687
+ input: 10,
9688
+ output: 50,
9689
+ cacheRead: 1,
9690
+ cacheWrite: 12.5,
9691
+ },
9692
+ contextWindow: 1050000,
9693
+ maxTokens: 128000,
9694
+ },
9722
9695
  "grok-4.5": {
9723
9696
  id: "grok-4.5",
9724
9697
  name: "Grok 4.5",
@@ -9960,6 +9933,40 @@ export const MODELS = {
9960
9933
  contextWindow: 1048576,
9961
9934
  maxTokens: 131072,
9962
9935
  },
9936
+ "muse-spark-1.3": {
9937
+ id: "muse-spark-1.3",
9938
+ name: "Muse Spark 1.3",
9939
+ api: "openai-responses",
9940
+ provider: "opencode",
9941
+ baseUrl: "https://opencode.ai/zen/v1",
9942
+ reasoning: true,
9943
+ input: ["text", "image"],
9944
+ cost: {
9945
+ input: 1.25,
9946
+ output: 4.25,
9947
+ cacheRead: 0.15,
9948
+ cacheWrite: 0,
9949
+ },
9950
+ contextWindow: 1048576,
9951
+ maxTokens: 131072,
9952
+ },
9953
+ "muse-spark-1.3-contributor-free": {
9954
+ id: "muse-spark-1.3-contributor-free",
9955
+ name: "Muse Spark 1.3 Free",
9956
+ api: "openai-responses",
9957
+ provider: "opencode",
9958
+ baseUrl: "https://opencode.ai/zen/v1",
9959
+ reasoning: true,
9960
+ input: ["text", "image"],
9961
+ cost: {
9962
+ input: 0,
9963
+ output: 0,
9964
+ cacheRead: 0,
9965
+ cacheWrite: 0,
9966
+ },
9967
+ contextWindow: 1048576,
9968
+ maxTokens: 131072,
9969
+ },
9963
9970
  "nemotron-3-ultra-free": {
9964
9971
  id: "nemotron-3-ultra-free",
9965
9972
  name: "Nemotron 3 Ultra Free",
@@ -10205,7 +10212,7 @@ export const MODELS = {
10205
10212
  cacheWrite: 0,
10206
10213
  },
10207
10214
  contextWindow: 256000,
10208
- maxTokens: 64000,
10215
+ maxTokens: 128000,
10209
10216
  },
10210
10217
  "hy4-preview": {
10211
10218
  id: "hy4-preview",
@@ -10359,12 +10366,29 @@ export const MODELS = {
10359
10366
  cacheRead: 0.06,
10360
10367
  cacheWrite: 0,
10361
10368
  },
10362
- contextWindow: 1000000,
10369
+ contextWindow: 1000000,
10370
+ maxTokens: 131072,
10371
+ },
10372
+ "muse-spark-1.2-contributor": {
10373
+ id: "muse-spark-1.2-contributor",
10374
+ name: "Muse Spark 1.2 Contributor",
10375
+ api: "openai-responses",
10376
+ provider: "opencode-go",
10377
+ baseUrl: "https://opencode.ai/zen/go/v1",
10378
+ reasoning: true,
10379
+ input: ["text", "image"],
10380
+ cost: {
10381
+ input: 0.1,
10382
+ output: 0.2,
10383
+ cacheRead: 0.002,
10384
+ cacheWrite: 0,
10385
+ },
10386
+ contextWindow: 1048576,
10363
10387
  maxTokens: 131072,
10364
10388
  },
10365
- "muse-spark-1.2-contributor": {
10366
- id: "muse-spark-1.2-contributor",
10367
- name: "Muse Spark 1.2 Contributor",
10389
+ "muse-spark-1.3-contributor": {
10390
+ id: "muse-spark-1.3-contributor",
10391
+ name: "Muse Spark 1.3 Contributor",
10368
10392
  api: "openai-responses",
10369
10393
  provider: "opencode-go",
10370
10394
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -10379,6 +10403,23 @@ export const MODELS = {
10379
10403
  contextWindow: 1048576,
10380
10404
  maxTokens: 131072,
10381
10405
  },
10406
+ "omen-alpha": {
10407
+ id: "omen-alpha",
10408
+ name: "Omen Alpha",
10409
+ api: "openai-completions",
10410
+ provider: "opencode-go",
10411
+ baseUrl: "https://opencode.ai/zen/go/v1",
10412
+ reasoning: true,
10413
+ input: ["text", "image"],
10414
+ cost: {
10415
+ input: 0.2,
10416
+ output: 0.66,
10417
+ cacheRead: 0.04,
10418
+ cacheWrite: 0,
10419
+ },
10420
+ contextWindow: 500000,
10421
+ maxTokens: 128000,
10422
+ },
10382
10423
  "qwen3.6-plus": {
10383
10424
  id: "qwen3.6-plus",
10384
10425
  name: "Qwen3.6 Plus",
@@ -11262,13 +11303,13 @@ export const MODELS = {
11262
11303
  reasoning: false,
11263
11304
  input: ["text"],
11264
11305
  cost: {
11265
- input: 0.2574,
11266
- output: 1.0287,
11306
+ input: 0.32,
11307
+ output: 0.8899999999999999,
11267
11308
  cacheRead: 0,
11268
11309
  cacheWrite: 0,
11269
11310
  },
11270
11311
  contextWindow: 163840,
11271
- maxTokens: 16000,
11312
+ maxTokens: 16384,
11272
11313
  },
11273
11314
  "deepseek/deepseek-chat-v3-0324": {
11274
11315
  id: "deepseek/deepseek-chat-v3-0324",
@@ -11279,9 +11320,9 @@ export const MODELS = {
11279
11320
  reasoning: false,
11280
11321
  input: ["text"],
11281
11322
  cost: {
11282
- input: 0.25,
11283
- output: 1,
11284
- cacheRead: 0,
11323
+ input: 0.29,
11324
+ output: 1.1400000000000001,
11325
+ cacheRead: 0.11,
11285
11326
  cacheWrite: 0,
11286
11327
  },
11287
11328
  contextWindow: 163840,
@@ -11476,9 +11517,9 @@ export const MODELS = {
11476
11517
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11477
11518
  input: ["text"],
11478
11519
  cost: {
11479
- input: 1.04226,
11480
- output: 2.08452,
11481
- cacheRead: 0.086855,
11520
+ input: 0.9552599999999999,
11521
+ output: 1.9105199999999998,
11522
+ cacheRead: 0.07960500000000001,
11482
11523
  cacheWrite: 0,
11483
11524
  },
11484
11525
  contextWindow: 1048576,
@@ -11495,9 +11536,9 @@ export const MODELS = {
11495
11536
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11496
11537
  input: ["text"],
11497
11538
  cost: {
11498
- input: 1.1154,
11499
- output: 3.3461999999999996,
11500
- cacheRead: 0.03718,
11539
+ input: 1.0494,
11540
+ output: 3.1482,
11541
+ cacheRead: 0.03498,
11501
11542
  cacheWrite: 0,
11502
11543
  },
11503
11544
  contextWindow: 1048576,
@@ -11956,10 +11997,44 @@ export const MODELS = {
11956
11997
  reasoning: true,
11957
11998
  input: ["text", "image"],
11958
11999
  cost: {
11959
- input: 0.1875,
11960
- output: 0.9375,
11961
- cacheRead: 0.01875,
11962
- cacheWrite: 0.0208333333333333,
12000
+ input: 0.375,
12001
+ output: 1.875,
12002
+ cacheRead: 0.0375,
12003
+ cacheWrite: 0.0416666666666667,
12004
+ },
12005
+ contextWindow: 1048576,
12006
+ maxTokens: 65536,
12007
+ },
12008
+ "google/gemini-3.8-flash": {
12009
+ id: "google/gemini-3.8-flash",
12010
+ name: "Google: Gemini 3.8 Flash",
12011
+ api: "openai-completions",
12012
+ provider: "openrouter",
12013
+ baseUrl: "https://openrouter.ai/api/v1",
12014
+ reasoning: true,
12015
+ input: ["text", "image"],
12016
+ cost: {
12017
+ input: 0.75,
12018
+ output: 3.75,
12019
+ cacheRead: 0.075,
12020
+ cacheWrite: 0.0416666666666667,
12021
+ },
12022
+ contextWindow: 1048576,
12023
+ maxTokens: 65536,
12024
+ },
12025
+ "google/gemini-3.8-flash:batch": {
12026
+ id: "google/gemini-3.8-flash:batch",
12027
+ name: "Google: Gemini 3.8 Flash (batch)",
12028
+ api: "openai-completions",
12029
+ provider: "openrouter",
12030
+ baseUrl: "https://openrouter.ai/api/v1",
12031
+ reasoning: true,
12032
+ input: ["text", "image"],
12033
+ cost: {
12034
+ input: 0.375,
12035
+ output: 1.875,
12036
+ cacheRead: 0.0375,
12037
+ cacheWrite: 0.0416666666666667,
11963
12038
  },
11964
12039
  contextWindow: 1048576,
11965
12040
  maxTokens: 65536,
@@ -12083,23 +12158,6 @@ export const MODELS = {
12083
12158
  contextWindow: 262144,
12084
12159
  maxTokens: 32768,
12085
12160
  },
12086
- "ibm-granite/granite-4.1-8b": {
12087
- id: "ibm-granite/granite-4.1-8b",
12088
- name: "IBM: Granite 4.1 8B",
12089
- api: "openai-completions",
12090
- provider: "openrouter",
12091
- baseUrl: "https://openrouter.ai/api/v1",
12092
- reasoning: false,
12093
- input: ["text"],
12094
- cost: {
12095
- input: 0.049999999999999996,
12096
- output: 0.09999999999999999,
12097
- cacheRead: 0.049999999999999996,
12098
- cacheWrite: 0,
12099
- },
12100
- contextWindow: 131072,
12101
- maxTokens: 117964,
12102
- },
12103
12161
  "ibm-granite/granite-4.2-8b": {
12104
12162
  id: "ibm-granite/granite-4.2-8b",
12105
12163
  name: "IBM: Granite 4.2 8B",
@@ -12155,7 +12213,7 @@ export const MODELS = {
12155
12213
  },
12156
12214
  "inclusionai/ling-3.0-flash": {
12157
12215
  id: "inclusionai/ling-3.0-flash",
12158
- name: "Ling-3.0-flash",
12216
+ name: "inclusionAI: Ling 3.0 Flash",
12159
12217
  api: "openai-completions",
12160
12218
  provider: "openrouter",
12161
12219
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12170,9 +12228,43 @@ export const MODELS = {
12170
12228
  contextWindow: 262144,
12171
12229
  maxTokens: 32768,
12172
12230
  },
12231
+ "inclusionai/ling-3.0-flash-fin": {
12232
+ id: "inclusionai/ling-3.0-flash-fin",
12233
+ name: "inclusionAI: Ling 3.0 Flash Fin",
12234
+ api: "openai-completions",
12235
+ provider: "openrouter",
12236
+ baseUrl: "https://openrouter.ai/api/v1",
12237
+ reasoning: true,
12238
+ input: ["text"],
12239
+ cost: {
12240
+ input: 0.06,
12241
+ output: 0.18,
12242
+ cacheRead: 0.012,
12243
+ cacheWrite: 0,
12244
+ },
12245
+ contextWindow: 262144,
12246
+ maxTokens: 235929,
12247
+ },
12173
12248
  "inclusionai/ling-3.0-flash-fin:free": {
12174
12249
  id: "inclusionai/ling-3.0-flash-fin:free",
12175
- name: "Ling 3.0 Flash Fin (free)",
12250
+ name: "inclusionAI: Ling 3.0 Flash Fin (free)",
12251
+ api: "openai-completions",
12252
+ provider: "openrouter",
12253
+ baseUrl: "https://openrouter.ai/api/v1",
12254
+ reasoning: true,
12255
+ input: ["text"],
12256
+ cost: {
12257
+ input: 0,
12258
+ output: 0,
12259
+ cacheRead: 0,
12260
+ cacheWrite: 0,
12261
+ },
12262
+ contextWindow: 262144,
12263
+ maxTokens: 32768,
12264
+ },
12265
+ "inclusionai/ling-3.0-flash-sante:free": {
12266
+ id: "inclusionai/ling-3.0-flash-sante:free",
12267
+ name: "inclusionAI: Ling 3.0 Flash Sante (free)",
12176
12268
  api: "openai-completions",
12177
12269
  provider: "openrouter",
12178
12270
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12298,13 +12390,13 @@ export const MODELS = {
12298
12390
  reasoning: false,
12299
12391
  input: ["text"],
12300
12392
  cost: {
12301
- input: 0.71,
12302
- output: 0.71,
12303
- cacheRead: 0.71,
12393
+ input: 0.09999999999999999,
12394
+ output: 0.32,
12395
+ cacheRead: 0,
12304
12396
  cacheWrite: 0,
12305
12397
  },
12306
12398
  contextWindow: 131072,
12307
- maxTokens: 115200,
12399
+ maxTokens: 16384,
12308
12400
  },
12309
12401
  "meta-llama/llama-4-maverick": {
12310
12402
  id: "meta-llama/llama-4-maverick",
@@ -12350,12 +12442,12 @@ export const MODELS = {
12350
12442
  input: ["text", "image"],
12351
12443
  cost: {
12352
12444
  input: 0.3,
12353
- output: 1.2,
12445
+ output: 1.1,
12354
12446
  cacheRead: 0.04,
12355
12447
  cacheWrite: 0,
12356
12448
  },
12357
12449
  contextWindow: 131072,
12358
- maxTokens: 16384,
12450
+ maxTokens: 117964,
12359
12451
  },
12360
12452
  "meta/muse-glimmer-30b:batch": {
12361
12453
  id: "meta/muse-glimmer-30b:batch",
@@ -12425,6 +12517,40 @@ export const MODELS = {
12425
12517
  contextWindow: 1048576,
12426
12518
  maxTokens: 943718,
12427
12519
  },
12520
+ "meta/muse-spark-1.3": {
12521
+ id: "meta/muse-spark-1.3",
12522
+ name: "Meta: Muse Spark 1.3",
12523
+ api: "openai-completions",
12524
+ provider: "openrouter",
12525
+ baseUrl: "https://openrouter.ai/api/v1",
12526
+ reasoning: true,
12527
+ input: ["text", "image"],
12528
+ cost: {
12529
+ input: 1.25,
12530
+ output: 4.25,
12531
+ cacheRead: 0.15,
12532
+ cacheWrite: 0,
12533
+ },
12534
+ contextWindow: 1048576,
12535
+ maxTokens: 943718,
12536
+ },
12537
+ "meta/muse-spark-1.3-contributor": {
12538
+ id: "meta/muse-spark-1.3-contributor",
12539
+ name: "Meta: Muse Spark 1.3 Contributor",
12540
+ api: "openai-completions",
12541
+ provider: "openrouter",
12542
+ baseUrl: "https://openrouter.ai/api/v1",
12543
+ reasoning: true,
12544
+ input: ["text", "image"],
12545
+ cost: {
12546
+ input: 0.09999999999999999,
12547
+ output: 0.19999999999999998,
12548
+ cacheRead: 0.002,
12549
+ cacheWrite: 0,
12550
+ },
12551
+ contextWindow: 1048576,
12552
+ maxTokens: 943718,
12553
+ },
12428
12554
  "minimax/minimax-m1": {
12429
12555
  id: "minimax/minimax-m1",
12430
12556
  name: "MiniMax: MiniMax M1",
@@ -12434,7 +12560,7 @@ export const MODELS = {
12434
12560
  reasoning: true,
12435
12561
  input: ["text"],
12436
12562
  cost: {
12437
- input: 0.55,
12563
+ input: 0.39999999999999997,
12438
12564
  output: 2.2,
12439
12565
  cacheRead: 0,
12440
12566
  cacheWrite: 0,
@@ -12510,23 +12636,6 @@ export const MODELS = {
12510
12636
  contextWindow: 204800,
12511
12637
  maxTokens: 131072,
12512
12638
  },
12513
- "minimax/minimax-m2.7:free": {
12514
- id: "minimax/minimax-m2.7:free",
12515
- name: "MiniMax: MiniMax M2.7 (free)",
12516
- api: "openai-completions",
12517
- provider: "openrouter",
12518
- baseUrl: "https://openrouter.ai/api/v1",
12519
- reasoning: true,
12520
- input: ["text"],
12521
- cost: {
12522
- input: 0,
12523
- output: 0,
12524
- cacheRead: 0,
12525
- cacheWrite: 0,
12526
- },
12527
- contextWindow: 196608,
12528
- maxTokens: 176947,
12529
- },
12530
12639
  "minimax/minimax-m3": {
12531
12640
  id: "minimax/minimax-m3",
12532
12641
  name: "MiniMax: MiniMax M3",
@@ -12561,23 +12670,6 @@ export const MODELS = {
12561
12670
  contextWindow: 524288,
12562
12671
  maxTokens: 471859,
12563
12672
  },
12564
- "minimax/minimax-m3:free": {
12565
- id: "minimax/minimax-m3:free",
12566
- name: "MiniMax: MiniMax M3 (free)",
12567
- api: "openai-completions",
12568
- provider: "openrouter",
12569
- baseUrl: "https://openrouter.ai/api/v1",
12570
- reasoning: true,
12571
- input: ["text", "image"],
12572
- cost: {
12573
- input: 0,
12574
- output: 0,
12575
- cacheRead: 0,
12576
- cacheWrite: 0,
12577
- },
12578
- contextWindow: 1048576,
12579
- maxTokens: 943718,
12580
- },
12581
12673
  "mistralai/codestral-2508": {
12582
12674
  id: "mistralai/codestral-2508",
12583
12675
  name: "Mistral: Codestral 2508",
@@ -12929,11 +13021,11 @@ export const MODELS = {
12929
13021
  cost: {
12930
13022
  input: 0.6,
12931
13023
  output: 2.5,
12932
- cacheRead: 0.15,
13024
+ cacheRead: 0,
12933
13025
  cacheWrite: 0,
12934
13026
  },
12935
13027
  contextWindow: 262144,
12936
- maxTokens: 100352,
13028
+ maxTokens: 235929,
12937
13029
  },
12938
13030
  "moonshotai/kimi-k2.5": {
12939
13031
  id: "moonshotai/kimi-k2.5",
@@ -12979,9 +13071,9 @@ export const MODELS = {
12979
13071
  reasoning: true,
12980
13072
  input: ["text", "image"],
12981
13073
  cost: {
12982
- input: 0.66,
12983
- output: 3.4,
12984
- cacheRead: 0.18,
13074
+ input: 0.71,
13075
+ output: 3.5,
13076
+ cacheRead: 0.15,
12985
13077
  cacheWrite: 0,
12986
13078
  },
12987
13079
  contextWindow: 262144,
@@ -13066,11 +13158,11 @@ export const MODELS = {
13066
13158
  cost: {
13067
13159
  input: 0.049999999999999996,
13068
13160
  output: 0.19999999999999998,
13069
- cacheRead: 0.024999999999999998,
13161
+ cacheRead: 0.03,
13070
13162
  cacheWrite: 0,
13071
13163
  },
13072
13164
  contextWindow: 262144,
13073
- maxTokens: 228000,
13165
+ maxTokens: 235929,
13074
13166
  },
13075
13167
  "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free": {
13076
13168
  id: "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free",
@@ -13140,23 +13232,6 @@ export const MODELS = {
13140
13232
  contextWindow: 262144,
13141
13233
  maxTokens: 32768,
13142
13234
  },
13143
- "nvidia/nemotron-3-ultra-550b-a55b:batch": {
13144
- id: "nvidia/nemotron-3-ultra-550b-a55b:batch",
13145
- name: "NVIDIA: Nemotron 3 Ultra (batch)",
13146
- api: "openai-completions",
13147
- provider: "openrouter",
13148
- baseUrl: "https://openrouter.ai/api/v1",
13149
- reasoning: true,
13150
- input: ["text"],
13151
- cost: {
13152
- input: 0.6,
13153
- output: 3.5999999999999996,
13154
- cacheRead: 0.19999999999999998,
13155
- cacheWrite: 0,
13156
- },
13157
- contextWindow: 512288,
13158
- maxTokens: 461059,
13159
- },
13160
13235
  "nvidia/nemotron-3-ultra-550b-a55b:free": {
13161
13236
  id: "nvidia/nemotron-3-ultra-550b-a55b:free",
13162
13237
  name: "NVIDIA: Nemotron 3 Ultra (free)",
@@ -14353,12 +14428,84 @@ export const MODELS = {
14353
14428
  reasoning: true,
14354
14429
  input: ["text", "image"],
14355
14430
  cost: {
14356
- input: 0.625,
14357
- output: 5,
14358
- cacheRead: 0.0625,
14359
- cacheWrite: 0,
14431
+ input: 0.625,
14432
+ output: 5,
14433
+ cacheRead: 0.0625,
14434
+ cacheWrite: 0,
14435
+ },
14436
+ contextWindow: 400000,
14437
+ maxTokens: 128000,
14438
+ },
14439
+ "openai/gpt-6-astra": {
14440
+ id: "openai/gpt-6-astra",
14441
+ name: "OpenAI: GPT-6 Astra",
14442
+ api: "openai-completions",
14443
+ provider: "openrouter",
14444
+ baseUrl: "https://openrouter.ai/api/v1",
14445
+ reasoning: true,
14446
+ thinkingLevelMap: { "xhigh": "xhigh" },
14447
+ input: ["text", "image"],
14448
+ cost: {
14449
+ input: 10,
14450
+ output: 50,
14451
+ cacheRead: 1,
14452
+ cacheWrite: 12.5,
14453
+ },
14454
+ contextWindow: 1050000,
14455
+ maxTokens: 128000,
14456
+ },
14457
+ "openai/gpt-6-astra-pro": {
14458
+ id: "openai/gpt-6-astra-pro",
14459
+ name: "OpenAI: GPT-6 Astra Pro",
14460
+ api: "openai-completions",
14461
+ provider: "openrouter",
14462
+ baseUrl: "https://openrouter.ai/api/v1",
14463
+ reasoning: true,
14464
+ thinkingLevelMap: { "xhigh": "xhigh" },
14465
+ input: ["text", "image"],
14466
+ cost: {
14467
+ input: 10,
14468
+ output: 50,
14469
+ cacheRead: 1,
14470
+ cacheWrite: 12.5,
14471
+ },
14472
+ contextWindow: 1050000,
14473
+ maxTokens: 128000,
14474
+ },
14475
+ "openai/gpt-6-astra-pro:batch": {
14476
+ id: "openai/gpt-6-astra-pro:batch",
14477
+ name: "OpenAI: GPT-6 Astra Pro (batch)",
14478
+ api: "openai-completions",
14479
+ provider: "openrouter",
14480
+ baseUrl: "https://openrouter.ai/api/v1",
14481
+ reasoning: true,
14482
+ thinkingLevelMap: { "xhigh": "xhigh" },
14483
+ input: ["text", "image"],
14484
+ cost: {
14485
+ input: 5,
14486
+ output: 25,
14487
+ cacheRead: 0.5,
14488
+ cacheWrite: 6.25,
14489
+ },
14490
+ contextWindow: 1050000,
14491
+ maxTokens: 128000,
14492
+ },
14493
+ "openai/gpt-6-astra:batch": {
14494
+ id: "openai/gpt-6-astra:batch",
14495
+ name: "OpenAI: GPT-6 Astra (batch)",
14496
+ api: "openai-completions",
14497
+ provider: "openrouter",
14498
+ baseUrl: "https://openrouter.ai/api/v1",
14499
+ reasoning: true,
14500
+ thinkingLevelMap: { "xhigh": "xhigh" },
14501
+ input: ["text", "image"],
14502
+ cost: {
14503
+ input: 5,
14504
+ output: 25,
14505
+ cacheRead: 0.5,
14506
+ cacheWrite: 6.25,
14360
14507
  },
14361
- contextWindow: 400000,
14508
+ contextWindow: 1050000,
14362
14509
  maxTokens: 128000,
14363
14510
  },
14364
14511
  "openai/gpt-audio": {
@@ -14846,13 +14993,13 @@ export const MODELS = {
14846
14993
  reasoning: true,
14847
14994
  input: ["text"],
14848
14995
  cost: {
14849
- input: 0.12,
14850
- output: 0.24,
14996
+ input: 0.22749999999999998,
14997
+ output: 0.9099999999999999,
14851
14998
  cacheRead: 0,
14852
14999
  cacheWrite: 0,
14853
15000
  },
14854
15001
  contextWindow: 131072,
14855
- maxTokens: 16384,
15002
+ maxTokens: 8192,
14856
15003
  },
14857
15004
  "qwen/qwen3-235b-a22b": {
14858
15005
  id: "qwen/qwen3-235b-a22b",
@@ -14880,13 +15027,13 @@ export const MODELS = {
14880
15027
  reasoning: false,
14881
15028
  input: ["text"],
14882
15029
  cost: {
14883
- input: 0.0875,
14884
- output: 0.35,
14885
- cacheRead: 0.0175,
15030
+ input: 0.09,
15031
+ output: 0.55,
15032
+ cacheRead: 0,
14886
15033
  cacheWrite: 0,
14887
15034
  },
14888
15035
  contextWindow: 262144,
14889
- maxTokens: 235929,
15036
+ maxTokens: 16384,
14890
15037
  },
14891
15038
  "qwen/qwen3-235b-a22b-thinking-2507": {
14892
15039
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -15141,7 +15288,7 @@ export const MODELS = {
15141
15288
  cacheWrite: 0,
15142
15289
  },
15143
15290
  contextWindow: 262144,
15144
- maxTokens: 32768,
15291
+ maxTokens: 235929,
15145
15292
  },
15146
15293
  "qwen/qwen3-vl-235b-a22b-instruct": {
15147
15294
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -15305,13 +15452,13 @@ export const MODELS = {
15305
15452
  reasoning: true,
15306
15453
  input: ["text", "image"],
15307
15454
  cost: {
15308
- input: 0.25,
15455
+ input: 0.3125,
15309
15456
  output: 1.25,
15310
- cacheRead: 0.25,
15457
+ cacheRead: 0.15625,
15311
15458
  cacheWrite: 0,
15312
15459
  },
15313
15460
  contextWindow: 262144,
15314
- maxTokens: 235929,
15461
+ maxTokens: 16384,
15315
15462
  },
15316
15463
  "qwen/qwen3.5-397b-a17b": {
15317
15464
  id: "qwen/qwen3.5-397b-a17b",
@@ -15424,13 +15571,13 @@ export const MODELS = {
15424
15571
  reasoning: true,
15425
15572
  input: ["text", "image"],
15426
15573
  cost: {
15427
- input: 0.6,
15428
- output: 3.5999999999999996,
15429
- cacheRead: 0.12,
15574
+ input: 0.3,
15575
+ output: 2,
15576
+ cacheRead: 0.03,
15430
15577
  cacheWrite: 0,
15431
15578
  },
15432
15579
  contextWindow: 262144,
15433
- maxTokens: 235929,
15580
+ maxTokens: 65536,
15434
15581
  },
15435
15582
  "qwen/qwen3.6-35b-a3b": {
15436
15583
  id: "qwen/qwen3.6-35b-a3b",
@@ -15562,11 +15709,11 @@ export const MODELS = {
15562
15709
  cost: {
15563
15710
  input: 2,
15564
15711
  output: 6,
15565
- cacheRead: 0.19999999999999998,
15712
+ cacheRead: 0.25,
15566
15713
  cacheWrite: 0,
15567
15714
  },
15568
15715
  contextWindow: 1048576,
15569
- maxTokens: 131072,
15716
+ maxTokens: 262144,
15570
15717
  },
15571
15718
  "qwen/qwen3.8-2.4t-a95b:batch": {
15572
15719
  id: "qwen/qwen3.8-2.4t-a95b:batch",
@@ -15594,10 +15741,10 @@ export const MODELS = {
15594
15741
  reasoning: true,
15595
15742
  input: ["text", "image"],
15596
15743
  cost: {
15597
- input: 0.425,
15598
- output: 2.5500000000000003,
15744
+ input: 0.42,
15745
+ output: 3,
15599
15746
  cacheRead: 0.08499999999999999,
15600
- cacheWrite: 0.53125,
15747
+ cacheWrite: 0,
15601
15748
  },
15602
15749
  contextWindow: 1000000,
15603
15750
  maxTokens: 131072,
@@ -15619,9 +15766,9 @@ export const MODELS = {
15619
15766
  contextWindow: 1000000,
15620
15767
  maxTokens: 131072,
15621
15768
  },
15622
- "qwen/qwen3.8-max": {
15623
- id: "qwen/qwen3.8-max",
15624
- name: "Qwen: Qwen3.8 Max",
15769
+ "qwen/qwen3.8-max-0902": {
15770
+ id: "qwen/qwen3.8-max-0902",
15771
+ name: "Qwen: Qwen3.8 Max (0902)",
15625
15772
  api: "openai-completions",
15626
15773
  provider: "openrouter",
15627
15774
  baseUrl: "https://openrouter.ai/api/v1",
@@ -15821,7 +15968,7 @@ export const MODELS = {
15821
15968
  cacheWrite: 0,
15822
15969
  },
15823
15970
  contextWindow: 1024000,
15824
- maxTokens: 26214,
15971
+ maxTokens: 819200,
15825
15972
  },
15826
15973
  "thinkingmachines/inkling": {
15827
15974
  id: "thinkingmachines/inkling",
@@ -15993,6 +16140,23 @@ export const MODELS = {
15993
16140
  contextWindow: 1000000,
15994
16141
  maxTokens: 900000,
15995
16142
  },
16143
+ "x-ai/grok-4.3:batch": {
16144
+ id: "x-ai/grok-4.3:batch",
16145
+ name: "SpaceXAI: Grok 4.3 (batch)",
16146
+ api: "openai-completions",
16147
+ provider: "openrouter",
16148
+ baseUrl: "https://openrouter.ai/api/v1",
16149
+ reasoning: true,
16150
+ input: ["text", "image"],
16151
+ cost: {
16152
+ input: 1,
16153
+ output: 2,
16154
+ cacheRead: 0.16,
16155
+ cacheWrite: 0,
16156
+ },
16157
+ contextWindow: 1000000,
16158
+ maxTokens: 900000,
16159
+ },
15996
16160
  "x-ai/grok-4.5": {
15997
16161
  id: "x-ai/grok-4.5",
15998
16162
  name: "SpaceXAI: Grok 4.5",
@@ -16138,13 +16302,13 @@ export const MODELS = {
16138
16302
  reasoning: true,
16139
16303
  input: ["text"],
16140
16304
  cost: {
16141
- input: 0.43,
16142
- output: 1.75,
16143
- cacheRead: 0.08,
16305
+ input: 0.55,
16306
+ output: 2.2,
16307
+ cacheRead: 0.11,
16144
16308
  cacheWrite: 0,
16145
16309
  },
16146
16310
  contextWindow: 204800,
16147
- maxTokens: 16384,
16311
+ maxTokens: 131072,
16148
16312
  },
16149
16313
  "z-ai/glm-4.6v": {
16150
16314
  id: "z-ai/glm-4.6v",
@@ -16189,13 +16353,13 @@ export const MODELS = {
16189
16353
  reasoning: true,
16190
16354
  input: ["text"],
16191
16355
  cost: {
16192
- input: 0.06,
16356
+ input: 0.060500000000000005,
16193
16357
  output: 0.39999999999999997,
16194
- cacheRead: 0.01,
16358
+ cacheRead: 0,
16195
16359
  cacheWrite: 0,
16196
16360
  },
16197
16361
  contextWindow: 202752,
16198
- maxTokens: 16384,
16362
+ maxTokens: 117964,
16199
16363
  },
16200
16364
  "z-ai/glm-5": {
16201
16365
  id: "z-ai/glm-5",
@@ -16265,23 +16429,6 @@ export const MODELS = {
16265
16429
  contextWindow: 1048576,
16266
16430
  maxTokens: 131072,
16267
16431
  },
16268
- "z-ai/glm-5.2:free": {
16269
- id: "z-ai/glm-5.2:free",
16270
- name: "Z.ai: GLM 5.2 (free)",
16271
- api: "openai-completions",
16272
- provider: "openrouter",
16273
- baseUrl: "https://openrouter.ai/api/v1",
16274
- reasoning: true,
16275
- input: ["text"],
16276
- cost: {
16277
- input: 0,
16278
- output: 0,
16279
- cacheRead: 0,
16280
- cacheWrite: 0,
16281
- },
16282
- contextWindow: 256000,
16283
- maxTokens: 230400,
16284
- },
16285
16432
  "z-ai/glm-5.3": {
16286
16433
  id: "z-ai/glm-5.3",
16287
16434
  name: "Z.ai: GLM 5.3",
@@ -16297,7 +16444,7 @@ export const MODELS = {
16297
16444
  cacheWrite: 0,
16298
16445
  },
16299
16446
  contextWindow: 1310720,
16300
- maxTokens: 131072,
16447
+ maxTokens: 943718,
16301
16448
  },
16302
16449
  "z-ai/glm-5.3-flash": {
16303
16450
  id: "z-ai/glm-5.3-flash",
@@ -16480,9 +16627,9 @@ export const MODELS = {
16480
16627
  reasoning: true,
16481
16628
  input: ["text", "image"],
16482
16629
  cost: {
16483
- input: 2.5500000000000003,
16484
- output: 12.75,
16485
- cacheRead: 0.25599998999999996,
16630
+ input: 2.5,
16631
+ output: 14,
16632
+ cacheRead: 0.29,
16486
16633
  cacheWrite: 0,
16487
16634
  },
16488
16635
  contextWindow: 1048576,
@@ -16548,9 +16695,9 @@ export const MODELS = {
16548
16695
  reasoning: true,
16549
16696
  input: ["text", "image"],
16550
16697
  cost: {
16551
- input: 0.075,
16552
- output: 0.25,
16553
- cacheRead: 0.015,
16698
+ input: 0.07125000000000001,
16699
+ output: 0.2375,
16700
+ cacheRead: 0.01425,
16554
16701
  cacheWrite: 0,
16555
16702
  },
16556
16703
  contextWindow: 1310720,
@@ -16565,9 +16712,9 @@ export const MODELS = {
16565
16712
  reasoning: true,
16566
16713
  input: ["text"],
16567
16714
  cost: {
16568
- input: 1.17,
16569
- output: 3.9600000000000004,
16570
- cacheRead: 0.234,
16715
+ input: 1.12,
16716
+ output: 3.52,
16717
+ cacheRead: 0.20800000000000002,
16571
16718
  cacheWrite: 0,
16572
16719
  },
16573
16720
  contextWindow: 1310720,
@@ -18255,6 +18402,23 @@ export const MODELS = {
18255
18402
  contextWindow: 1000000,
18256
18403
  maxTokens: 65536,
18257
18404
  },
18405
+ "google/gemini-3.8-flash": {
18406
+ id: "google/gemini-3.8-flash",
18407
+ name: "Gemini 3.8 Flash",
18408
+ api: "anthropic-messages",
18409
+ provider: "vercel-ai-gateway",
18410
+ baseUrl: "https://ai-gateway.vercel.sh",
18411
+ reasoning: true,
18412
+ input: ["text", "image"],
18413
+ cost: {
18414
+ input: 0.75,
18415
+ output: 3.75,
18416
+ cacheRead: 0.075,
18417
+ cacheWrite: 0,
18418
+ },
18419
+ contextWindow: 1000000,
18420
+ maxTokens: 65536,
18421
+ },
18258
18422
  "google/gemma-4-26b-a4b-it": {
18259
18423
  id: "google/gemma-4-26b-a4b-it",
18260
18424
  name: "Google Gemma 4 26B A4B",
@@ -18374,6 +18538,40 @@ export const MODELS = {
18374
18538
  contextWindow: 256000,
18375
18539
  maxTokens: 32000,
18376
18540
  },
18541
+ "inclusionai/ling-3.0-flash-sante": {
18542
+ id: "inclusionai/ling-3.0-flash-sante",
18543
+ name: "Ling 3.0 Flash Sante",
18544
+ api: "anthropic-messages",
18545
+ provider: "vercel-ai-gateway",
18546
+ baseUrl: "https://ai-gateway.vercel.sh",
18547
+ reasoning: true,
18548
+ input: ["text"],
18549
+ cost: {
18550
+ input: 0,
18551
+ output: 0,
18552
+ cacheRead: 0,
18553
+ cacheWrite: 0,
18554
+ },
18555
+ contextWindow: 256000,
18556
+ maxTokens: 32000,
18557
+ },
18558
+ "inclusionai/ling-3.0-flash-sante-free": {
18559
+ id: "inclusionai/ling-3.0-flash-sante-free",
18560
+ name: "Ling 3.0 Flash Sante (Free)",
18561
+ api: "anthropic-messages",
18562
+ provider: "vercel-ai-gateway",
18563
+ baseUrl: "https://ai-gateway.vercel.sh",
18564
+ reasoning: true,
18565
+ input: ["text"],
18566
+ cost: {
18567
+ input: 0,
18568
+ output: 0,
18569
+ cacheRead: 0,
18570
+ cacheWrite: 0,
18571
+ },
18572
+ contextWindow: 256000,
18573
+ maxTokens: 32000,
18574
+ },
18377
18575
  "interfaze/interfaze-beta": {
18378
18576
  id: "interfaze/interfaze-beta",
18379
18577
  name: "Interfaze Beta",
@@ -18612,6 +18810,40 @@ export const MODELS = {
18612
18810
  contextWindow: 1048576,
18613
18811
  maxTokens: 1048576,
18614
18812
  },
18813
+ "meta/muse-spark-1.3": {
18814
+ id: "meta/muse-spark-1.3",
18815
+ name: "Muse Spark 1.3",
18816
+ api: "anthropic-messages",
18817
+ provider: "vercel-ai-gateway",
18818
+ baseUrl: "https://ai-gateway.vercel.sh",
18819
+ reasoning: true,
18820
+ input: ["text", "image"],
18821
+ cost: {
18822
+ input: 1.25,
18823
+ output: 4.25,
18824
+ cacheRead: 0.15,
18825
+ cacheWrite: 0,
18826
+ },
18827
+ contextWindow: 1048576,
18828
+ maxTokens: 1048576,
18829
+ },
18830
+ "meta/muse-spark-1.3-contributor": {
18831
+ id: "meta/muse-spark-1.3-contributor",
18832
+ name: "Muse Spark 1.3 Contributor",
18833
+ api: "anthropic-messages",
18834
+ provider: "vercel-ai-gateway",
18835
+ baseUrl: "https://ai-gateway.vercel.sh",
18836
+ reasoning: true,
18837
+ input: ["text", "image"],
18838
+ cost: {
18839
+ input: 0.09999999999999999,
18840
+ output: 0.19999999999999998,
18841
+ cacheRead: 0.002,
18842
+ cacheWrite: 0,
18843
+ },
18844
+ contextWindow: 1048576,
18845
+ maxTokens: 1048576,
18846
+ },
18615
18847
  "minimax/minimax-m2": {
18616
18848
  id: "minimax/minimax-m2",
18617
18849
  name: "MiniMax M2",
@@ -18714,23 +18946,6 @@ export const MODELS = {
18714
18946
  contextWindow: 204800,
18715
18947
  maxTokens: 131000,
18716
18948
  },
18717
- "minimax/minimax-m2.7-free": {
18718
- id: "minimax/minimax-m2.7-free",
18719
- name: "MiniMax M2.7 (Free)",
18720
- api: "anthropic-messages",
18721
- provider: "vercel-ai-gateway",
18722
- baseUrl: "https://ai-gateway.vercel.sh",
18723
- reasoning: true,
18724
- input: ["text"],
18725
- cost: {
18726
- input: 0,
18727
- output: 0,
18728
- cacheRead: 0,
18729
- cacheWrite: 0,
18730
- },
18731
- contextWindow: 196608,
18732
- maxTokens: 196608,
18733
- },
18734
18949
  "minimax/minimax-m2.7-highspeed": {
18735
18950
  id: "minimax/minimax-m2.7-highspeed",
18736
18951
  name: "MiniMax M2.7 High Speed",
@@ -18765,23 +18980,6 @@ export const MODELS = {
18765
18980
  contextWindow: 512000,
18766
18981
  maxTokens: 512000,
18767
18982
  },
18768
- "minimax/minimax-m3-free": {
18769
- id: "minimax/minimax-m3-free",
18770
- name: "MiniMax M3 (Free)",
18771
- api: "anthropic-messages",
18772
- provider: "vercel-ai-gateway",
18773
- baseUrl: "https://ai-gateway.vercel.sh",
18774
- reasoning: true,
18775
- input: ["text", "image"],
18776
- cost: {
18777
- input: 0,
18778
- output: 0,
18779
- cacheRead: 0,
18780
- cacheWrite: 0,
18781
- },
18782
- contextWindow: 1048576,
18783
- maxTokens: 1048576,
18784
- },
18785
18983
  "mistral/codestral": {
18786
18984
  id: "mistral/codestral",
18787
18985
  name: "Mistral Codestral",
@@ -19933,7 +20131,7 @@ export const MODELS = {
19933
20131
  input: 0.39999999999999997,
19934
20132
  output: 2.4,
19935
20133
  cacheRead: 0.04,
19936
- cacheWrite: 0.25,
20134
+ cacheWrite: 0.5,
19937
20135
  },
19938
20136
  contextWindow: 1050000,
19939
20137
  maxTokens: 128000,
@@ -19969,7 +20167,7 @@ export const MODELS = {
19969
20167
  input: 4,
19970
20168
  output: 20,
19971
20169
  cacheRead: 0.39999999999999997,
19972
- cacheWrite: 2.5,
20170
+ cacheWrite: 5,
19973
20171
  },
19974
20172
  contextWindow: 1050000,
19975
20173
  maxTokens: 128000,
@@ -20005,7 +20203,43 @@ export const MODELS = {
20005
20203
  input: 4,
20006
20204
  output: 24,
20007
20205
  cacheRead: 0.39999999999999997,
20008
- cacheWrite: 2.5,
20206
+ cacheWrite: 5,
20207
+ },
20208
+ contextWindow: 1050000,
20209
+ maxTokens: 128000,
20210
+ },
20211
+ "openai/gpt-6-astra": {
20212
+ id: "openai/gpt-6-astra",
20213
+ name: "GPT-6 Astra",
20214
+ api: "anthropic-messages",
20215
+ provider: "vercel-ai-gateway",
20216
+ baseUrl: "https://ai-gateway.vercel.sh",
20217
+ reasoning: true,
20218
+ thinkingLevelMap: { "xhigh": "xhigh" },
20219
+ input: ["text", "image"],
20220
+ cost: {
20221
+ input: 10,
20222
+ output: 50,
20223
+ cacheRead: 1,
20224
+ cacheWrite: 12.5,
20225
+ },
20226
+ contextWindow: 1050000,
20227
+ maxTokens: 128000,
20228
+ },
20229
+ "openai/gpt-6-astra-fast": {
20230
+ id: "openai/gpt-6-astra-fast",
20231
+ name: "GPT-6 Astra (Fast)",
20232
+ api: "anthropic-messages",
20233
+ provider: "vercel-ai-gateway",
20234
+ baseUrl: "https://ai-gateway.vercel.sh",
20235
+ reasoning: true,
20236
+ thinkingLevelMap: { "xhigh": "xhigh" },
20237
+ input: ["text", "image"],
20238
+ cost: {
20239
+ input: 20,
20240
+ output: 100,
20241
+ cacheRead: 2,
20242
+ cacheWrite: 25,
20009
20243
  },
20010
20244
  contextWindow: 1050000,
20011
20245
  maxTokens: 128000,
@@ -20835,14 +21069,31 @@ export const MODELS = {
20835
21069
  reasoning: true,
20836
21070
  input: ["text"],
20837
21071
  cost: {
20838
- input: 1.4,
20839
- output: 4.4,
20840
- cacheRead: 0.14,
21072
+ input: 0.7,
21073
+ output: 2.2,
21074
+ cacheRead: 0.13,
20841
21075
  cacheWrite: 0,
20842
21076
  },
20843
21077
  contextWindow: 1000000,
20844
21078
  maxTokens: 1000000,
20845
21079
  },
21080
+ "zai/glm-5.3-fast": {
21081
+ id: "zai/glm-5.3-fast",
21082
+ name: "GLM 5.3 Fast",
21083
+ api: "anthropic-messages",
21084
+ provider: "vercel-ai-gateway",
21085
+ baseUrl: "https://ai-gateway.vercel.sh",
21086
+ reasoning: true,
21087
+ input: ["text"],
21088
+ cost: {
21089
+ input: 2.0999999999999996,
21090
+ output: 6.6000000000000005,
21091
+ cacheRead: 0.21,
21092
+ cacheWrite: 0,
21093
+ },
21094
+ contextWindow: 1048576,
21095
+ maxTokens: 262144,
21096
+ },
20846
21097
  "zai/glm-5.3-flash": {
20847
21098
  id: "zai/glm-5.3-flash",
20848
21099
  name: "GLM 5.3 Flash",
@@ -20860,6 +21111,23 @@ export const MODELS = {
20860
21111
  contextWindow: 1000000,
20861
21112
  maxTokens: 131000,
20862
21113
  },
21114
+ "zai/glm-5.3-promo-50": {
21115
+ id: "zai/glm-5.3-promo-50",
21116
+ name: "GLM 5.3 (50% off)",
21117
+ api: "anthropic-messages",
21118
+ provider: "vercel-ai-gateway",
21119
+ baseUrl: "https://ai-gateway.vercel.sh",
21120
+ reasoning: true,
21121
+ input: ["text"],
21122
+ cost: {
21123
+ input: 0.7,
21124
+ output: 2.2,
21125
+ cacheRead: 0.13,
21126
+ cacheWrite: 0,
21127
+ },
21128
+ contextWindow: 1048576,
21129
+ maxTokens: 1048576,
21130
+ },
20863
21131
  "zai/glm-5v-turbo": {
20864
21132
  id: "zai/glm-5v-turbo",
20865
21133
  name: "GLM 5V Turbo",