@caupulican/pi-ai 0.80.23 → 0.80.26

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.
@@ -3746,6 +3746,24 @@ export const MODELS = {
3746
3746
  contextWindow: 131072,
3747
3747
  maxTokens: 131072,
3748
3748
  },
3749
+ "@cf/zai-org/glm-5.2": {
3750
+ id: "@cf/zai-org/glm-5.2",
3751
+ name: "Glm 5.2",
3752
+ api: "openai-completions",
3753
+ provider: "cloudflare-workers-ai",
3754
+ baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
3755
+ compat: { "sendSessionAffinityHeaders": true },
3756
+ reasoning: true,
3757
+ input: ["text"],
3758
+ cost: {
3759
+ input: 1.4,
3760
+ output: 4.4,
3761
+ cacheRead: 0.26,
3762
+ cacheWrite: 0,
3763
+ },
3764
+ contextWindow: 262144,
3765
+ maxTokens: 262144,
3766
+ },
3749
3767
  },
3750
3768
  "deepseek": {
3751
3769
  "deepseek-v4-flash": {
@@ -3800,7 +3818,7 @@ export const MODELS = {
3800
3818
  cost: {
3801
3819
  input: 0.14,
3802
3820
  output: 0.28,
3803
- cacheRead: 0.03,
3821
+ cacheRead: 0.028,
3804
3822
  cacheWrite: 0,
3805
3823
  },
3806
3824
  contextWindow: 1000000,
@@ -3842,6 +3860,24 @@ export const MODELS = {
3842
3860
  contextWindow: 202800,
3843
3861
  maxTokens: 131072,
3844
3862
  },
3863
+ "accounts/fireworks/models/glm-5p2": {
3864
+ id: "accounts/fireworks/models/glm-5p2",
3865
+ name: "GLM 5.2",
3866
+ api: "anthropic-messages",
3867
+ provider: "fireworks",
3868
+ baseUrl: "https://api.fireworks.ai/inference",
3869
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
3870
+ reasoning: true,
3871
+ input: ["text"],
3872
+ cost: {
3873
+ input: 1.4,
3874
+ output: 4.4,
3875
+ cacheRead: 0.26,
3876
+ cacheWrite: 0,
3877
+ },
3878
+ contextWindow: 1048576,
3879
+ maxTokens: 131072,
3880
+ },
3845
3881
  "accounts/fireworks/models/gpt-oss-120b": {
3846
3882
  id: "accounts/fireworks/models/gpt-oss-120b",
3847
3883
  name: "GPT OSS 120B",
@@ -3878,9 +3914,9 @@ export const MODELS = {
3878
3914
  contextWindow: 131072,
3879
3915
  maxTokens: 32768,
3880
3916
  },
3881
- "accounts/fireworks/models/kimi-k2p5": {
3882
- id: "accounts/fireworks/models/kimi-k2p5",
3883
- name: "Kimi K2.5",
3917
+ "accounts/fireworks/models/kimi-k2p6": {
3918
+ id: "accounts/fireworks/models/kimi-k2p6",
3919
+ name: "Kimi K2.6",
3884
3920
  api: "anthropic-messages",
3885
3921
  provider: "fireworks",
3886
3922
  baseUrl: "https://api.fireworks.ai/inference",
@@ -3888,17 +3924,17 @@ export const MODELS = {
3888
3924
  reasoning: true,
3889
3925
  input: ["text", "image"],
3890
3926
  cost: {
3891
- input: 0.6,
3892
- output: 3,
3893
- cacheRead: 0.1,
3927
+ input: 0.95,
3928
+ output: 4,
3929
+ cacheRead: 0.16,
3894
3930
  cacheWrite: 0,
3895
3931
  },
3896
- contextWindow: 256000,
3897
- maxTokens: 256000,
3932
+ contextWindow: 262000,
3933
+ maxTokens: 262000,
3898
3934
  },
3899
- "accounts/fireworks/models/kimi-k2p6": {
3900
- id: "accounts/fireworks/models/kimi-k2p6",
3901
- name: "Kimi K2.6",
3935
+ "accounts/fireworks/models/kimi-k2p7-code": {
3936
+ id: "accounts/fireworks/models/kimi-k2p7-code",
3937
+ name: "Kimi K2.7 Code",
3902
3938
  api: "anthropic-messages",
3903
3939
  provider: "fireworks",
3904
3940
  baseUrl: "https://api.fireworks.ai/inference",
@@ -3908,15 +3944,15 @@ export const MODELS = {
3908
3944
  cost: {
3909
3945
  input: 0.95,
3910
3946
  output: 4,
3911
- cacheRead: 0.16,
3947
+ cacheRead: 0.19,
3912
3948
  cacheWrite: 0,
3913
3949
  },
3914
3950
  contextWindow: 262000,
3915
3951
  maxTokens: 262000,
3916
3952
  },
3917
- "accounts/fireworks/models/minimax-m2p5": {
3918
- id: "accounts/fireworks/models/minimax-m2p5",
3919
- name: "MiniMax-M2.5",
3953
+ "accounts/fireworks/models/minimax-m2p7": {
3954
+ id: "accounts/fireworks/models/minimax-m2p7",
3955
+ name: "MiniMax-M2.7",
3920
3956
  api: "anthropic-messages",
3921
3957
  provider: "fireworks",
3922
3958
  baseUrl: "https://api.fireworks.ai/inference",
@@ -3926,15 +3962,15 @@ export const MODELS = {
3926
3962
  cost: {
3927
3963
  input: 0.3,
3928
3964
  output: 1.2,
3929
- cacheRead: 0.03,
3965
+ cacheRead: 0.06,
3930
3966
  cacheWrite: 0,
3931
3967
  },
3932
3968
  contextWindow: 196608,
3933
3969
  maxTokens: 196608,
3934
3970
  },
3935
- "accounts/fireworks/models/minimax-m2p7": {
3936
- id: "accounts/fireworks/models/minimax-m2p7",
3937
- name: "MiniMax-M2.7",
3971
+ "accounts/fireworks/models/minimax-m3": {
3972
+ id: "accounts/fireworks/models/minimax-m3",
3973
+ name: "MiniMax-M3",
3938
3974
  api: "anthropic-messages",
3939
3975
  provider: "fireworks",
3940
3976
  baseUrl: "https://api.fireworks.ai/inference",
@@ -3947,12 +3983,12 @@ export const MODELS = {
3947
3983
  cacheRead: 0.06,
3948
3984
  cacheWrite: 0,
3949
3985
  },
3950
- contextWindow: 196608,
3951
- maxTokens: 196608,
3986
+ contextWindow: 512000,
3987
+ maxTokens: 512000,
3952
3988
  },
3953
- "accounts/fireworks/models/qwen3p6-plus": {
3954
- id: "accounts/fireworks/models/qwen3p6-plus",
3955
- name: "Qwen 3.6 Plus",
3989
+ "accounts/fireworks/models/qwen3p7-plus": {
3990
+ id: "accounts/fireworks/models/qwen3p7-plus",
3991
+ name: "Qwen 3.7 Plus",
3956
3992
  api: "anthropic-messages",
3957
3993
  provider: "fireworks",
3958
3994
  baseUrl: "https://api.fireworks.ai/inference",
@@ -3960,9 +3996,9 @@ export const MODELS = {
3960
3996
  reasoning: true,
3961
3997
  input: ["text", "image"],
3962
3998
  cost: {
3963
- input: 0.5,
3964
- output: 3,
3965
- cacheRead: 0.1,
3999
+ input: 0.4,
4000
+ output: 1.6,
4001
+ cacheRead: 0.08,
3966
4002
  cacheWrite: 0,
3967
4003
  },
3968
4004
  contextWindow: 262144,
@@ -4022,6 +4058,24 @@ export const MODELS = {
4022
4058
  contextWindow: 262000,
4023
4059
  maxTokens: 262000,
4024
4060
  },
4061
+ "accounts/fireworks/routers/kimi-k2p7-code-fast": {
4062
+ id: "accounts/fireworks/routers/kimi-k2p7-code-fast",
4063
+ name: "Kimi K2.7 Code Fast",
4064
+ api: "anthropic-messages",
4065
+ provider: "fireworks",
4066
+ baseUrl: "https://api.fireworks.ai/inference",
4067
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4068
+ reasoning: true,
4069
+ input: ["text", "image"],
4070
+ cost: {
4071
+ input: 1.9,
4072
+ output: 8,
4073
+ cacheRead: 0.38,
4074
+ cacheWrite: 0,
4075
+ },
4076
+ contextWindow: 262000,
4077
+ maxTokens: 262000,
4078
+ },
4025
4079
  },
4026
4080
  "github-copilot": {
4027
4081
  "claude-fable-5": {
@@ -4746,6 +4800,42 @@ export const MODELS = {
4746
4800
  contextWindow: 262144,
4747
4801
  maxTokens: 32768,
4748
4802
  },
4803
+ "gemma-4-E2B-it": {
4804
+ id: "gemma-4-E2B-it",
4805
+ name: "Gemma 4 E2B IT",
4806
+ api: "google-generative-ai",
4807
+ provider: "google",
4808
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4809
+ reasoning: true,
4810
+ thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4811
+ input: ["text", "image"],
4812
+ cost: {
4813
+ input: 0,
4814
+ output: 0,
4815
+ cacheRead: 0,
4816
+ cacheWrite: 0,
4817
+ },
4818
+ contextWindow: 131072,
4819
+ maxTokens: 8192,
4820
+ },
4821
+ "gemma-4-E4B-it": {
4822
+ id: "gemma-4-E4B-it",
4823
+ name: "Gemma 4 E4B IT",
4824
+ api: "google-generative-ai",
4825
+ provider: "google",
4826
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4827
+ reasoning: true,
4828
+ thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4829
+ input: ["text", "image"],
4830
+ cost: {
4831
+ input: 0,
4832
+ output: 0,
4833
+ cacheRead: 0,
4834
+ cacheWrite: 0,
4835
+ },
4836
+ contextWindow: 131072,
4837
+ maxTokens: 8192,
4838
+ },
4749
4839
  },
4750
4840
  "google-vertex": {
4751
4841
  "gemini-1.5-flash": {
@@ -6299,6 +6389,24 @@ export const MODELS = {
6299
6389
  contextWindow: 262144,
6300
6390
  maxTokens: 262144,
6301
6391
  },
6392
+ "kimi-k2.7-code-highspeed": {
6393
+ id: "kimi-k2.7-code-highspeed",
6394
+ name: "Kimi K2.7 Code HighSpeed",
6395
+ api: "openai-completions",
6396
+ provider: "moonshotai",
6397
+ baseUrl: "https://api.moonshot.ai/v1",
6398
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6399
+ reasoning: true,
6400
+ input: ["text", "image"],
6401
+ cost: {
6402
+ input: 1.9,
6403
+ output: 8,
6404
+ cacheRead: 0.38,
6405
+ cacheWrite: 0,
6406
+ },
6407
+ contextWindow: 262144,
6408
+ maxTokens: 262144,
6409
+ },
6302
6410
  },
6303
6411
  "moonshotai-cn": {
6304
6412
  "kimi-k2-0711-preview": {
@@ -6427,6 +6535,42 @@ export const MODELS = {
6427
6535
  contextWindow: 262144,
6428
6536
  maxTokens: 262144,
6429
6537
  },
6538
+ "kimi-k2.7-code": {
6539
+ id: "kimi-k2.7-code",
6540
+ name: "Kimi K2.7 Code",
6541
+ api: "openai-completions",
6542
+ provider: "moonshotai-cn",
6543
+ baseUrl: "https://api.moonshot.cn/v1",
6544
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6545
+ reasoning: true,
6546
+ input: ["text", "image"],
6547
+ cost: {
6548
+ input: 0.95,
6549
+ output: 4,
6550
+ cacheRead: 0.19,
6551
+ cacheWrite: 0,
6552
+ },
6553
+ contextWindow: 262144,
6554
+ maxTokens: 262144,
6555
+ },
6556
+ "kimi-k2.7-code-highspeed": {
6557
+ id: "kimi-k2.7-code-highspeed",
6558
+ name: "Kimi K2.7 Code HighSpeed",
6559
+ api: "openai-completions",
6560
+ provider: "moonshotai-cn",
6561
+ baseUrl: "https://api.moonshot.cn/v1",
6562
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6563
+ reasoning: true,
6564
+ input: ["text", "image"],
6565
+ cost: {
6566
+ input: 1.9,
6567
+ output: 8,
6568
+ cacheRead: 0.38,
6569
+ cacheWrite: 0,
6570
+ },
6571
+ contextWindow: 262144,
6572
+ maxTokens: 262144,
6573
+ },
6430
6574
  },
6431
6575
  "openai": {
6432
6576
  "gpt-4": {
@@ -7296,23 +7440,6 @@ export const MODELS = {
7296
7440
  contextWindow: 200000,
7297
7441
  maxTokens: 32000,
7298
7442
  },
7299
- "claude-fable-5": {
7300
- id: "claude-fable-5",
7301
- name: "Claude Fable 5",
7302
- api: "anthropic-messages",
7303
- provider: "opencode",
7304
- baseUrl: "https://opencode.ai/zen",
7305
- reasoning: true,
7306
- input: ["text", "image"],
7307
- cost: {
7308
- input: 10,
7309
- output: 50,
7310
- cacheRead: 1,
7311
- cacheWrite: 12.5,
7312
- },
7313
- contextWindow: 1000000,
7314
- maxTokens: 128000,
7315
- },
7316
7443
  "claude-haiku-4-5": {
7317
7444
  id: "claude-haiku-4-5",
7318
7445
  name: "Claude Haiku 4.5",
@@ -8119,26 +8246,26 @@ export const MODELS = {
8119
8246
  contextWindow: 1000000,
8120
8247
  maxTokens: 384000,
8121
8248
  },
8122
- "glm-5": {
8123
- id: "glm-5",
8124
- name: "GLM-5",
8249
+ "glm-5.1": {
8250
+ id: "glm-5.1",
8251
+ name: "GLM-5.1",
8125
8252
  api: "openai-completions",
8126
8253
  provider: "opencode-go",
8127
8254
  baseUrl: "https://opencode.ai/zen/go/v1",
8128
8255
  reasoning: true,
8129
8256
  input: ["text"],
8130
8257
  cost: {
8131
- input: 1,
8132
- output: 3.2,
8133
- cacheRead: 0.2,
8258
+ input: 1.4,
8259
+ output: 4.4,
8260
+ cacheRead: 0.26,
8134
8261
  cacheWrite: 0,
8135
8262
  },
8136
8263
  contextWindow: 202752,
8137
8264
  maxTokens: 32768,
8138
8265
  },
8139
- "glm-5.1": {
8140
- id: "glm-5.1",
8141
- name: "GLM-5.1",
8266
+ "glm-5.2": {
8267
+ id: "glm-5.2",
8268
+ name: "GLM-5.2",
8142
8269
  api: "openai-completions",
8143
8270
  provider: "opencode-go",
8144
8271
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -8150,8 +8277,8 @@ export const MODELS = {
8150
8277
  cacheRead: 0.26,
8151
8278
  cacheWrite: 0,
8152
8279
  },
8153
- contextWindow: 202752,
8154
- maxTokens: 32768,
8280
+ contextWindow: 1000000,
8281
+ maxTokens: 131072,
8155
8282
  },
8156
8283
  "kimi-k2.6": {
8157
8284
  id: "kimi-k2.6",
@@ -8861,6 +8988,23 @@ export const MODELS = {
8861
8988
  contextWindow: 128000,
8862
8989
  maxTokens: 4000,
8863
8990
  },
8991
+ "cohere/north-mini-code:free": {
8992
+ id: "cohere/north-mini-code:free",
8993
+ name: "Cohere: North Mini Code (free)",
8994
+ api: "openai-completions",
8995
+ provider: "openrouter",
8996
+ baseUrl: "https://openrouter.ai/api/v1",
8997
+ reasoning: true,
8998
+ input: ["text"],
8999
+ cost: {
9000
+ input: 0,
9001
+ output: 0,
9002
+ cacheRead: 0,
9003
+ cacheWrite: 0,
9004
+ },
9005
+ contextWindow: 256000,
9006
+ maxTokens: 64000,
9007
+ },
8864
9008
  "deepseek/deepseek-chat": {
8865
9009
  id: "deepseek/deepseek-chat",
8866
9010
  name: "DeepSeek: DeepSeek V3",
@@ -9008,13 +9152,13 @@ export const MODELS = {
9008
9152
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
9009
9153
  input: ["text"],
9010
9154
  cost: {
9011
- input: 0.098,
9012
- output: 0.196,
9155
+ input: 0.09,
9156
+ output: 0.18,
9013
9157
  cacheRead: 0.02,
9014
9158
  cacheWrite: 0,
9015
9159
  },
9016
9160
  contextWindow: 1048576,
9017
- maxTokens: 4096,
9161
+ maxTokens: 65536,
9018
9162
  },
9019
9163
  "deepseek/deepseek-v4-pro": {
9020
9164
  id: "deepseek/deepseek-v4-pro",
@@ -9169,7 +9313,7 @@ export const MODELS = {
9169
9313
  cacheWrite: 0.08333333333333334,
9170
9314
  },
9171
9315
  contextWindow: 1048576,
9172
- maxTokens: 65536,
9316
+ maxTokens: 65535,
9173
9317
  },
9174
9318
  "google/gemini-3.1-flash-lite": {
9175
9319
  id: "google/gemini-3.1-flash-lite",
@@ -9356,7 +9500,7 @@ export const MODELS = {
9356
9500
  cacheWrite: 0,
9357
9501
  },
9358
9502
  contextWindow: 262144,
9359
- maxTokens: 32768,
9503
+ maxTokens: 8192,
9360
9504
  },
9361
9505
  "ibm-granite/granite-4.1-8b": {
9362
9506
  id: "ibm-granite/granite-4.1-8b",
@@ -10049,13 +10193,13 @@ export const MODELS = {
10049
10193
  reasoning: true,
10050
10194
  input: ["text", "image"],
10051
10195
  cost: {
10052
- input: 0.95,
10053
- output: 4,
10054
- cacheRead: 0.19,
10196
+ input: 0.74,
10197
+ output: 3.5,
10198
+ cacheRead: 0.15,
10055
10199
  cacheWrite: 0,
10056
10200
  },
10057
10201
  contextWindow: 262144,
10058
- maxTokens: 4096,
10202
+ maxTokens: 16384,
10059
10203
  },
10060
10204
  "nex-agi/nex-n2-pro:free": {
10061
10205
  id: "nex-agi/nex-n2-pro:free",
@@ -10186,8 +10330,8 @@ export const MODELS = {
10186
10330
  input: ["text"],
10187
10331
  cost: {
10188
10332
  input: 0.5,
10189
- output: 2.5,
10190
- cacheRead: 0.15,
10333
+ output: 2.2,
10334
+ cacheRead: 0.09999999999999999,
10191
10335
  cacheWrite: 0,
10192
10336
  },
10193
10337
  contextWindow: 1000000,
@@ -11002,7 +11146,7 @@ export const MODELS = {
11002
11146
  cacheWrite: 0,
11003
11147
  },
11004
11148
  contextWindow: 131072,
11005
- maxTokens: 8192,
11149
+ maxTokens: 32768,
11006
11150
  },
11007
11151
  "openai/gpt-oss-safeguard-20b": {
11008
11152
  id: "openai/gpt-oss-safeguard-20b",
@@ -11852,7 +11996,7 @@ export const MODELS = {
11852
11996
  cacheWrite: 0,
11853
11997
  },
11854
11998
  contextWindow: 262144,
11855
- maxTokens: 262144,
11999
+ maxTokens: 81920,
11856
12000
  },
11857
12001
  "qwen/qwen3.5-397b-a17b": {
11858
12002
  id: "qwen/qwen3.5-397b-a17b",
@@ -11863,13 +12007,13 @@ export const MODELS = {
11863
12007
  reasoning: true,
11864
12008
  input: ["text", "image"],
11865
12009
  cost: {
11866
- input: 0.39,
11867
- output: 2.34,
12010
+ input: 0.385,
12011
+ output: 2.4499999999999997,
11868
12012
  cacheRead: 0,
11869
12013
  cacheWrite: 0,
11870
12014
  },
11871
- contextWindow: 262144,
11872
- maxTokens: 65536,
12015
+ contextWindow: 256000,
12016
+ maxTokens: 4096,
11873
12017
  },
11874
12018
  "qwen/qwen3.5-9b": {
11875
12019
  id: "qwen/qwen3.5-9b",
@@ -12152,13 +12296,13 @@ export const MODELS = {
12152
12296
  reasoning: true,
12153
12297
  input: ["text"],
12154
12298
  cost: {
12155
- input: 0.063,
12156
- output: 0.21,
12157
- cacheRead: 0.020999999999999998,
12299
+ input: 0.06599999999999999,
12300
+ output: 0.26,
12301
+ cacheRead: 0.029,
12158
12302
  cacheWrite: 0,
12159
12303
  },
12160
12304
  contextWindow: 262144,
12161
- maxTokens: 4096,
12305
+ maxTokens: 262144,
12162
12306
  },
12163
12307
  "thedrummer/rocinante-12b": {
12164
12308
  id: "thedrummer/rocinante-12b",
@@ -12262,23 +12406,6 @@ export const MODELS = {
12262
12406
  contextWindow: 256000,
12263
12407
  maxTokens: 4096,
12264
12408
  },
12265
- "xiaomi/mimo-v2-flash": {
12266
- id: "xiaomi/mimo-v2-flash",
12267
- name: "Xiaomi: MiMo-V2-Flash",
12268
- api: "openai-completions",
12269
- provider: "openrouter",
12270
- baseUrl: "https://openrouter.ai/api/v1",
12271
- reasoning: true,
12272
- input: ["text"],
12273
- cost: {
12274
- input: 0.09999999999999999,
12275
- output: 0.3,
12276
- cacheRead: 0.01,
12277
- cacheWrite: 0,
12278
- },
12279
- contextWindow: 262144,
12280
- maxTokens: 65536,
12281
- },
12282
12409
  "xiaomi/mimo-v2.5": {
12283
12410
  id: "xiaomi/mimo-v2.5",
12284
12411
  name: "Xiaomi: MiMo-V2.5",
@@ -12339,13 +12466,13 @@ export const MODELS = {
12339
12466
  reasoning: true,
12340
12467
  input: ["text"],
12341
12468
  cost: {
12342
- input: 0.125,
12469
+ input: 0.13,
12343
12470
  output: 0.85,
12344
- cacheRead: 0.06,
12471
+ cacheRead: 0.024999999999999998,
12345
12472
  cacheWrite: 0,
12346
12473
  },
12347
12474
  contextWindow: 131072,
12348
- maxTokens: 131070,
12475
+ maxTokens: 98304,
12349
12476
  },
12350
12477
  "z-ai/glm-4.5v": {
12351
12478
  id: "z-ai/glm-4.5v",
@@ -12483,6 +12610,23 @@ export const MODELS = {
12483
12610
  contextWindow: 202752,
12484
12611
  maxTokens: 4096,
12485
12612
  },
12613
+ "z-ai/glm-5.2": {
12614
+ id: "z-ai/glm-5.2",
12615
+ name: "Z.ai: GLM 5.2",
12616
+ api: "openai-completions",
12617
+ provider: "openrouter",
12618
+ baseUrl: "https://openrouter.ai/api/v1",
12619
+ reasoning: true,
12620
+ input: ["text"],
12621
+ cost: {
12622
+ input: 1.4,
12623
+ output: 4.4,
12624
+ cacheRead: 0.26,
12625
+ cacheWrite: 0,
12626
+ },
12627
+ contextWindow: 1048576,
12628
+ maxTokens: 262144,
12629
+ },
12486
12630
  "~anthropic/claude-fable-latest": {
12487
12631
  id: "~anthropic/claude-fable-latest",
12488
12632
  name: "Anthropic: Claude Fable Latest",
@@ -12657,6 +12801,25 @@ export const MODELS = {
12657
12801
  contextWindow: 202752,
12658
12802
  maxTokens: 131072,
12659
12803
  },
12804
+ "MiniMaxAI/MiniMax-M3": {
12805
+ id: "MiniMaxAI/MiniMax-M3",
12806
+ name: "MiniMax-M3",
12807
+ api: "openai-completions",
12808
+ provider: "together",
12809
+ baseUrl: "https://api.together.ai/v1",
12810
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12811
+ reasoning: true,
12812
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12813
+ input: ["text", "image"],
12814
+ cost: {
12815
+ input: 0.3,
12816
+ output: 1.2,
12817
+ cacheRead: 0.06,
12818
+ cacheWrite: 0,
12819
+ },
12820
+ contextWindow: 524288,
12821
+ maxTokens: 250000,
12822
+ },
12660
12823
  "Qwen/Qwen2.5-7B-Instruct-Turbo": {
12661
12824
  id: "Qwen/Qwen2.5-7B-Instruct-Turbo",
12662
12825
  name: "Qwen 2.5 7B Instruct Turbo",
@@ -12760,8 +12923,8 @@ export const MODELS = {
12760
12923
  reasoning: false,
12761
12924
  input: ["text"],
12762
12925
  cost: {
12763
- input: 2.5,
12764
- output: 7.5,
12926
+ input: 1.25,
12927
+ output: 3.75,
12765
12928
  cacheRead: 0,
12766
12929
  cacheWrite: 0,
12767
12930
  },
@@ -12861,6 +13024,25 @@ export const MODELS = {
12861
13024
  contextWindow: 262144,
12862
13025
  maxTokens: 131000,
12863
13026
  },
13027
+ "moonshotai/Kimi-K2.7-Code": {
13028
+ id: "moonshotai/Kimi-K2.7-Code",
13029
+ name: "Kimi K2.7 Code",
13030
+ api: "openai-completions",
13031
+ provider: "together",
13032
+ baseUrl: "https://api.together.ai/v1",
13033
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13034
+ reasoning: true,
13035
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13036
+ input: ["text"],
13037
+ cost: {
13038
+ input: 0.95,
13039
+ output: 4,
13040
+ cacheRead: 0.19,
13041
+ cacheWrite: 0,
13042
+ },
13043
+ contextWindow: 262144,
13044
+ maxTokens: 131072,
13045
+ },
12864
13046
  "nvidia/nemotron-3-ultra-550b-a55b": {
12865
13047
  id: "nvidia/nemotron-3-ultra-550b-a55b",
12866
13048
  name: "Nemotron 3 Ultra 550B A55B",
@@ -13366,23 +13548,6 @@ export const MODELS = {
13366
13548
  contextWindow: 200000,
13367
13549
  maxTokens: 8192,
13368
13550
  },
13369
- "anthropic/claude-fable-5": {
13370
- id: "anthropic/claude-fable-5",
13371
- name: "Claude Fable 5",
13372
- api: "anthropic-messages",
13373
- provider: "vercel-ai-gateway",
13374
- baseUrl: "https://ai-gateway.vercel.sh",
13375
- reasoning: true,
13376
- input: ["text", "image"],
13377
- cost: {
13378
- input: 10,
13379
- output: 50,
13380
- cacheRead: 1,
13381
- cacheWrite: 12.5,
13382
- },
13383
- contextWindow: 1000000,
13384
- maxTokens: 128000,
13385
- },
13386
13551
  "anthropic/claude-haiku-4.5": {
13387
13552
  id: "anthropic/claude-haiku-4.5",
13388
13553
  name: "Claude Haiku 4.5",
@@ -14419,13 +14584,13 @@ export const MODELS = {
14419
14584
  reasoning: false,
14420
14585
  input: ["text"],
14421
14586
  cost: {
14422
- input: 0.02,
14423
- output: 0.04,
14587
+ input: 0.15,
14588
+ output: 0.15,
14424
14589
  cacheRead: 0,
14425
14590
  cacheWrite: 0,
14426
14591
  },
14427
- contextWindow: 131072,
14428
- maxTokens: 131072,
14592
+ contextWindow: 128000,
14593
+ maxTokens: 128000,
14429
14594
  },
14430
14595
  "mistral/mistral-small": {
14431
14596
  id: "mistral/mistral-small",
@@ -14563,6 +14728,23 @@ export const MODELS = {
14563
14728
  contextWindow: 256000,
14564
14729
  maxTokens: 32768,
14565
14730
  },
14731
+ "moonshotai/kimi-k2.7-code-highspeed": {
14732
+ id: "moonshotai/kimi-k2.7-code-highspeed",
14733
+ name: "Kimi K2.7 Code High Speed",
14734
+ api: "anthropic-messages",
14735
+ provider: "vercel-ai-gateway",
14736
+ baseUrl: "https://ai-gateway.vercel.sh",
14737
+ reasoning: true,
14738
+ input: ["text", "image"],
14739
+ cost: {
14740
+ input: 1.9,
14741
+ output: 8,
14742
+ cacheRead: 0.38,
14743
+ cacheWrite: 0,
14744
+ },
14745
+ contextWindow: 262144,
14746
+ maxTokens: 32768,
14747
+ },
14566
14748
  "nvidia/nemotron-3-super-120b-a12b": {
14567
14749
  id: "nvidia/nemotron-3-super-120b-a12b",
14568
14750
  name: "NVIDIA Nemotron 3 Super 120B A12B",
@@ -15799,6 +15981,23 @@ export const MODELS = {
15799
15981
  contextWindow: 202800,
15800
15982
  maxTokens: 64000,
15801
15983
  },
15984
+ "zai/glm-5.2": {
15985
+ id: "zai/glm-5.2",
15986
+ name: "GLM 5.2",
15987
+ api: "anthropic-messages",
15988
+ provider: "vercel-ai-gateway",
15989
+ baseUrl: "https://ai-gateway.vercel.sh",
15990
+ reasoning: true,
15991
+ input: ["text"],
15992
+ cost: {
15993
+ input: 1.5,
15994
+ output: 4.5,
15995
+ cacheRead: 0.3,
15996
+ cacheWrite: 0,
15997
+ },
15998
+ contextWindow: 1000000,
15999
+ maxTokens: 128000,
16000
+ },
15802
16001
  "zai/glm-5v-turbo": {
15803
16002
  id: "zai/glm-5v-turbo",
15804
16003
  name: "GLM 5V Turbo",
@@ -16397,6 +16596,24 @@ export const MODELS = {
16397
16596
  contextWindow: 200000,
16398
16597
  maxTokens: 131072,
16399
16598
  },
16599
+ "glm-5.2": {
16600
+ id: "glm-5.2",
16601
+ name: "GLM-5.2",
16602
+ api: "openai-completions",
16603
+ provider: "zai",
16604
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
16605
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
16606
+ reasoning: true,
16607
+ input: ["text"],
16608
+ cost: {
16609
+ input: 0,
16610
+ output: 0,
16611
+ cacheRead: 0,
16612
+ cacheWrite: 0,
16613
+ },
16614
+ contextWindow: 1000000,
16615
+ maxTokens: 131072,
16616
+ },
16400
16617
  "glm-5v-turbo": {
16401
16618
  id: "glm-5v-turbo",
16402
16619
  name: "GLM-5V-Turbo",