@dreb/ai 2.40.2 → 2.42.0

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.
@@ -3662,9 +3662,9 @@ export const MODELS = {
3662
3662
  reasoning: true,
3663
3663
  input: ["text", "image"],
3664
3664
  cost: {
3665
- input: 0.3,
3666
- output: 2.5,
3667
- cacheRead: 0.075,
3665
+ input: 1.5,
3666
+ output: 9,
3667
+ cacheRead: 0.15,
3668
3668
  cacheWrite: 0,
3669
3669
  },
3670
3670
  contextWindow: 1048576,
@@ -3679,8 +3679,8 @@ export const MODELS = {
3679
3679
  reasoning: true,
3680
3680
  input: ["text", "image"],
3681
3681
  cost: {
3682
- input: 0.1,
3683
- output: 0.4,
3682
+ input: 0.25,
3683
+ output: 1.5,
3684
3684
  cacheRead: 0.025,
3685
3685
  cacheWrite: 0,
3686
3686
  },
@@ -5200,7 +5200,7 @@ export const MODELS = {
5200
5200
  provider: "kimi-coding",
5201
5201
  baseUrl: "https://api.kimi.com/coding",
5202
5202
  reasoning: true,
5203
- input: ["text", "image"],
5203
+ input: ["text"],
5204
5204
  cost: {
5205
5205
  input: 0,
5206
5206
  output: 0,
@@ -5210,9 +5210,9 @@ export const MODELS = {
5210
5210
  contextWindow: 262144,
5211
5211
  maxTokens: 32768,
5212
5212
  },
5213
- "k2p6": {
5214
- id: "k2p6",
5215
- name: "Kimi K2.6",
5213
+ "k2p7": {
5214
+ id: "k2p7",
5215
+ name: "Kimi K2.7 Code",
5216
5216
  api: "anthropic-messages",
5217
5217
  provider: "kimi-coding",
5218
5218
  baseUrl: "https://api.kimi.com/coding",
@@ -5227,9 +5227,26 @@ export const MODELS = {
5227
5227
  contextWindow: 262144,
5228
5228
  maxTokens: 32768,
5229
5229
  },
5230
- "k2p7": {
5231
- id: "k2p7",
5232
- name: "Kimi K2.7 Code",
5230
+ "k3": {
5231
+ id: "k3",
5232
+ name: "Kimi K3",
5233
+ api: "anthropic-messages",
5234
+ provider: "kimi-coding",
5235
+ baseUrl: "https://api.kimi.com/coding",
5236
+ reasoning: true,
5237
+ input: ["text", "image"],
5238
+ cost: {
5239
+ input: 0,
5240
+ output: 0,
5241
+ cacheRead: 0,
5242
+ cacheWrite: 0,
5243
+ },
5244
+ contextWindow: 1048576,
5245
+ maxTokens: 131072,
5246
+ },
5247
+ "kimi-for-coding-highspeed": {
5248
+ id: "kimi-for-coding-highspeed",
5249
+ name: "Kimi For Coding HighSpeed",
5233
5250
  api: "anthropic-messages",
5234
5251
  provider: "kimi-coding",
5235
5252
  baseUrl: "https://api.kimi.com/coding",
@@ -7769,9 +7786,9 @@ export const MODELS = {
7769
7786
  reasoning: true,
7770
7787
  input: ["text"],
7771
7788
  cost: {
7772
- input: 0.14,
7773
- output: 0.28,
7774
- cacheRead: 0.0028,
7789
+ input: 0.435,
7790
+ output: 0.87,
7791
+ cacheRead: 0.003625,
7775
7792
  cacheWrite: 0,
7776
7793
  },
7777
7794
  contextWindow: 1000000,
@@ -7786,9 +7803,9 @@ export const MODELS = {
7786
7803
  reasoning: true,
7787
7804
  input: ["text"],
7788
7805
  cost: {
7789
- input: 1.74,
7790
- output: 3.48,
7791
- cacheRead: 0.0145,
7806
+ input: 0.435,
7807
+ output: 0.87,
7808
+ cacheRead: 0.003625,
7792
7809
  cacheWrite: 0,
7793
7810
  },
7794
7811
  contextWindow: 1000000,
@@ -7828,6 +7845,23 @@ export const MODELS = {
7828
7845
  contextWindow: 1000000,
7829
7846
  maxTokens: 131072,
7830
7847
  },
7848
+ "grok-4.5": {
7849
+ id: "grok-4.5",
7850
+ name: "Grok 4.5",
7851
+ api: "openai-completions",
7852
+ provider: "opencode-go",
7853
+ baseUrl: "https://opencode.ai/zen/go/v1",
7854
+ reasoning: true,
7855
+ input: ["text", "image"],
7856
+ cost: {
7857
+ input: 2,
7858
+ output: 6,
7859
+ cacheRead: 0.5,
7860
+ cacheWrite: 0,
7861
+ },
7862
+ contextWindow: 500000,
7863
+ maxTokens: 500000,
7864
+ },
7831
7865
  "kimi-k2.6": {
7832
7866
  id: "kimi-k2.6",
7833
7867
  name: "Kimi K2.6",
@@ -7862,6 +7896,23 @@ export const MODELS = {
7862
7896
  contextWindow: 262144,
7863
7897
  maxTokens: 262144,
7864
7898
  },
7899
+ "kimi-k3": {
7900
+ id: "kimi-k3",
7901
+ name: "Kimi K3",
7902
+ api: "openai-completions",
7903
+ provider: "opencode-go",
7904
+ baseUrl: "https://opencode.ai/zen/go/v1",
7905
+ reasoning: true,
7906
+ input: ["text", "image"],
7907
+ cost: {
7908
+ input: 3,
7909
+ output: 15,
7910
+ cacheRead: 0.3,
7911
+ cacheWrite: 0,
7912
+ },
7913
+ contextWindow: 1048576,
7914
+ maxTokens: 131072,
7915
+ },
7865
7916
  "mimo-v2.5": {
7866
7917
  id: "mimo-v2.5",
7867
7918
  name: "MiMo V2.5",
@@ -7888,9 +7939,9 @@ export const MODELS = {
7888
7939
  reasoning: true,
7889
7940
  input: ["text"],
7890
7941
  cost: {
7891
- input: 1.74,
7892
- output: 3.48,
7893
- cacheRead: 0.0145,
7942
+ input: 0.435,
7943
+ output: 0.87,
7944
+ cacheRead: 0.003625,
7894
7945
  cacheWrite: 0,
7895
7946
  },
7896
7947
  contextWindow: 1048576,
@@ -8587,13 +8638,13 @@ export const MODELS = {
8587
8638
  reasoning: false,
8588
8639
  input: ["text"],
8589
8640
  cost: {
8590
- input: 0.24,
8591
- output: 0.8999999999999999,
8641
+ input: 0.27,
8642
+ output: 1.12,
8592
8643
  cacheRead: 0.135,
8593
8644
  cacheWrite: 0,
8594
8645
  },
8595
8646
  contextWindow: 163840,
8596
- maxTokens: 16384,
8647
+ maxTokens: 65536,
8597
8648
  },
8598
8649
  "deepseek/deepseek-chat-v3.1": {
8599
8650
  id: "deepseek/deepseek-chat-v3.1",
@@ -8656,11 +8707,11 @@ export const MODELS = {
8656
8707
  input: ["text"],
8657
8708
  cost: {
8658
8709
  input: 0.27,
8659
- output: 0.95,
8660
- cacheRead: 0.13,
8710
+ output: 1,
8711
+ cacheRead: 0.135,
8661
8712
  cacheWrite: 0,
8662
8713
  },
8663
- contextWindow: 163840,
8714
+ contextWindow: 131072,
8664
8715
  maxTokens: 32768,
8665
8716
  },
8666
8717
  "deepseek/deepseek-v3.2": {
@@ -8672,13 +8723,13 @@ export const MODELS = {
8672
8723
  reasoning: true,
8673
8724
  input: ["text"],
8674
8725
  cost: {
8675
- input: 0.2145,
8676
- output: 0.32175,
8677
- cacheRead: 0.02145,
8726
+ input: 0.26899999999999996,
8727
+ output: 0.39999999999999997,
8728
+ cacheRead: 0.13449999999999998,
8678
8729
  cacheWrite: 0,
8679
8730
  },
8680
- contextWindow: 131072,
8681
- maxTokens: 64000,
8731
+ contextWindow: 163840,
8732
+ maxTokens: 65536,
8682
8733
  },
8683
8734
  "deepseek/deepseek-v3.2-exp": {
8684
8735
  id: "deepseek/deepseek-v3.2-exp",
@@ -8706,13 +8757,13 @@ export const MODELS = {
8706
8757
  reasoning: true,
8707
8758
  input: ["text"],
8708
8759
  cost: {
8709
- input: 0.09,
8710
- output: 0.18,
8711
- cacheRead: 0.018,
8760
+ input: 0.098,
8761
+ output: 0.196,
8762
+ cacheRead: 0.0196,
8712
8763
  cacheWrite: 0,
8713
8764
  },
8714
8765
  contextWindow: 1048576,
8715
- maxTokens: 65536,
8766
+ maxTokens: 4096,
8716
8767
  },
8717
8768
  "deepseek/deepseek-v4-pro": {
8718
8769
  id: "deepseek/deepseek-v4-pro",
@@ -8961,13 +9012,13 @@ export const MODELS = {
8961
9012
  reasoning: false,
8962
9013
  input: ["text", "image"],
8963
9014
  cost: {
8964
- input: 0.08,
8965
- output: 0.16,
9015
+ input: 0.09999999999999999,
9016
+ output: 0.3,
8966
9017
  cacheRead: 0,
8967
9018
  cacheWrite: 0,
8968
9019
  },
8969
9020
  contextWindow: 131072,
8970
- maxTokens: 16384,
9021
+ maxTokens: 4096,
8971
9022
  },
8972
9023
  "google/gemma-4-26b-a4b-it": {
8973
9024
  id: "google/gemma-4-26b-a4b-it",
@@ -8978,13 +9029,13 @@ export const MODELS = {
8978
9029
  reasoning: true,
8979
9030
  input: ["text", "image"],
8980
9031
  cost: {
8981
- input: 0.06,
8982
- output: 0.33,
9032
+ input: 0.09999999999999999,
9033
+ output: 0.3,
8983
9034
  cacheRead: 0,
8984
9035
  cacheWrite: 0,
8985
9036
  },
8986
9037
  contextWindow: 262144,
8987
- maxTokens: 4096,
9038
+ maxTokens: 256000,
8988
9039
  },
8989
9040
  "google/gemma-4-26b-a4b-it:free": {
8990
9041
  id: "google/gemma-4-26b-a4b-it:free",
@@ -9012,13 +9063,13 @@ export const MODELS = {
9012
9063
  reasoning: true,
9013
9064
  input: ["text", "image"],
9014
9065
  cost: {
9015
- input: 0.06,
9016
- output: 0.35,
9017
- cacheRead: 0,
9066
+ input: 0.22,
9067
+ output: 0.55,
9068
+ cacheRead: 0.12,
9018
9069
  cacheWrite: 0,
9019
9070
  },
9020
9071
  contextWindow: 262144,
9021
- maxTokens: 8192,
9072
+ maxTokens: 262144,
9022
9073
  },
9023
9074
  "google/gemma-4-31b-it:free": {
9024
9075
  id: "google/gemma-4-31b-it:free",
@@ -9199,13 +9250,13 @@ export const MODELS = {
9199
9250
  reasoning: false,
9200
9251
  input: ["text"],
9201
9252
  cost: {
9202
- input: 0.02,
9203
- output: 0.03,
9204
- cacheRead: 0,
9253
+ input: 0.049999999999999996,
9254
+ output: 0.08,
9255
+ cacheRead: 0.024999999999999998,
9205
9256
  cacheWrite: 0,
9206
9257
  },
9207
9258
  contextWindow: 131072,
9208
- maxTokens: 16384,
9259
+ maxTokens: 131072,
9209
9260
  },
9210
9261
  "meta-llama/llama-3.3-70b-instruct": {
9211
9262
  id: "meta-llama/llama-3.3-70b-instruct",
@@ -9216,13 +9267,13 @@ export const MODELS = {
9216
9267
  reasoning: false,
9217
9268
  input: ["text"],
9218
9269
  cost: {
9219
- input: 0.09999999999999999,
9220
- output: 0.32,
9270
+ input: 0.13,
9271
+ output: 0.39999999999999997,
9221
9272
  cacheRead: 0,
9222
9273
  cacheWrite: 0,
9223
9274
  },
9224
9275
  contextWindow: 131072,
9225
- maxTokens: 16384,
9276
+ maxTokens: 128000,
9226
9277
  },
9227
9278
  "meta-llama/llama-3.3-70b-instruct:free": {
9228
9279
  id: "meta-llama/llama-3.3-70b-instruct:free",
@@ -9275,6 +9326,23 @@ export const MODELS = {
9275
9326
  contextWindow: 10000000,
9276
9327
  maxTokens: 16384,
9277
9328
  },
9329
+ "meta/muse-spark-1.1": {
9330
+ id: "meta/muse-spark-1.1",
9331
+ name: "Meta: Muse Spark 1.1",
9332
+ api: "openai-completions",
9333
+ provider: "openrouter",
9334
+ baseUrl: "https://openrouter.ai/api/v1",
9335
+ reasoning: true,
9336
+ input: ["text", "image"],
9337
+ cost: {
9338
+ input: 1.25,
9339
+ output: 4.25,
9340
+ cacheRead: 0.15,
9341
+ cacheWrite: 0,
9342
+ },
9343
+ contextWindow: 1048576,
9344
+ maxTokens: 4096,
9345
+ },
9278
9346
  "minimax/minimax-m1": {
9279
9347
  id: "minimax/minimax-m1",
9280
9348
  name: "MiniMax: MiniMax M1",
@@ -9284,7 +9352,7 @@ export const MODELS = {
9284
9352
  reasoning: true,
9285
9353
  input: ["text"],
9286
9354
  cost: {
9287
- input: 0.39999999999999997,
9355
+ input: 0.55,
9288
9356
  output: 2.2,
9289
9357
  cacheRead: 0,
9290
9358
  cacheWrite: 0,
@@ -9352,13 +9420,13 @@ export const MODELS = {
9352
9420
  reasoning: true,
9353
9421
  input: ["text"],
9354
9422
  cost: {
9355
- input: 0.24,
9356
- output: 0.96,
9357
- cacheRead: 0,
9423
+ input: 0.25,
9424
+ output: 1,
9425
+ cacheRead: 0.049999999999999996,
9358
9426
  cacheWrite: 0,
9359
9427
  },
9360
9428
  contextWindow: 204800,
9361
- maxTokens: 196608,
9429
+ maxTokens: 131072,
9362
9430
  },
9363
9431
  "minimax/minimax-m3": {
9364
9432
  id: "minimax/minimax-m3",
@@ -9375,7 +9443,7 @@ export const MODELS = {
9375
9443
  cacheWrite: 0,
9376
9444
  },
9377
9445
  contextWindow: 1048576,
9378
- maxTokens: 131072,
9446
+ maxTokens: 512000,
9379
9447
  },
9380
9448
  "mistralai/codestral-2508": {
9381
9449
  id: "mistralai/codestral-2508",
@@ -9573,13 +9641,13 @@ export const MODELS = {
9573
9641
  reasoning: false,
9574
9642
  input: ["text"],
9575
9643
  cost: {
9576
- input: 0.02,
9644
+ input: 0.019000000000000003,
9577
9645
  output: 0.03,
9578
9646
  cacheRead: 0,
9579
9647
  cacheWrite: 0,
9580
9648
  },
9581
9649
  contextWindow: 131072,
9582
- maxTokens: 4096,
9650
+ maxTokens: 16384,
9583
9651
  },
9584
9652
  "mistralai/mistral-saba": {
9585
9653
  id: "mistralai/mistral-saba",
@@ -9624,13 +9692,13 @@ export const MODELS = {
9624
9692
  reasoning: false,
9625
9693
  input: ["text", "image"],
9626
9694
  cost: {
9627
- input: 0.075,
9628
- output: 0.19999999999999998,
9629
- cacheRead: 0,
9695
+ input: 0.09999999999999999,
9696
+ output: 0.3,
9697
+ cacheRead: 0.01,
9630
9698
  cacheWrite: 0,
9631
9699
  },
9632
- contextWindow: 128000,
9633
- maxTokens: 16384,
9700
+ contextWindow: 131072,
9701
+ maxTokens: 4096,
9634
9702
  },
9635
9703
  "mistralai/mixtral-8x22b-instruct": {
9636
9704
  id: "mistralai/mixtral-8x22b-instruct",
@@ -9711,11 +9779,11 @@ export const MODELS = {
9711
9779
  cost: {
9712
9780
  input: 0.6,
9713
9781
  output: 2.5,
9714
- cacheRead: 0.15,
9782
+ cacheRead: 0,
9715
9783
  cacheWrite: 0,
9716
9784
  },
9717
9785
  contextWindow: 262144,
9718
- maxTokens: 100352,
9786
+ maxTokens: 262144,
9719
9787
  },
9720
9788
  "moonshotai/kimi-k2.5": {
9721
9789
  id: "moonshotai/kimi-k2.5",
@@ -9743,13 +9811,13 @@ export const MODELS = {
9743
9811
  reasoning: true,
9744
9812
  input: ["text", "image"],
9745
9813
  cost: {
9746
- input: 0.66,
9747
- output: 3.41,
9748
- cacheRead: 0.15,
9814
+ input: 0.95,
9815
+ output: 4,
9816
+ cacheRead: 0.16,
9749
9817
  cacheWrite: 0,
9750
9818
  },
9751
9819
  contextWindow: 262144,
9752
- maxTokens: 262144,
9820
+ maxTokens: 4096,
9753
9821
  },
9754
9822
  "moonshotai/kimi-k2.7-code": {
9755
9823
  id: "moonshotai/kimi-k2.7-code",
@@ -9760,14 +9828,31 @@ export const MODELS = {
9760
9828
  reasoning: true,
9761
9829
  input: ["text", "image"],
9762
9830
  cost: {
9763
- input: 0.719,
9764
- output: 3.49,
9765
- cacheRead: 0.149,
9831
+ input: 0.75,
9832
+ output: 3.5,
9833
+ cacheRead: 0.16,
9766
9834
  cacheWrite: 0,
9767
9835
  },
9768
9836
  contextWindow: 262144,
9769
9837
  maxTokens: 262144,
9770
9838
  },
9839
+ "moonshotai/kimi-k3": {
9840
+ id: "moonshotai/kimi-k3",
9841
+ name: "MoonshotAI: Kimi K3",
9842
+ api: "openai-completions",
9843
+ provider: "openrouter",
9844
+ baseUrl: "https://openrouter.ai/api/v1",
9845
+ reasoning: true,
9846
+ input: ["text", "image"],
9847
+ cost: {
9848
+ input: 3,
9849
+ output: 15,
9850
+ cacheRead: 0.3,
9851
+ cacheWrite: 0,
9852
+ },
9853
+ contextWindow: 1048576,
9854
+ maxTokens: 4096,
9855
+ },
9771
9856
  "nex-agi/nex-n2-mini": {
9772
9857
  id: "nex-agi/nex-n2-mini",
9773
9858
  name: "Nex AGI: Nex-N2-Mini",
@@ -9802,23 +9887,6 @@ export const MODELS = {
9802
9887
  contextWindow: 262144,
9803
9888
  maxTokens: 262144,
9804
9889
  },
9805
- "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
9806
- id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
9807
- name: "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5",
9808
- api: "openai-completions",
9809
- provider: "openrouter",
9810
- baseUrl: "https://openrouter.ai/api/v1",
9811
- reasoning: true,
9812
- input: ["text"],
9813
- cost: {
9814
- input: 0.39999999999999997,
9815
- output: 0.39999999999999997,
9816
- cacheRead: 0,
9817
- cacheWrite: 0,
9818
- },
9819
- contextWindow: 131072,
9820
- maxTokens: 16384,
9821
- },
9822
9890
  "nvidia/nemotron-3-nano-30b-a3b": {
9823
9891
  id: "nvidia/nemotron-3-nano-30b-a3b",
9824
9892
  name: "NVIDIA: Nemotron 3 Nano 30B A3B",
@@ -9879,9 +9947,9 @@ export const MODELS = {
9879
9947
  reasoning: true,
9880
9948
  input: ["text"],
9881
9949
  cost: {
9882
- input: 0.08,
9883
- output: 0.44999999999999996,
9884
- cacheRead: 0,
9950
+ input: 0.21,
9951
+ output: 0.45499999999999996,
9952
+ cacheRead: 0.06,
9885
9953
  cacheWrite: 0,
9886
9954
  },
9887
9955
  contextWindow: 1000000,
@@ -9913,13 +9981,13 @@ export const MODELS = {
9913
9981
  reasoning: true,
9914
9982
  input: ["text"],
9915
9983
  cost: {
9916
- input: 0.5,
9917
- output: 2.2,
9918
- cacheRead: 0.09999999999999999,
9984
+ input: 0,
9985
+ output: 0,
9986
+ cacheRead: 0,
9919
9987
  cacheWrite: 0,
9920
9988
  },
9921
9989
  contextWindow: 1000000,
9922
- maxTokens: 16384,
9990
+ maxTokens: 202800,
9923
9991
  },
9924
9992
  "nvidia/nemotron-3-ultra-550b-a55b:free": {
9925
9993
  id: "nvidia/nemotron-3-ultra-550b-a55b:free",
@@ -10089,7 +10157,7 @@ export const MODELS = {
10089
10157
  cacheWrite: 0,
10090
10158
  },
10091
10159
  contextWindow: 1047576,
10092
- maxTokens: 4096,
10160
+ maxTokens: 32768,
10093
10161
  },
10094
10162
  "openai/gpt-4.1-mini": {
10095
10163
  id: "openai/gpt-4.1-mini",
@@ -10136,7 +10204,7 @@ export const MODELS = {
10136
10204
  cost: {
10137
10205
  input: 2.5,
10138
10206
  output: 10,
10139
- cacheRead: 0,
10207
+ cacheRead: 1.25,
10140
10208
  cacheWrite: 0,
10141
10209
  },
10142
10210
  contextWindow: 128000,
@@ -10340,11 +10408,11 @@ export const MODELS = {
10340
10408
  cost: {
10341
10409
  input: 1.25,
10342
10410
  output: 10,
10343
- cacheRead: 0.13,
10411
+ cacheRead: 0.125,
10344
10412
  cacheWrite: 0,
10345
10413
  },
10346
10414
  contextWindow: 128000,
10347
- maxTokens: 32000,
10415
+ maxTokens: 16384,
10348
10416
  },
10349
10417
  "openai/gpt-5.1-codex": {
10350
10418
  id: "openai/gpt-5.1-codex",
@@ -10357,7 +10425,7 @@ export const MODELS = {
10357
10425
  cost: {
10358
10426
  input: 1.25,
10359
10427
  output: 10,
10360
- cacheRead: 0.13,
10428
+ cacheRead: 0.125,
10361
10429
  cacheWrite: 0,
10362
10430
  },
10363
10431
  contextWindow: 400000,
@@ -10763,8 +10831,8 @@ export const MODELS = {
10763
10831
  reasoning: true,
10764
10832
  input: ["text"],
10765
10833
  cost: {
10766
- input: 0.03,
10767
- output: 0.15,
10834
+ input: 0.037,
10835
+ output: 0.16999999999999998,
10768
10836
  cacheRead: 0,
10769
10837
  cacheWrite: 0,
10770
10838
  },
@@ -10780,13 +10848,13 @@ export const MODELS = {
10780
10848
  reasoning: true,
10781
10849
  input: ["text"],
10782
10850
  cost: {
10783
- input: 0.029,
10784
- output: 0.14,
10785
- cacheRead: 0,
10851
+ input: 0.03,
10852
+ output: 0.13,
10853
+ cacheRead: 0.03,
10786
10854
  cacheWrite: 0,
10787
10855
  },
10788
10856
  contextWindow: 131072,
10789
- maxTokens: 4096,
10857
+ maxTokens: 131072,
10790
10858
  },
10791
10859
  "openai/gpt-oss-20b:free": {
10792
10860
  id: "openai/gpt-oss-20b:free",
@@ -11171,13 +11239,13 @@ export const MODELS = {
11171
11239
  reasoning: true,
11172
11240
  input: ["text"],
11173
11241
  cost: {
11174
- input: 0.09999999999999999,
11242
+ input: 0.12,
11175
11243
  output: 0.24,
11176
11244
  cacheRead: 0,
11177
11245
  cacheWrite: 0,
11178
11246
  },
11179
11247
  contextWindow: 131702,
11180
- maxTokens: 40960,
11248
+ maxTokens: 16384,
11181
11249
  },
11182
11250
  "qwen/qwen3-235b-a22b": {
11183
11251
  id: "qwen/qwen3-235b-a22b",
@@ -11256,13 +11324,13 @@ export const MODELS = {
11256
11324
  reasoning: false,
11257
11325
  input: ["text"],
11258
11326
  cost: {
11259
- input: 0.04815,
11260
- output: 0.19305,
11327
+ input: 0.09999999999999999,
11328
+ output: 0.3,
11261
11329
  cacheRead: 0,
11262
11330
  cacheWrite: 0,
11263
11331
  },
11264
- contextWindow: 131072,
11265
- maxTokens: 32000,
11332
+ contextWindow: 262144,
11333
+ maxTokens: 4096,
11266
11334
  },
11267
11335
  "qwen/qwen3-30b-a3b-thinking-2507": {
11268
11336
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -11324,9 +11392,9 @@ export const MODELS = {
11324
11392
  reasoning: false,
11325
11393
  input: ["text"],
11326
11394
  cost: {
11327
- input: 0.22,
11328
- output: 1.7999999999999998,
11329
- cacheRead: 0,
11395
+ input: 0.3,
11396
+ output: 1,
11397
+ cacheRead: 0.09999999999999999,
11330
11398
  cacheWrite: 0,
11331
11399
  },
11332
11400
  contextWindow: 1048576,
@@ -11460,13 +11528,13 @@ export const MODELS = {
11460
11528
  reasoning: false,
11461
11529
  input: ["text"],
11462
11530
  cost: {
11463
- input: 0.09,
11531
+ input: 0.09999999999999999,
11464
11532
  output: 1.1,
11465
- cacheRead: 0,
11533
+ cacheRead: 0.07,
11466
11534
  cacheWrite: 0,
11467
11535
  },
11468
11536
  contextWindow: 262144,
11469
- maxTokens: 16384,
11537
+ maxTokens: 262144,
11470
11538
  },
11471
11539
  "qwen/qwen3-next-80b-a3b-instruct:free": {
11472
11540
  id: "qwen/qwen3-next-80b-a3b-instruct:free",
@@ -11511,13 +11579,13 @@ export const MODELS = {
11511
11579
  reasoning: false,
11512
11580
  input: ["text", "image"],
11513
11581
  cost: {
11514
- input: 0.19999999999999998,
11515
- output: 0.88,
11516
- cacheRead: 0.11,
11582
+ input: 0.21,
11583
+ output: 1.9,
11584
+ cacheRead: 0.09999999999999999,
11517
11585
  cacheWrite: 0,
11518
11586
  },
11519
- contextWindow: 262144,
11520
- maxTokens: 16384,
11587
+ contextWindow: 131072,
11588
+ maxTokens: 32768,
11521
11589
  },
11522
11590
  "qwen/qwen3-vl-235b-a22b-thinking": {
11523
11591
  id: "qwen/qwen3-vl-235b-a22b-thinking",
@@ -11636,7 +11704,7 @@ export const MODELS = {
11636
11704
  cacheWrite: 0,
11637
11705
  },
11638
11706
  contextWindow: 262144,
11639
- maxTokens: 262144,
11707
+ maxTokens: 65536,
11640
11708
  },
11641
11709
  "qwen/qwen3.5-27b": {
11642
11710
  id: "qwen/qwen3.5-27b",
@@ -11666,11 +11734,11 @@ export const MODELS = {
11666
11734
  cost: {
11667
11735
  input: 0.14,
11668
11736
  output: 1,
11669
- cacheRead: 0.049999999999999996,
11737
+ cacheRead: 0,
11670
11738
  cacheWrite: 0,
11671
11739
  },
11672
11740
  contextWindow: 262144,
11673
- maxTokens: 81920,
11741
+ maxTokens: 262144,
11674
11742
  },
11675
11743
  "qwen/qwen3.5-397b-a17b": {
11676
11744
  id: "qwen/qwen3.5-397b-a17b",
@@ -11681,13 +11749,13 @@ export const MODELS = {
11681
11749
  reasoning: true,
11682
11750
  input: ["text", "image"],
11683
11751
  cost: {
11684
- input: 0.385,
11685
- output: 2.4499999999999997,
11686
- cacheRead: 0.111,
11752
+ input: 0.39,
11753
+ output: 2.34,
11754
+ cacheRead: 0,
11687
11755
  cacheWrite: 0,
11688
11756
  },
11689
- contextWindow: 256000,
11690
- maxTokens: 4096,
11757
+ contextWindow: 262144,
11758
+ maxTokens: 65536,
11691
11759
  },
11692
11760
  "qwen/qwen3.5-9b": {
11693
11761
  id: "qwen/qwen3.5-9b",
@@ -11766,13 +11834,13 @@ export const MODELS = {
11766
11834
  reasoning: true,
11767
11835
  input: ["text", "image"],
11768
11836
  cost: {
11769
- input: 0.28900000000000003,
11770
- output: 2.4,
11837
+ input: 0.44999999999999996,
11838
+ output: 2.7,
11771
11839
  cacheRead: 0,
11772
11840
  cacheWrite: 0,
11773
11841
  },
11774
11842
  contextWindow: 262144,
11775
- maxTokens: 131072,
11843
+ maxTokens: 65536,
11776
11844
  },
11777
11845
  "qwen/qwen3.6-35b-a3b": {
11778
11846
  id: "qwen/qwen3.6-35b-a3b",
@@ -11851,10 +11919,10 @@ export const MODELS = {
11851
11919
  reasoning: true,
11852
11920
  input: ["text"],
11853
11921
  cost: {
11854
- input: 1.25,
11855
- output: 3.75,
11856
- cacheRead: 0.25,
11857
- cacheWrite: 1.5625,
11922
+ input: 1.475,
11923
+ output: 4.425,
11924
+ cacheRead: 0.295,
11925
+ cacheWrite: 1.84375,
11858
11926
  },
11859
11927
  contextWindow: 1000000,
11860
11928
  maxTokens: 65536,
@@ -11987,13 +12055,13 @@ export const MODELS = {
11987
12055
  reasoning: true,
11988
12056
  input: ["text"],
11989
12057
  cost: {
11990
- input: 0.14,
11991
- output: 0.58,
11992
- cacheRead: 0.035,
12058
+ input: 0.19999999999999998,
12059
+ output: 0.7999999999999999,
12060
+ cacheRead: 0.049999999999999996,
11993
12061
  cacheWrite: 0,
11994
12062
  },
11995
12063
  contextWindow: 262144,
11996
- maxTokens: 4096,
12064
+ maxTokens: 131072,
11997
12065
  },
11998
12066
  "tencent/hy3-preview": {
11999
12067
  id: "tencent/hy3-preview",
@@ -12140,13 +12208,13 @@ export const MODELS = {
12140
12208
  reasoning: true,
12141
12209
  input: ["text", "image"],
12142
12210
  cost: {
12143
- input: 0.105,
12211
+ input: 0.14,
12144
12212
  output: 0.28,
12145
- cacheRead: 0.028,
12213
+ cacheRead: 0.0028,
12146
12214
  cacheWrite: 0,
12147
12215
  },
12148
12216
  contextWindow: 1048576,
12149
- maxTokens: 4096,
12217
+ maxTokens: 131072,
12150
12218
  },
12151
12219
  "xiaomi/mimo-v2.5-pro": {
12152
12220
  id: "xiaomi/mimo-v2.5-pro",
@@ -12225,13 +12293,13 @@ export const MODELS = {
12225
12293
  reasoning: true,
12226
12294
  input: ["text"],
12227
12295
  cost: {
12228
- input: 0.43,
12229
- output: 1.75,
12230
- cacheRead: 0.08,
12296
+ input: 0.5,
12297
+ output: 2,
12298
+ cacheRead: 0.09999999999999999,
12231
12299
  cacheWrite: 0,
12232
12300
  },
12233
- contextWindow: 200000,
12234
- maxTokens: 16384,
12301
+ contextWindow: 202752,
12302
+ maxTokens: 131072,
12235
12303
  },
12236
12304
  "z-ai/glm-4.6v": {
12237
12305
  id: "z-ai/glm-4.6v",
@@ -12276,13 +12344,13 @@ export const MODELS = {
12276
12344
  reasoning: true,
12277
12345
  input: ["text"],
12278
12346
  cost: {
12279
- input: 0.06,
12347
+ input: 0.060500000000000005,
12280
12348
  output: 0.39999999999999997,
12281
- cacheRead: 0.01,
12349
+ cacheRead: 0,
12282
12350
  cacheWrite: 0,
12283
12351
  },
12284
- contextWindow: 202752,
12285
- maxTokens: 16384,
12352
+ contextWindow: 200000,
12353
+ maxTokens: 131072,
12286
12354
  },
12287
12355
  "z-ai/glm-5": {
12288
12356
  id: "z-ai/glm-5",
@@ -12293,13 +12361,13 @@ export const MODELS = {
12293
12361
  reasoning: true,
12294
12362
  input: ["text"],
12295
12363
  cost: {
12296
- input: 0.6,
12297
- output: 1.92,
12298
- cacheRead: 0.12,
12364
+ input: 0.95,
12365
+ output: 3.15,
12366
+ cacheRead: 0.19,
12299
12367
  cacheWrite: 0,
12300
12368
  },
12301
12369
  contextWindow: 202752,
12302
- maxTokens: 128000,
12370
+ maxTokens: 202752,
12303
12371
  },
12304
12372
  "z-ai/glm-5-turbo": {
12305
12373
  id: "z-ai/glm-5-turbo",
@@ -12315,7 +12383,7 @@ export const MODELS = {
12315
12383
  cacheRead: 0.24,
12316
12384
  cacheWrite: 0,
12317
12385
  },
12318
- contextWindow: 262144,
12386
+ contextWindow: 202752,
12319
12387
  maxTokens: 131072,
12320
12388
  },
12321
12389
  "z-ai/glm-5.1": {
@@ -12344,13 +12412,13 @@ export const MODELS = {
12344
12412
  reasoning: true,
12345
12413
  input: ["text"],
12346
12414
  cost: {
12347
- input: 0.9099999999999999,
12348
- output: 2.8600000000000003,
12349
- cacheRead: 0.16899999999999998,
12415
+ input: 0.9254,
12416
+ output: 2.9084,
12417
+ cacheRead: 0.17186,
12350
12418
  cacheWrite: 0,
12351
12419
  },
12352
12420
  contextWindow: 1048576,
12353
- maxTokens: 128000,
12421
+ maxTokens: 131072,
12354
12422
  },
12355
12423
  "z-ai/glm-5v-turbo": {
12356
12424
  id: "z-ai/glm-5v-turbo",
@@ -12480,13 +12548,13 @@ export const MODELS = {
12480
12548
  reasoning: true,
12481
12549
  input: ["text", "image"],
12482
12550
  cost: {
12483
- input: 0.66,
12484
- output: 3.41,
12485
- cacheRead: 0.15,
12551
+ input: 3,
12552
+ output: 15,
12553
+ cacheRead: 0.3,
12486
12554
  cacheWrite: 0,
12487
12555
  },
12488
- contextWindow: 262144,
12489
- maxTokens: 262144,
12556
+ contextWindow: 1048576,
12557
+ maxTokens: 4096,
12490
12558
  },
12491
12559
  "~openai/gpt-latest": {
12492
12560
  id: "~openai/gpt-latest",
@@ -13153,6 +13221,23 @@ export const MODELS = {
13153
13221
  contextWindow: 1000000,
13154
13222
  maxTokens: 128000,
13155
13223
  },
13224
+ "anthropic/claude-opus-4.7-fast": {
13225
+ id: "anthropic/claude-opus-4.7-fast",
13226
+ name: "Claude Opus 4.7 (Fast)",
13227
+ api: "anthropic-messages",
13228
+ provider: "vercel-ai-gateway",
13229
+ baseUrl: "https://ai-gateway.vercel.sh",
13230
+ reasoning: true,
13231
+ input: ["text", "image"],
13232
+ cost: {
13233
+ input: 30,
13234
+ output: 150,
13235
+ cacheRead: 3,
13236
+ cacheWrite: 37.5,
13237
+ },
13238
+ contextWindow: 1000000,
13239
+ maxTokens: 128000,
13240
+ },
13156
13241
  "anthropic/claude-opus-4.8": {
13157
13242
  id: "anthropic/claude-opus-4.8",
13158
13243
  name: "Claude Opus 4.8",
@@ -13170,6 +13255,23 @@ export const MODELS = {
13170
13255
  contextWindow: 1000000,
13171
13256
  maxTokens: 128000,
13172
13257
  },
13258
+ "anthropic/claude-opus-4.8-fast": {
13259
+ id: "anthropic/claude-opus-4.8-fast",
13260
+ name: "Claude Opus 4.8 (Fast)",
13261
+ api: "anthropic-messages",
13262
+ provider: "vercel-ai-gateway",
13263
+ baseUrl: "https://ai-gateway.vercel.sh",
13264
+ reasoning: true,
13265
+ input: ["text", "image"],
13266
+ cost: {
13267
+ input: 10,
13268
+ output: 50,
13269
+ cacheRead: 1,
13270
+ cacheWrite: 12.5,
13271
+ },
13272
+ contextWindow: 1000000,
13273
+ maxTokens: 128000,
13274
+ },
13173
13275
  "anthropic/claude-sonnet-4": {
13174
13276
  id: "anthropic/claude-sonnet-4",
13175
13277
  name: "Claude Sonnet 4",
@@ -13799,40 +13901,6 @@ export const MODELS = {
13799
13901
  contextWindow: 128000,
13800
13902
  maxTokens: 8192,
13801
13903
  },
13802
- "meta/llama-3.2-11b": {
13803
- id: "meta/llama-3.2-11b",
13804
- name: "Llama 3.2 11B Vision Instruct",
13805
- api: "anthropic-messages",
13806
- provider: "vercel-ai-gateway",
13807
- baseUrl: "https://ai-gateway.vercel.sh",
13808
- reasoning: false,
13809
- input: ["text", "image"],
13810
- cost: {
13811
- input: 0.16,
13812
- output: 0.16,
13813
- cacheRead: 0,
13814
- cacheWrite: 0,
13815
- },
13816
- contextWindow: 128000,
13817
- maxTokens: 8192,
13818
- },
13819
- "meta/llama-3.2-90b": {
13820
- id: "meta/llama-3.2-90b",
13821
- name: "Llama 3.2 90B Vision Instruct",
13822
- api: "anthropic-messages",
13823
- provider: "vercel-ai-gateway",
13824
- baseUrl: "https://ai-gateway.vercel.sh",
13825
- reasoning: false,
13826
- input: ["text", "image"],
13827
- cost: {
13828
- input: 0.72,
13829
- output: 0.72,
13830
- cacheRead: 0,
13831
- cacheWrite: 0,
13832
- },
13833
- contextWindow: 128000,
13834
- maxTokens: 8192,
13835
- },
13836
13904
  "meta/llama-3.3-70b": {
13837
13905
  id: "meta/llama-3.3-70b",
13838
13906
  name: "Llama 3.3 70B Instruct",
@@ -14377,6 +14445,23 @@ export const MODELS = {
14377
14445
  contextWindow: 262144,
14378
14446
  maxTokens: 32768,
14379
14447
  },
14448
+ "moonshotai/kimi-k3": {
14449
+ id: "moonshotai/kimi-k3",
14450
+ name: "Kimi K3",
14451
+ api: "anthropic-messages",
14452
+ provider: "vercel-ai-gateway",
14453
+ baseUrl: "https://ai-gateway.vercel.sh",
14454
+ reasoning: true,
14455
+ input: ["text", "image"],
14456
+ cost: {
14457
+ input: 3,
14458
+ output: 15,
14459
+ cacheRead: 0.3,
14460
+ cacheWrite: 0,
14461
+ },
14462
+ contextWindow: 1000000,
14463
+ maxTokens: 131072,
14464
+ },
14380
14465
  "nvidia/nemotron-3-nano-30b-a3b": {
14381
14466
  id: "nvidia/nemotron-3-nano-30b-a3b",
14382
14467
  name: "Nemotron 3 Nano 30B A3B",
@@ -15227,6 +15312,23 @@ export const MODELS = {
15227
15312
  contextWindow: 256000,
15228
15313
  maxTokens: 256000,
15229
15314
  },
15315
+ "thinkingmachines/inkling": {
15316
+ id: "thinkingmachines/inkling",
15317
+ name: "Inkling",
15318
+ api: "anthropic-messages",
15319
+ provider: "vercel-ai-gateway",
15320
+ baseUrl: "https://ai-gateway.vercel.sh",
15321
+ reasoning: true,
15322
+ input: ["text", "image"],
15323
+ cost: {
15324
+ input: 1,
15325
+ output: 4.05,
15326
+ cacheRead: 0.16999999999999998,
15327
+ cacheWrite: 0,
15328
+ },
15329
+ contextWindow: 256000,
15330
+ maxTokens: 256000,
15331
+ },
15230
15332
  "xai/grok-4.1-fast-non-reasoning": {
15231
15333
  id: "xai/grok-4.1-fast-non-reasoning",
15232
15334
  name: "Grok 4.1 Fast Non-Reasoning",