@caupulican/pi-ai 0.80.34 → 0.80.38

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.
@@ -4077,6 +4077,46 @@ export const MODELS = {
4077
4077
  maxTokens: 262000,
4078
4078
  },
4079
4079
  },
4080
+ "fugu": {
4081
+ "fugu": {
4082
+ id: "fugu",
4083
+ name: "Fugu",
4084
+ api: "openai-responses",
4085
+ provider: "fugu",
4086
+ baseUrl: "https://api.sakana.ai/v1",
4087
+ compat: { "supportsLongCacheRetention": false },
4088
+ reasoning: true,
4089
+ thinkingLevelMap: { "off": null, "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
4090
+ input: ["text", "image"],
4091
+ cost: {
4092
+ input: 0,
4093
+ output: 0,
4094
+ cacheRead: 0,
4095
+ cacheWrite: 0,
4096
+ },
4097
+ contextWindow: 1000000,
4098
+ maxTokens: 10000,
4099
+ },
4100
+ "fugu-ultra": {
4101
+ id: "fugu-ultra",
4102
+ name: "Fugu Ultra",
4103
+ api: "openai-responses",
4104
+ provider: "fugu",
4105
+ baseUrl: "https://api.sakana.ai/v1",
4106
+ compat: { "supportsLongCacheRetention": false },
4107
+ reasoning: true,
4108
+ thinkingLevelMap: { "off": null, "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
4109
+ input: ["text", "image"],
4110
+ cost: {
4111
+ input: 5,
4112
+ output: 30,
4113
+ cacheRead: 0.5,
4114
+ cacheWrite: 0,
4115
+ },
4116
+ contextWindow: 1000000,
4117
+ maxTokens: 10000,
4118
+ },
4119
+ },
4080
4120
  "github-copilot": {
4081
4121
  "claude-fable-5": {
4082
4122
  id: "claude-fable-5",
@@ -5132,6 +5172,24 @@ export const MODELS = {
5132
5172
  },
5133
5173
  },
5134
5174
  "huggingface": {
5175
+ "MiniMaxAI/MiniMax-M2": {
5176
+ id: "MiniMaxAI/MiniMax-M2",
5177
+ name: "MiniMax-M2",
5178
+ api: "openai-completions",
5179
+ provider: "huggingface",
5180
+ baseUrl: "https://router.huggingface.co/v1",
5181
+ compat: { "supportsDeveloperRole": false },
5182
+ reasoning: true,
5183
+ input: ["text"],
5184
+ cost: {
5185
+ input: 0.3,
5186
+ output: 1.2,
5187
+ cacheRead: 0,
5188
+ cacheWrite: 0,
5189
+ },
5190
+ contextWindow: 204800,
5191
+ maxTokens: 128000,
5192
+ },
5135
5193
  "MiniMaxAI/MiniMax-M2.1": {
5136
5194
  id: "MiniMaxAI/MiniMax-M2.1",
5137
5195
  name: "MiniMax-M2.1",
@@ -5186,6 +5244,42 @@ export const MODELS = {
5186
5244
  contextWindow: 204800,
5187
5245
  maxTokens: 131072,
5188
5246
  },
5247
+ "MiniMaxAI/MiniMax-M3": {
5248
+ id: "MiniMaxAI/MiniMax-M3",
5249
+ name: "MiniMax-M3",
5250
+ api: "openai-completions",
5251
+ provider: "huggingface",
5252
+ baseUrl: "https://router.huggingface.co/v1",
5253
+ compat: { "supportsDeveloperRole": false },
5254
+ reasoning: true,
5255
+ input: ["text", "image"],
5256
+ cost: {
5257
+ input: 0.3,
5258
+ output: 1.2,
5259
+ cacheRead: 0,
5260
+ cacheWrite: 0,
5261
+ },
5262
+ contextWindow: 524288,
5263
+ maxTokens: 128000,
5264
+ },
5265
+ "Qwen/Qwen3-235B-A22B": {
5266
+ id: "Qwen/Qwen3-235B-A22B",
5267
+ name: "Qwen3 235B-A22B",
5268
+ api: "openai-completions",
5269
+ provider: "huggingface",
5270
+ baseUrl: "https://router.huggingface.co/v1",
5271
+ compat: { "supportsDeveloperRole": false },
5272
+ reasoning: true,
5273
+ input: ["text"],
5274
+ cost: {
5275
+ input: 0.2,
5276
+ output: 0.8,
5277
+ cacheRead: 0,
5278
+ cacheWrite: 0,
5279
+ },
5280
+ contextWindow: 40960,
5281
+ maxTokens: 16384,
5282
+ },
5189
5283
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
5190
5284
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
5191
5285
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -5204,6 +5298,42 @@ export const MODELS = {
5204
5298
  contextWindow: 262144,
5205
5299
  maxTokens: 131072,
5206
5300
  },
5301
+ "Qwen/Qwen3-32B": {
5302
+ id: "Qwen/Qwen3-32B",
5303
+ name: "Qwen3 32B",
5304
+ api: "openai-completions",
5305
+ provider: "huggingface",
5306
+ baseUrl: "https://router.huggingface.co/v1",
5307
+ compat: { "supportsDeveloperRole": false },
5308
+ reasoning: true,
5309
+ input: ["text"],
5310
+ cost: {
5311
+ input: 0.29,
5312
+ output: 0.59,
5313
+ cacheRead: 0,
5314
+ cacheWrite: 0,
5315
+ },
5316
+ contextWindow: 131072,
5317
+ maxTokens: 16384,
5318
+ },
5319
+ "Qwen/Qwen3-Coder-30B-A3B-Instruct": {
5320
+ id: "Qwen/Qwen3-Coder-30B-A3B-Instruct",
5321
+ name: "Qwen3-Coder 30B-A3B Instruct",
5322
+ api: "openai-completions",
5323
+ provider: "huggingface",
5324
+ baseUrl: "https://router.huggingface.co/v1",
5325
+ compat: { "supportsDeveloperRole": false },
5326
+ reasoning: false,
5327
+ input: ["text"],
5328
+ cost: {
5329
+ input: 0.07,
5330
+ output: 0.26,
5331
+ cacheRead: 0,
5332
+ cacheWrite: 0,
5333
+ },
5334
+ contextWindow: 262144,
5335
+ maxTokens: 65536,
5336
+ },
5207
5337
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
5208
5338
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
5209
5339
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -5276,6 +5406,60 @@ export const MODELS = {
5276
5406
  contextWindow: 262144,
5277
5407
  maxTokens: 131072,
5278
5408
  },
5409
+ "Qwen/Qwen3.5-122B-A10B": {
5410
+ id: "Qwen/Qwen3.5-122B-A10B",
5411
+ name: "Qwen3.5 122B-A10B",
5412
+ api: "openai-completions",
5413
+ provider: "huggingface",
5414
+ baseUrl: "https://router.huggingface.co/v1",
5415
+ compat: { "supportsDeveloperRole": false },
5416
+ reasoning: true,
5417
+ input: ["text", "image"],
5418
+ cost: {
5419
+ input: 0.4,
5420
+ output: 3.2,
5421
+ cacheRead: 0,
5422
+ cacheWrite: 0,
5423
+ },
5424
+ contextWindow: 262144,
5425
+ maxTokens: 65536,
5426
+ },
5427
+ "Qwen/Qwen3.5-27B": {
5428
+ id: "Qwen/Qwen3.5-27B",
5429
+ name: "Qwen3.5 27B",
5430
+ api: "openai-completions",
5431
+ provider: "huggingface",
5432
+ baseUrl: "https://router.huggingface.co/v1",
5433
+ compat: { "supportsDeveloperRole": false },
5434
+ reasoning: true,
5435
+ input: ["text", "image"],
5436
+ cost: {
5437
+ input: 0.3,
5438
+ output: 2.4,
5439
+ cacheRead: 0,
5440
+ cacheWrite: 0,
5441
+ },
5442
+ contextWindow: 262144,
5443
+ maxTokens: 65536,
5444
+ },
5445
+ "Qwen/Qwen3.5-35B-A3B": {
5446
+ id: "Qwen/Qwen3.5-35B-A3B",
5447
+ name: "Qwen3.5 35B-A3B",
5448
+ api: "openai-completions",
5449
+ provider: "huggingface",
5450
+ baseUrl: "https://router.huggingface.co/v1",
5451
+ compat: { "supportsDeveloperRole": false },
5452
+ reasoning: true,
5453
+ input: ["text", "image"],
5454
+ cost: {
5455
+ input: 0.25,
5456
+ output: 2,
5457
+ cacheRead: 0,
5458
+ cacheWrite: 0,
5459
+ },
5460
+ contextWindow: 262144,
5461
+ maxTokens: 65536,
5462
+ },
5279
5463
  "Qwen/Qwen3.5-397B-A17B": {
5280
5464
  id: "Qwen/Qwen3.5-397B-A17B",
5281
5465
  name: "Qwen3.5-397B-A17B",
@@ -5294,6 +5478,42 @@ export const MODELS = {
5294
5478
  contextWindow: 262144,
5295
5479
  maxTokens: 32768,
5296
5480
  },
5481
+ "Qwen/Qwen3.5-9B": {
5482
+ id: "Qwen/Qwen3.5-9B",
5483
+ name: "Qwen3.5 9B",
5484
+ api: "openai-completions",
5485
+ provider: "huggingface",
5486
+ baseUrl: "https://router.huggingface.co/v1",
5487
+ compat: { "supportsDeveloperRole": false },
5488
+ reasoning: true,
5489
+ input: ["text", "image"],
5490
+ cost: {
5491
+ input: 0.17,
5492
+ output: 0.25,
5493
+ cacheRead: 0,
5494
+ cacheWrite: 0,
5495
+ },
5496
+ contextWindow: 262144,
5497
+ maxTokens: 65536,
5498
+ },
5499
+ "Qwen/Qwen3.6-35B-A3B": {
5500
+ id: "Qwen/Qwen3.6-35B-A3B",
5501
+ name: "Qwen3.6 35B-A3B",
5502
+ api: "openai-completions",
5503
+ provider: "huggingface",
5504
+ baseUrl: "https://router.huggingface.co/v1",
5505
+ compat: { "supportsDeveloperRole": false },
5506
+ reasoning: true,
5507
+ input: ["text", "image"],
5508
+ cost: {
5509
+ input: 0.15,
5510
+ output: 0.95,
5511
+ cacheRead: 0,
5512
+ cacheWrite: 0,
5513
+ },
5514
+ contextWindow: 262144,
5515
+ maxTokens: 65536,
5516
+ },
5297
5517
  "XiaomiMiMo/MiMo-V2-Flash": {
5298
5518
  id: "XiaomiMiMo/MiMo-V2-Flash",
5299
5519
  name: "MiMo-V2-Flash",
@@ -5312,6 +5532,24 @@ export const MODELS = {
5312
5532
  contextWindow: 262144,
5313
5533
  maxTokens: 4096,
5314
5534
  },
5535
+ "deepseek-ai/DeepSeek-R1": {
5536
+ id: "deepseek-ai/DeepSeek-R1",
5537
+ name: "DeepSeek-R1",
5538
+ api: "openai-completions",
5539
+ provider: "huggingface",
5540
+ baseUrl: "https://router.huggingface.co/v1",
5541
+ compat: { "supportsDeveloperRole": false },
5542
+ reasoning: true,
5543
+ input: ["text"],
5544
+ cost: {
5545
+ input: 0.7,
5546
+ output: 2.5,
5547
+ cacheRead: 0,
5548
+ cacheWrite: 0,
5549
+ },
5550
+ contextWindow: 64000,
5551
+ maxTokens: 32768,
5552
+ },
5315
5553
  "deepseek-ai/DeepSeek-R1-0528": {
5316
5554
  id: "deepseek-ai/DeepSeek-R1-0528",
5317
5555
  name: "DeepSeek-R1-0528",
@@ -5348,6 +5586,24 @@ export const MODELS = {
5348
5586
  contextWindow: 163840,
5349
5587
  maxTokens: 65536,
5350
5588
  },
5589
+ "deepseek-ai/DeepSeek-V4-Flash": {
5590
+ id: "deepseek-ai/DeepSeek-V4-Flash",
5591
+ name: "DeepSeek V4 Flash",
5592
+ api: "openai-completions",
5593
+ provider: "huggingface",
5594
+ baseUrl: "https://router.huggingface.co/v1",
5595
+ compat: { "supportsDeveloperRole": false },
5596
+ reasoning: true,
5597
+ input: ["text"],
5598
+ cost: {
5599
+ input: 0.14,
5600
+ output: 0.28,
5601
+ cacheRead: 0,
5602
+ cacheWrite: 0,
5603
+ },
5604
+ contextWindow: 1048576,
5605
+ maxTokens: 384000,
5606
+ },
5351
5607
  "deepseek-ai/DeepSeek-V4-Pro": {
5352
5608
  id: "deepseek-ai/DeepSeek-V4-Pro",
5353
5609
  name: "DeepSeek V4 Pro",
@@ -5366,6 +5622,60 @@ export const MODELS = {
5366
5622
  contextWindow: 1048576,
5367
5623
  maxTokens: 393216,
5368
5624
  },
5625
+ "google/gemma-4-26B-A4B-it": {
5626
+ id: "google/gemma-4-26B-A4B-it",
5627
+ name: "Gemma 4 26B A4B IT",
5628
+ api: "openai-completions",
5629
+ provider: "huggingface",
5630
+ baseUrl: "https://router.huggingface.co/v1",
5631
+ compat: { "supportsDeveloperRole": false },
5632
+ reasoning: true,
5633
+ input: ["text", "image"],
5634
+ cost: {
5635
+ input: 0.13,
5636
+ output: 0.4,
5637
+ cacheRead: 0,
5638
+ cacheWrite: 0,
5639
+ },
5640
+ contextWindow: 262144,
5641
+ maxTokens: 32768,
5642
+ },
5643
+ "google/gemma-4-31B-it": {
5644
+ id: "google/gemma-4-31B-it",
5645
+ name: "Gemma 4 31B IT",
5646
+ api: "openai-completions",
5647
+ provider: "huggingface",
5648
+ baseUrl: "https://router.huggingface.co/v1",
5649
+ compat: { "supportsDeveloperRole": false },
5650
+ reasoning: true,
5651
+ input: ["text", "image"],
5652
+ cost: {
5653
+ input: 0.14,
5654
+ output: 0.4,
5655
+ cacheRead: 0,
5656
+ cacheWrite: 0,
5657
+ },
5658
+ contextWindow: 262144,
5659
+ maxTokens: 32768,
5660
+ },
5661
+ "meta-llama/Llama-3.3-70B-Instruct": {
5662
+ id: "meta-llama/Llama-3.3-70B-Instruct",
5663
+ name: "Llama-3.3-70B-Instruct",
5664
+ api: "openai-completions",
5665
+ provider: "huggingface",
5666
+ baseUrl: "https://router.huggingface.co/v1",
5667
+ compat: { "supportsDeveloperRole": false },
5668
+ reasoning: false,
5669
+ input: ["text"],
5670
+ cost: {
5671
+ input: 0.59,
5672
+ output: 0.79,
5673
+ cacheRead: 0,
5674
+ cacheWrite: 0,
5675
+ },
5676
+ contextWindow: 131072,
5677
+ maxTokens: 4096,
5678
+ },
5369
5679
  "moonshotai/Kimi-K2-Instruct": {
5370
5680
  id: "moonshotai/Kimi-K2-Instruct",
5371
5681
  name: "Kimi-K2-Instruct",
@@ -5456,6 +5766,114 @@ export const MODELS = {
5456
5766
  contextWindow: 262144,
5457
5767
  maxTokens: 262144,
5458
5768
  },
5769
+ "moonshotai/Kimi-K2.7-Code": {
5770
+ id: "moonshotai/Kimi-K2.7-Code",
5771
+ name: "Kimi K2.7 Code",
5772
+ api: "openai-completions",
5773
+ provider: "huggingface",
5774
+ baseUrl: "https://router.huggingface.co/v1",
5775
+ compat: { "supportsDeveloperRole": false },
5776
+ reasoning: true,
5777
+ input: ["text", "image"],
5778
+ cost: {
5779
+ input: 0.95,
5780
+ output: 4,
5781
+ cacheRead: 0,
5782
+ cacheWrite: 0,
5783
+ },
5784
+ contextWindow: 262144,
5785
+ maxTokens: 262144,
5786
+ },
5787
+ "stepfun-ai/Step-3.5-Flash": {
5788
+ id: "stepfun-ai/Step-3.5-Flash",
5789
+ name: "Step 3.5 Flash",
5790
+ api: "openai-completions",
5791
+ provider: "huggingface",
5792
+ baseUrl: "https://router.huggingface.co/v1",
5793
+ compat: { "supportsDeveloperRole": false },
5794
+ reasoning: true,
5795
+ input: ["text"],
5796
+ cost: {
5797
+ input: 0.1,
5798
+ output: 0.3,
5799
+ cacheRead: 0,
5800
+ cacheWrite: 0,
5801
+ },
5802
+ contextWindow: 262144,
5803
+ maxTokens: 256000,
5804
+ },
5805
+ "zai-org/GLM-4.5": {
5806
+ id: "zai-org/GLM-4.5",
5807
+ name: "GLM-4.5",
5808
+ api: "openai-completions",
5809
+ provider: "huggingface",
5810
+ baseUrl: "https://router.huggingface.co/v1",
5811
+ compat: { "supportsDeveloperRole": false },
5812
+ reasoning: true,
5813
+ input: ["text"],
5814
+ cost: {
5815
+ input: 0.6,
5816
+ output: 2.2,
5817
+ cacheRead: 0,
5818
+ cacheWrite: 0,
5819
+ },
5820
+ contextWindow: 131072,
5821
+ maxTokens: 98304,
5822
+ },
5823
+ "zai-org/GLM-4.5-Air": {
5824
+ id: "zai-org/GLM-4.5-Air",
5825
+ name: "GLM-4.5-Air",
5826
+ api: "openai-completions",
5827
+ provider: "huggingface",
5828
+ baseUrl: "https://router.huggingface.co/v1",
5829
+ compat: { "supportsDeveloperRole": false },
5830
+ reasoning: true,
5831
+ input: ["text"],
5832
+ cost: {
5833
+ input: 0.13,
5834
+ output: 0.85,
5835
+ cacheRead: 0,
5836
+ cacheWrite: 0,
5837
+ },
5838
+ contextWindow: 131072,
5839
+ maxTokens: 98304,
5840
+ },
5841
+ "zai-org/GLM-4.5V": {
5842
+ id: "zai-org/GLM-4.5V",
5843
+ name: "GLM-4.5V",
5844
+ api: "openai-completions",
5845
+ provider: "huggingface",
5846
+ baseUrl: "https://router.huggingface.co/v1",
5847
+ compat: { "supportsDeveloperRole": false },
5848
+ reasoning: true,
5849
+ input: ["text", "image"],
5850
+ cost: {
5851
+ input: 0.6,
5852
+ output: 1.8,
5853
+ cacheRead: 0,
5854
+ cacheWrite: 0,
5855
+ },
5856
+ contextWindow: 65536,
5857
+ maxTokens: 16384,
5858
+ },
5859
+ "zai-org/GLM-4.6": {
5860
+ id: "zai-org/GLM-4.6",
5861
+ name: "GLM-4.6",
5862
+ api: "openai-completions",
5863
+ provider: "huggingface",
5864
+ baseUrl: "https://router.huggingface.co/v1",
5865
+ compat: { "supportsDeveloperRole": false },
5866
+ reasoning: true,
5867
+ input: ["text"],
5868
+ cost: {
5869
+ input: 0.55,
5870
+ output: 2.2,
5871
+ cacheRead: 0,
5872
+ cacheWrite: 0,
5873
+ },
5874
+ contextWindow: 204800,
5875
+ maxTokens: 131072,
5876
+ },
5459
5877
  "zai-org/GLM-4.7": {
5460
5878
  id: "zai-org/GLM-4.7",
5461
5879
  name: "GLM-4.7",
@@ -5528,6 +5946,24 @@ export const MODELS = {
5528
5946
  contextWindow: 202752,
5529
5947
  maxTokens: 131072,
5530
5948
  },
5949
+ "zai-org/GLM-5.2": {
5950
+ id: "zai-org/GLM-5.2",
5951
+ name: "GLM-5.2",
5952
+ api: "openai-completions",
5953
+ provider: "huggingface",
5954
+ baseUrl: "https://router.huggingface.co/v1",
5955
+ compat: { "supportsDeveloperRole": false },
5956
+ reasoning: true,
5957
+ input: ["text"],
5958
+ cost: {
5959
+ input: 1.4,
5960
+ output: 4.4,
5961
+ cacheRead: 0,
5962
+ cacheWrite: 0,
5963
+ },
5964
+ contextWindow: 262144,
5965
+ maxTokens: 131072,
5966
+ },
5531
5967
  },
5532
5968
  "kimi-coding": {
5533
5969
  "k2p7": {
@@ -7690,6 +8126,23 @@ export const MODELS = {
7690
8126
  contextWindow: 204800,
7691
8127
  maxTokens: 131072,
7692
8128
  },
8129
+ "glm-5.2": {
8130
+ id: "glm-5.2",
8131
+ name: "GLM-5.2",
8132
+ api: "openai-completions",
8133
+ provider: "opencode",
8134
+ baseUrl: "https://opencode.ai/zen/v1",
8135
+ reasoning: true,
8136
+ input: ["text"],
8137
+ cost: {
8138
+ input: 1.4,
8139
+ output: 4.4,
8140
+ cacheRead: 0.26,
8141
+ cacheWrite: 0,
8142
+ },
8143
+ contextWindow: 1000000,
8144
+ maxTokens: 131072,
8145
+ },
7693
8146
  "gpt-5": {
7694
8147
  id: "gpt-5",
7695
8148
  name: "GPT-5",
@@ -8502,23 +8955,6 @@ export const MODELS = {
8502
8955
  contextWindow: 200000,
8503
8956
  maxTokens: 4096,
8504
8957
  },
8505
- "anthropic/claude-3.5-haiku": {
8506
- id: "anthropic/claude-3.5-haiku",
8507
- name: "Anthropic: Claude 3.5 Haiku",
8508
- api: "openai-completions",
8509
- provider: "openrouter",
8510
- baseUrl: "https://openrouter.ai/api/v1",
8511
- reasoning: false,
8512
- input: ["text", "image"],
8513
- cost: {
8514
- input: 0.7999999999999999,
8515
- output: 4,
8516
- cacheRead: 0.08,
8517
- cacheWrite: 1,
8518
- },
8519
- contextWindow: 200000,
8520
- maxTokens: 8192,
8521
- },
8522
8958
  "anthropic/claude-fable-5": {
8523
8959
  id: "anthropic/claude-fable-5",
8524
8960
  name: "Anthropic: Claude Fable 5",
@@ -9124,23 +9560,6 @@ export const MODELS = {
9124
9560
  contextWindow: 1048576,
9125
9561
  maxTokens: 384000,
9126
9562
  },
9127
- "essentialai/rnj-1-instruct": {
9128
- id: "essentialai/rnj-1-instruct",
9129
- name: "EssentialAI: Rnj 1 Instruct",
9130
- api: "openai-completions",
9131
- provider: "openrouter",
9132
- baseUrl: "https://openrouter.ai/api/v1",
9133
- reasoning: false,
9134
- input: ["text"],
9135
- cost: {
9136
- input: 0.15,
9137
- output: 0.15,
9138
- cacheRead: 0,
9139
- cacheWrite: 0,
9140
- },
9141
- contextWindow: 32768,
9142
- maxTokens: 4096,
9143
- },
9144
9563
  "google/gemini-2.5-flash": {
9145
9564
  id: "google/gemini-2.5-flash",
9146
9565
  name: "Google: Gemini 2.5 Flash",
@@ -9763,13 +10182,13 @@ export const MODELS = {
9763
10182
  reasoning: true,
9764
10183
  input: ["text"],
9765
10184
  cost: {
9766
- input: 0.25,
9767
- output: 1,
9768
- cacheRead: 0.049999999999999996,
10185
+ input: 0.24,
10186
+ output: 0.96,
10187
+ cacheRead: 0,
9769
10188
  cacheWrite: 0,
9770
10189
  },
9771
10190
  contextWindow: 204800,
9772
- maxTokens: 131072,
10191
+ maxTokens: 196608,
9773
10192
  },
9774
10193
  "minimax/minimax-m3": {
9775
10194
  id: "minimax/minimax-m3",
@@ -10156,12 +10575,12 @@ export const MODELS = {
10156
10575
  input: ["text", "image"],
10157
10576
  cost: {
10158
10577
  input: 0.66,
10159
- output: 3.5,
10160
- cacheRead: 0.33,
10578
+ output: 3.41,
10579
+ cacheRead: 0.144,
10161
10580
  cacheWrite: 0,
10162
10581
  },
10163
10582
  contextWindow: 262144,
10164
- maxTokens: 65535,
10583
+ maxTokens: 262144,
10165
10584
  },
10166
10585
  "moonshotai/kimi-k2.7-code": {
10167
10586
  id: "moonshotai/kimi-k2.7-code",
@@ -10172,30 +10591,13 @@ export const MODELS = {
10172
10591
  reasoning: true,
10173
10592
  input: ["text", "image"],
10174
10593
  cost: {
10175
- input: 0.612,
10176
- output: 3.0690000000000004,
10177
- cacheRead: 0.1296,
10178
- cacheWrite: 0,
10179
- },
10180
- contextWindow: 262144,
10181
- maxTokens: 262144,
10182
- },
10183
- "nex-agi/nex-n2-pro:free": {
10184
- id: "nex-agi/nex-n2-pro:free",
10185
- name: "Nex AGI: Nex-N2-Pro (free)",
10186
- api: "openai-completions",
10187
- provider: "openrouter",
10188
- baseUrl: "https://openrouter.ai/api/v1",
10189
- reasoning: true,
10190
- input: ["text", "image"],
10191
- cost: {
10192
- input: 0,
10193
- output: 0,
10194
- cacheRead: 0,
10594
+ input: 0.74,
10595
+ output: 3.5,
10596
+ cacheRead: 0.15,
10195
10597
  cacheWrite: 0,
10196
10598
  },
10197
10599
  contextWindow: 262144,
10198
- maxTokens: 262144,
10600
+ maxTokens: 16384,
10199
10601
  },
10200
10602
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
10201
10603
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -11433,23 +11835,6 @@ export const MODELS = {
11433
11835
  contextWindow: 262144,
11434
11836
  maxTokens: 32768,
11435
11837
  },
11436
- "prime-intellect/intellect-3": {
11437
- id: "prime-intellect/intellect-3",
11438
- name: "Prime Intellect: INTELLECT-3",
11439
- api: "openai-completions",
11440
- provider: "openrouter",
11441
- baseUrl: "https://openrouter.ai/api/v1",
11442
- reasoning: true,
11443
- input: ["text"],
11444
- cost: {
11445
- input: 0.19999999999999998,
11446
- output: 1.1,
11447
- cacheRead: 0,
11448
- cacheWrite: 0,
11449
- },
11450
- contextWindow: 131072,
11451
- maxTokens: 131072,
11452
- },
11453
11838
  "qwen/qwen-2.5-72b-instruct": {
11454
11839
  id: "qwen/qwen-2.5-72b-instruct",
11455
11840
  name: "Qwen2.5 72B Instruct",
@@ -12351,23 +12736,6 @@ export const MODELS = {
12351
12736
  contextWindow: 262144,
12352
12737
  maxTokens: 4096,
12353
12738
  },
12354
- "thedrummer/rocinante-12b": {
12355
- id: "thedrummer/rocinante-12b",
12356
- name: "TheDrummer: Rocinante 12B",
12357
- api: "openai-completions",
12358
- provider: "openrouter",
12359
- baseUrl: "https://openrouter.ai/api/v1",
12360
- reasoning: false,
12361
- input: ["text"],
12362
- cost: {
12363
- input: 0.16999999999999998,
12364
- output: 0.43,
12365
- cacheRead: 0,
12366
- cacheWrite: 0,
12367
- },
12368
- contextWindow: 32768,
12369
- maxTokens: 32768,
12370
- },
12371
12739
  "thedrummer/unslopnemo-12b": {
12372
12740
  id: "thedrummer/unslopnemo-12b",
12373
12741
  name: "TheDrummer: UnslopNemo 12B",
@@ -12462,13 +12830,13 @@ export const MODELS = {
12462
12830
  reasoning: true,
12463
12831
  input: ["text", "image"],
12464
12832
  cost: {
12465
- input: 0.14,
12833
+ input: 0.105,
12466
12834
  output: 0.28,
12467
- cacheRead: 0.0028,
12835
+ cacheRead: 0,
12468
12836
  cacheWrite: 0,
12469
12837
  },
12470
12838
  contextWindow: 1048576,
12471
- maxTokens: 131072,
12839
+ maxTokens: 4096,
12472
12840
  },
12473
12841
  "xiaomi/mimo-v2.5-pro": {
12474
12842
  id: "xiaomi/mimo-v2.5-pro",
@@ -12666,14 +13034,31 @@ export const MODELS = {
12666
13034
  reasoning: true,
12667
13035
  input: ["text"],
12668
13036
  cost: {
12669
- input: 1,
12670
- output: 4,
13037
+ input: 0.95,
13038
+ output: 3,
12671
13039
  cacheRead: 0.18,
12672
13040
  cacheWrite: 0,
12673
13041
  },
12674
13042
  contextWindow: 1048576,
12675
13043
  maxTokens: 32768,
12676
13044
  },
13045
+ "z-ai/glm-5v-turbo": {
13046
+ id: "z-ai/glm-5v-turbo",
13047
+ name: "Z.ai: GLM 5V Turbo",
13048
+ api: "openai-completions",
13049
+ provider: "openrouter",
13050
+ baseUrl: "https://openrouter.ai/api/v1",
13051
+ reasoning: true,
13052
+ input: ["text", "image"],
13053
+ cost: {
13054
+ input: 1.2,
13055
+ output: 4,
13056
+ cacheRead: 0.24,
13057
+ cacheWrite: 0,
13058
+ },
13059
+ contextWindow: 202752,
13060
+ maxTokens: 131072,
13061
+ },
12677
13062
  "~anthropic/claude-fable-latest": {
12678
13063
  id: "~anthropic/claude-fable-latest",
12679
13064
  name: "Anthropic: Claude Fable Latest",
@@ -12786,12 +13171,12 @@ export const MODELS = {
12786
13171
  input: ["text", "image"],
12787
13172
  cost: {
12788
13173
  input: 0.66,
12789
- output: 3.5,
12790
- cacheRead: 0.33,
13174
+ output: 3.41,
13175
+ cacheRead: 0.144,
12791
13176
  cacheWrite: 0,
12792
13177
  },
12793
13178
  contextWindow: 262144,
12794
- maxTokens: 65535,
13179
+ maxTokens: 262144,
12795
13180
  },
12796
13181
  "~openai/gpt-latest": {
12797
13182
  id: "~openai/gpt-latest",
@@ -15552,6 +15937,23 @@ export const MODELS = {
15552
15937
  contextWindow: 200000,
15553
15938
  maxTokens: 8000,
15554
15939
  },
15940
+ "sakana/fugu-ultra": {
15941
+ id: "sakana/fugu-ultra",
15942
+ name: "Fugu Ultra",
15943
+ api: "anthropic-messages",
15944
+ provider: "vercel-ai-gateway",
15945
+ baseUrl: "https://ai-gateway.vercel.sh",
15946
+ reasoning: true,
15947
+ input: ["text", "image"],
15948
+ cost: {
15949
+ input: 5,
15950
+ output: 30,
15951
+ cacheRead: 0.5,
15952
+ cacheWrite: 0,
15953
+ },
15954
+ contextWindow: 1000000,
15955
+ maxTokens: 1000000,
15956
+ },
15555
15957
  "stepfun/step-3.5-flash": {
15556
15958
  id: "stepfun/step-3.5-flash",
15557
15959
  name: "StepFun 3.5 Flash",