@caupulican/pi-ai 0.80.31 → 0.80.37

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.
@@ -4781,42 +4781,6 @@ export const MODELS = {
4781
4781
  contextWindow: 262144,
4782
4782
  maxTokens: 32768,
4783
4783
  },
4784
- "gemma-4-E2B-it": {
4785
- id: "gemma-4-E2B-it",
4786
- name: "Gemma 4 E2B IT",
4787
- api: "google-generative-ai",
4788
- provider: "google",
4789
- baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4790
- reasoning: true,
4791
- thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4792
- input: ["text", "image"],
4793
- cost: {
4794
- input: 0,
4795
- output: 0,
4796
- cacheRead: 0,
4797
- cacheWrite: 0,
4798
- },
4799
- contextWindow: 131072,
4800
- maxTokens: 8192,
4801
- },
4802
- "gemma-4-E4B-it": {
4803
- id: "gemma-4-E4B-it",
4804
- name: "Gemma 4 E4B IT",
4805
- api: "google-generative-ai",
4806
- provider: "google",
4807
- baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4808
- reasoning: true,
4809
- thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4810
- input: ["text", "image"],
4811
- cost: {
4812
- input: 0,
4813
- output: 0,
4814
- cacheRead: 0,
4815
- cacheWrite: 0,
4816
- },
4817
- contextWindow: 131072,
4818
- maxTokens: 8192,
4819
- },
4820
4784
  },
4821
4785
  "google-vertex": {
4822
4786
  "gemini-1.5-flash": {
@@ -5168,6 +5132,24 @@ export const MODELS = {
5168
5132
  },
5169
5133
  },
5170
5134
  "huggingface": {
5135
+ "MiniMaxAI/MiniMax-M2": {
5136
+ id: "MiniMaxAI/MiniMax-M2",
5137
+ name: "MiniMax-M2",
5138
+ api: "openai-completions",
5139
+ provider: "huggingface",
5140
+ baseUrl: "https://router.huggingface.co/v1",
5141
+ compat: { "supportsDeveloperRole": false },
5142
+ reasoning: true,
5143
+ input: ["text"],
5144
+ cost: {
5145
+ input: 0.3,
5146
+ output: 1.2,
5147
+ cacheRead: 0,
5148
+ cacheWrite: 0,
5149
+ },
5150
+ contextWindow: 204800,
5151
+ maxTokens: 128000,
5152
+ },
5171
5153
  "MiniMaxAI/MiniMax-M2.1": {
5172
5154
  id: "MiniMaxAI/MiniMax-M2.1",
5173
5155
  name: "MiniMax-M2.1",
@@ -5222,6 +5204,42 @@ export const MODELS = {
5222
5204
  contextWindow: 204800,
5223
5205
  maxTokens: 131072,
5224
5206
  },
5207
+ "MiniMaxAI/MiniMax-M3": {
5208
+ id: "MiniMaxAI/MiniMax-M3",
5209
+ name: "MiniMax-M3",
5210
+ api: "openai-completions",
5211
+ provider: "huggingface",
5212
+ baseUrl: "https://router.huggingface.co/v1",
5213
+ compat: { "supportsDeveloperRole": false },
5214
+ reasoning: true,
5215
+ input: ["text", "image"],
5216
+ cost: {
5217
+ input: 0.3,
5218
+ output: 1.2,
5219
+ cacheRead: 0,
5220
+ cacheWrite: 0,
5221
+ },
5222
+ contextWindow: 524288,
5223
+ maxTokens: 128000,
5224
+ },
5225
+ "Qwen/Qwen3-235B-A22B": {
5226
+ id: "Qwen/Qwen3-235B-A22B",
5227
+ name: "Qwen3 235B-A22B",
5228
+ api: "openai-completions",
5229
+ provider: "huggingface",
5230
+ baseUrl: "https://router.huggingface.co/v1",
5231
+ compat: { "supportsDeveloperRole": false },
5232
+ reasoning: true,
5233
+ input: ["text"],
5234
+ cost: {
5235
+ input: 0.2,
5236
+ output: 0.8,
5237
+ cacheRead: 0,
5238
+ cacheWrite: 0,
5239
+ },
5240
+ contextWindow: 40960,
5241
+ maxTokens: 16384,
5242
+ },
5225
5243
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
5226
5244
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
5227
5245
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -5240,6 +5258,42 @@ export const MODELS = {
5240
5258
  contextWindow: 262144,
5241
5259
  maxTokens: 131072,
5242
5260
  },
5261
+ "Qwen/Qwen3-32B": {
5262
+ id: "Qwen/Qwen3-32B",
5263
+ name: "Qwen3 32B",
5264
+ api: "openai-completions",
5265
+ provider: "huggingface",
5266
+ baseUrl: "https://router.huggingface.co/v1",
5267
+ compat: { "supportsDeveloperRole": false },
5268
+ reasoning: true,
5269
+ input: ["text"],
5270
+ cost: {
5271
+ input: 0.29,
5272
+ output: 0.59,
5273
+ cacheRead: 0,
5274
+ cacheWrite: 0,
5275
+ },
5276
+ contextWindow: 131072,
5277
+ maxTokens: 16384,
5278
+ },
5279
+ "Qwen/Qwen3-Coder-30B-A3B-Instruct": {
5280
+ id: "Qwen/Qwen3-Coder-30B-A3B-Instruct",
5281
+ name: "Qwen3-Coder 30B-A3B Instruct",
5282
+ api: "openai-completions",
5283
+ provider: "huggingface",
5284
+ baseUrl: "https://router.huggingface.co/v1",
5285
+ compat: { "supportsDeveloperRole": false },
5286
+ reasoning: false,
5287
+ input: ["text"],
5288
+ cost: {
5289
+ input: 0.07,
5290
+ output: 0.26,
5291
+ cacheRead: 0,
5292
+ cacheWrite: 0,
5293
+ },
5294
+ contextWindow: 262144,
5295
+ maxTokens: 65536,
5296
+ },
5243
5297
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
5244
5298
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
5245
5299
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -5312,6 +5366,60 @@ export const MODELS = {
5312
5366
  contextWindow: 262144,
5313
5367
  maxTokens: 131072,
5314
5368
  },
5369
+ "Qwen/Qwen3.5-122B-A10B": {
5370
+ id: "Qwen/Qwen3.5-122B-A10B",
5371
+ name: "Qwen3.5 122B-A10B",
5372
+ api: "openai-completions",
5373
+ provider: "huggingface",
5374
+ baseUrl: "https://router.huggingface.co/v1",
5375
+ compat: { "supportsDeveloperRole": false },
5376
+ reasoning: true,
5377
+ input: ["text", "image"],
5378
+ cost: {
5379
+ input: 0.4,
5380
+ output: 3.2,
5381
+ cacheRead: 0,
5382
+ cacheWrite: 0,
5383
+ },
5384
+ contextWindow: 262144,
5385
+ maxTokens: 65536,
5386
+ },
5387
+ "Qwen/Qwen3.5-27B": {
5388
+ id: "Qwen/Qwen3.5-27B",
5389
+ name: "Qwen3.5 27B",
5390
+ api: "openai-completions",
5391
+ provider: "huggingface",
5392
+ baseUrl: "https://router.huggingface.co/v1",
5393
+ compat: { "supportsDeveloperRole": false },
5394
+ reasoning: true,
5395
+ input: ["text", "image"],
5396
+ cost: {
5397
+ input: 0.3,
5398
+ output: 2.4,
5399
+ cacheRead: 0,
5400
+ cacheWrite: 0,
5401
+ },
5402
+ contextWindow: 262144,
5403
+ maxTokens: 65536,
5404
+ },
5405
+ "Qwen/Qwen3.5-35B-A3B": {
5406
+ id: "Qwen/Qwen3.5-35B-A3B",
5407
+ name: "Qwen3.5 35B-A3B",
5408
+ api: "openai-completions",
5409
+ provider: "huggingface",
5410
+ baseUrl: "https://router.huggingface.co/v1",
5411
+ compat: { "supportsDeveloperRole": false },
5412
+ reasoning: true,
5413
+ input: ["text", "image"],
5414
+ cost: {
5415
+ input: 0.25,
5416
+ output: 2,
5417
+ cacheRead: 0,
5418
+ cacheWrite: 0,
5419
+ },
5420
+ contextWindow: 262144,
5421
+ maxTokens: 65536,
5422
+ },
5315
5423
  "Qwen/Qwen3.5-397B-A17B": {
5316
5424
  id: "Qwen/Qwen3.5-397B-A17B",
5317
5425
  name: "Qwen3.5-397B-A17B",
@@ -5330,6 +5438,42 @@ export const MODELS = {
5330
5438
  contextWindow: 262144,
5331
5439
  maxTokens: 32768,
5332
5440
  },
5441
+ "Qwen/Qwen3.5-9B": {
5442
+ id: "Qwen/Qwen3.5-9B",
5443
+ name: "Qwen3.5 9B",
5444
+ api: "openai-completions",
5445
+ provider: "huggingface",
5446
+ baseUrl: "https://router.huggingface.co/v1",
5447
+ compat: { "supportsDeveloperRole": false },
5448
+ reasoning: true,
5449
+ input: ["text", "image"],
5450
+ cost: {
5451
+ input: 0.17,
5452
+ output: 0.25,
5453
+ cacheRead: 0,
5454
+ cacheWrite: 0,
5455
+ },
5456
+ contextWindow: 262144,
5457
+ maxTokens: 65536,
5458
+ },
5459
+ "Qwen/Qwen3.6-35B-A3B": {
5460
+ id: "Qwen/Qwen3.6-35B-A3B",
5461
+ name: "Qwen3.6 35B-A3B",
5462
+ api: "openai-completions",
5463
+ provider: "huggingface",
5464
+ baseUrl: "https://router.huggingface.co/v1",
5465
+ compat: { "supportsDeveloperRole": false },
5466
+ reasoning: true,
5467
+ input: ["text", "image"],
5468
+ cost: {
5469
+ input: 0.15,
5470
+ output: 0.95,
5471
+ cacheRead: 0,
5472
+ cacheWrite: 0,
5473
+ },
5474
+ contextWindow: 262144,
5475
+ maxTokens: 65536,
5476
+ },
5333
5477
  "XiaomiMiMo/MiMo-V2-Flash": {
5334
5478
  id: "XiaomiMiMo/MiMo-V2-Flash",
5335
5479
  name: "MiMo-V2-Flash",
@@ -5348,6 +5492,24 @@ export const MODELS = {
5348
5492
  contextWindow: 262144,
5349
5493
  maxTokens: 4096,
5350
5494
  },
5495
+ "deepseek-ai/DeepSeek-R1": {
5496
+ id: "deepseek-ai/DeepSeek-R1",
5497
+ name: "DeepSeek-R1",
5498
+ api: "openai-completions",
5499
+ provider: "huggingface",
5500
+ baseUrl: "https://router.huggingface.co/v1",
5501
+ compat: { "supportsDeveloperRole": false },
5502
+ reasoning: true,
5503
+ input: ["text"],
5504
+ cost: {
5505
+ input: 0.7,
5506
+ output: 2.5,
5507
+ cacheRead: 0,
5508
+ cacheWrite: 0,
5509
+ },
5510
+ contextWindow: 64000,
5511
+ maxTokens: 32768,
5512
+ },
5351
5513
  "deepseek-ai/DeepSeek-R1-0528": {
5352
5514
  id: "deepseek-ai/DeepSeek-R1-0528",
5353
5515
  name: "DeepSeek-R1-0528",
@@ -5384,6 +5546,24 @@ export const MODELS = {
5384
5546
  contextWindow: 163840,
5385
5547
  maxTokens: 65536,
5386
5548
  },
5549
+ "deepseek-ai/DeepSeek-V4-Flash": {
5550
+ id: "deepseek-ai/DeepSeek-V4-Flash",
5551
+ name: "DeepSeek V4 Flash",
5552
+ api: "openai-completions",
5553
+ provider: "huggingface",
5554
+ baseUrl: "https://router.huggingface.co/v1",
5555
+ compat: { "supportsDeveloperRole": false },
5556
+ reasoning: true,
5557
+ input: ["text"],
5558
+ cost: {
5559
+ input: 0.14,
5560
+ output: 0.28,
5561
+ cacheRead: 0,
5562
+ cacheWrite: 0,
5563
+ },
5564
+ contextWindow: 1048576,
5565
+ maxTokens: 384000,
5566
+ },
5387
5567
  "deepseek-ai/DeepSeek-V4-Pro": {
5388
5568
  id: "deepseek-ai/DeepSeek-V4-Pro",
5389
5569
  name: "DeepSeek V4 Pro",
@@ -5402,6 +5582,60 @@ export const MODELS = {
5402
5582
  contextWindow: 1048576,
5403
5583
  maxTokens: 393216,
5404
5584
  },
5585
+ "google/gemma-4-26B-A4B-it": {
5586
+ id: "google/gemma-4-26B-A4B-it",
5587
+ name: "Gemma 4 26B A4B IT",
5588
+ api: "openai-completions",
5589
+ provider: "huggingface",
5590
+ baseUrl: "https://router.huggingface.co/v1",
5591
+ compat: { "supportsDeveloperRole": false },
5592
+ reasoning: true,
5593
+ input: ["text", "image"],
5594
+ cost: {
5595
+ input: 0.13,
5596
+ output: 0.4,
5597
+ cacheRead: 0,
5598
+ cacheWrite: 0,
5599
+ },
5600
+ contextWindow: 262144,
5601
+ maxTokens: 32768,
5602
+ },
5603
+ "google/gemma-4-31B-it": {
5604
+ id: "google/gemma-4-31B-it",
5605
+ name: "Gemma 4 31B IT",
5606
+ api: "openai-completions",
5607
+ provider: "huggingface",
5608
+ baseUrl: "https://router.huggingface.co/v1",
5609
+ compat: { "supportsDeveloperRole": false },
5610
+ reasoning: true,
5611
+ input: ["text", "image"],
5612
+ cost: {
5613
+ input: 0.14,
5614
+ output: 0.4,
5615
+ cacheRead: 0,
5616
+ cacheWrite: 0,
5617
+ },
5618
+ contextWindow: 262144,
5619
+ maxTokens: 32768,
5620
+ },
5621
+ "meta-llama/Llama-3.3-70B-Instruct": {
5622
+ id: "meta-llama/Llama-3.3-70B-Instruct",
5623
+ name: "Llama-3.3-70B-Instruct",
5624
+ api: "openai-completions",
5625
+ provider: "huggingface",
5626
+ baseUrl: "https://router.huggingface.co/v1",
5627
+ compat: { "supportsDeveloperRole": false },
5628
+ reasoning: false,
5629
+ input: ["text"],
5630
+ cost: {
5631
+ input: 0.59,
5632
+ output: 0.79,
5633
+ cacheRead: 0,
5634
+ cacheWrite: 0,
5635
+ },
5636
+ contextWindow: 131072,
5637
+ maxTokens: 4096,
5638
+ },
5405
5639
  "moonshotai/Kimi-K2-Instruct": {
5406
5640
  id: "moonshotai/Kimi-K2-Instruct",
5407
5641
  name: "Kimi-K2-Instruct",
@@ -5492,6 +5726,114 @@ export const MODELS = {
5492
5726
  contextWindow: 262144,
5493
5727
  maxTokens: 262144,
5494
5728
  },
5729
+ "moonshotai/Kimi-K2.7-Code": {
5730
+ id: "moonshotai/Kimi-K2.7-Code",
5731
+ name: "Kimi K2.7 Code",
5732
+ api: "openai-completions",
5733
+ provider: "huggingface",
5734
+ baseUrl: "https://router.huggingface.co/v1",
5735
+ compat: { "supportsDeveloperRole": false },
5736
+ reasoning: true,
5737
+ input: ["text", "image"],
5738
+ cost: {
5739
+ input: 0.95,
5740
+ output: 4,
5741
+ cacheRead: 0,
5742
+ cacheWrite: 0,
5743
+ },
5744
+ contextWindow: 262144,
5745
+ maxTokens: 262144,
5746
+ },
5747
+ "stepfun-ai/Step-3.5-Flash": {
5748
+ id: "stepfun-ai/Step-3.5-Flash",
5749
+ name: "Step 3.5 Flash",
5750
+ api: "openai-completions",
5751
+ provider: "huggingface",
5752
+ baseUrl: "https://router.huggingface.co/v1",
5753
+ compat: { "supportsDeveloperRole": false },
5754
+ reasoning: true,
5755
+ input: ["text"],
5756
+ cost: {
5757
+ input: 0.1,
5758
+ output: 0.3,
5759
+ cacheRead: 0,
5760
+ cacheWrite: 0,
5761
+ },
5762
+ contextWindow: 262144,
5763
+ maxTokens: 256000,
5764
+ },
5765
+ "zai-org/GLM-4.5": {
5766
+ id: "zai-org/GLM-4.5",
5767
+ name: "GLM-4.5",
5768
+ api: "openai-completions",
5769
+ provider: "huggingface",
5770
+ baseUrl: "https://router.huggingface.co/v1",
5771
+ compat: { "supportsDeveloperRole": false },
5772
+ reasoning: true,
5773
+ input: ["text"],
5774
+ cost: {
5775
+ input: 0.6,
5776
+ output: 2.2,
5777
+ cacheRead: 0,
5778
+ cacheWrite: 0,
5779
+ },
5780
+ contextWindow: 131072,
5781
+ maxTokens: 98304,
5782
+ },
5783
+ "zai-org/GLM-4.5-Air": {
5784
+ id: "zai-org/GLM-4.5-Air",
5785
+ name: "GLM-4.5-Air",
5786
+ api: "openai-completions",
5787
+ provider: "huggingface",
5788
+ baseUrl: "https://router.huggingface.co/v1",
5789
+ compat: { "supportsDeveloperRole": false },
5790
+ reasoning: true,
5791
+ input: ["text"],
5792
+ cost: {
5793
+ input: 0.13,
5794
+ output: 0.85,
5795
+ cacheRead: 0,
5796
+ cacheWrite: 0,
5797
+ },
5798
+ contextWindow: 131072,
5799
+ maxTokens: 98304,
5800
+ },
5801
+ "zai-org/GLM-4.5V": {
5802
+ id: "zai-org/GLM-4.5V",
5803
+ name: "GLM-4.5V",
5804
+ api: "openai-completions",
5805
+ provider: "huggingface",
5806
+ baseUrl: "https://router.huggingface.co/v1",
5807
+ compat: { "supportsDeveloperRole": false },
5808
+ reasoning: true,
5809
+ input: ["text", "image"],
5810
+ cost: {
5811
+ input: 0.6,
5812
+ output: 1.8,
5813
+ cacheRead: 0,
5814
+ cacheWrite: 0,
5815
+ },
5816
+ contextWindow: 65536,
5817
+ maxTokens: 16384,
5818
+ },
5819
+ "zai-org/GLM-4.6": {
5820
+ id: "zai-org/GLM-4.6",
5821
+ name: "GLM-4.6",
5822
+ api: "openai-completions",
5823
+ provider: "huggingface",
5824
+ baseUrl: "https://router.huggingface.co/v1",
5825
+ compat: { "supportsDeveloperRole": false },
5826
+ reasoning: true,
5827
+ input: ["text"],
5828
+ cost: {
5829
+ input: 0.55,
5830
+ output: 2.2,
5831
+ cacheRead: 0,
5832
+ cacheWrite: 0,
5833
+ },
5834
+ contextWindow: 204800,
5835
+ maxTokens: 131072,
5836
+ },
5495
5837
  "zai-org/GLM-4.7": {
5496
5838
  id: "zai-org/GLM-4.7",
5497
5839
  name: "GLM-4.7",
@@ -5564,6 +5906,24 @@ export const MODELS = {
5564
5906
  contextWindow: 202752,
5565
5907
  maxTokens: 131072,
5566
5908
  },
5909
+ "zai-org/GLM-5.2": {
5910
+ id: "zai-org/GLM-5.2",
5911
+ name: "GLM-5.2",
5912
+ api: "openai-completions",
5913
+ provider: "huggingface",
5914
+ baseUrl: "https://router.huggingface.co/v1",
5915
+ compat: { "supportsDeveloperRole": false },
5916
+ reasoning: true,
5917
+ input: ["text"],
5918
+ cost: {
5919
+ input: 1.4,
5920
+ output: 4.4,
5921
+ cacheRead: 0,
5922
+ cacheWrite: 0,
5923
+ },
5924
+ contextWindow: 262144,
5925
+ maxTokens: 131072,
5926
+ },
5567
5927
  },
5568
5928
  "kimi-coding": {
5569
5929
  "k2p7": {
@@ -7726,6 +8086,23 @@ export const MODELS = {
7726
8086
  contextWindow: 204800,
7727
8087
  maxTokens: 131072,
7728
8088
  },
8089
+ "glm-5.2": {
8090
+ id: "glm-5.2",
8091
+ name: "GLM-5.2",
8092
+ api: "openai-completions",
8093
+ provider: "opencode",
8094
+ baseUrl: "https://opencode.ai/zen/v1",
8095
+ reasoning: true,
8096
+ input: ["text"],
8097
+ cost: {
8098
+ input: 1.4,
8099
+ output: 4.4,
8100
+ cacheRead: 0.26,
8101
+ cacheWrite: 0,
8102
+ },
8103
+ contextWindow: 1000000,
8104
+ maxTokens: 131072,
8105
+ },
7729
8106
  "gpt-5": {
7730
8107
  id: "gpt-5",
7731
8108
  name: "GPT-5",
@@ -8538,23 +8915,6 @@ export const MODELS = {
8538
8915
  contextWindow: 200000,
8539
8916
  maxTokens: 4096,
8540
8917
  },
8541
- "anthropic/claude-3.5-haiku": {
8542
- id: "anthropic/claude-3.5-haiku",
8543
- name: "Anthropic: Claude 3.5 Haiku",
8544
- api: "openai-completions",
8545
- provider: "openrouter",
8546
- baseUrl: "https://openrouter.ai/api/v1",
8547
- reasoning: false,
8548
- input: ["text", "image"],
8549
- cost: {
8550
- input: 0.7999999999999999,
8551
- output: 4,
8552
- cacheRead: 0.08,
8553
- cacheWrite: 1,
8554
- },
8555
- contextWindow: 200000,
8556
- maxTokens: 8192,
8557
- },
8558
8918
  "anthropic/claude-fable-5": {
8559
8919
  id: "anthropic/claude-fable-5",
8560
8920
  name: "Anthropic: Claude Fable 5",
@@ -8808,13 +9168,13 @@ export const MODELS = {
8808
9168
  reasoning: true,
8809
9169
  input: ["text"],
8810
9170
  cost: {
8811
- input: 0.22,
8812
- output: 0.85,
9171
+ input: 0.25,
9172
+ output: 0.7999999999999999,
8813
9173
  cacheRead: 0.06,
8814
9174
  cacheWrite: 0,
8815
9175
  },
8816
9176
  contextWindow: 262144,
8817
- maxTokens: 262144,
9177
+ maxTokens: 80000,
8818
9178
  },
8819
9179
  "arcee-ai/trinity-mini": {
8820
9180
  id: "arcee-ai/trinity-mini",
@@ -10191,9 +10551,9 @@ export const MODELS = {
10191
10551
  reasoning: true,
10192
10552
  input: ["text", "image"],
10193
10553
  cost: {
10194
- input: 0.67,
10195
- output: 3.5,
10196
- cacheRead: 0.19999999999999998,
10554
+ input: 0.66,
10555
+ output: 3.41,
10556
+ cacheRead: 0.144,
10197
10557
  cacheWrite: 0,
10198
10558
  },
10199
10559
  contextWindow: 262144,
@@ -10208,26 +10568,9 @@ export const MODELS = {
10208
10568
  reasoning: true,
10209
10569
  input: ["text", "image"],
10210
10570
  cost: {
10211
- input: 0.74,
10212
- output: 3.5,
10213
- cacheRead: 0.15,
10214
- cacheWrite: 0,
10215
- },
10216
- contextWindow: 262144,
10217
- maxTokens: 16384,
10218
- },
10219
- "nex-agi/nex-n2-pro:free": {
10220
- id: "nex-agi/nex-n2-pro:free",
10221
- name: "Nex AGI: Nex-N2-Pro (free)",
10222
- api: "openai-completions",
10223
- provider: "openrouter",
10224
- baseUrl: "https://openrouter.ai/api/v1",
10225
- reasoning: true,
10226
- input: ["text", "image"],
10227
- cost: {
10228
- input: 0,
10229
- output: 0,
10230
- cacheRead: 0,
10571
+ input: 0.6799999999999999,
10572
+ output: 3.41,
10573
+ cacheRead: 0.144,
10231
10574
  cacheWrite: 0,
10232
10575
  },
10233
10576
  contextWindow: 262144,
@@ -11469,23 +11812,6 @@ export const MODELS = {
11469
11812
  contextWindow: 262144,
11470
11813
  maxTokens: 32768,
11471
11814
  },
11472
- "prime-intellect/intellect-3": {
11473
- id: "prime-intellect/intellect-3",
11474
- name: "Prime Intellect: INTELLECT-3",
11475
- api: "openai-completions",
11476
- provider: "openrouter",
11477
- baseUrl: "https://openrouter.ai/api/v1",
11478
- reasoning: true,
11479
- input: ["text"],
11480
- cost: {
11481
- input: 0.19999999999999998,
11482
- output: 1.1,
11483
- cacheRead: 0,
11484
- cacheWrite: 0,
11485
- },
11486
- contextWindow: 131072,
11487
- maxTokens: 131072,
11488
- },
11489
11815
  "qwen/qwen-2.5-72b-instruct": {
11490
11816
  id: "qwen/qwen-2.5-72b-instruct",
11491
11817
  name: "Qwen2.5 72B Instruct",
@@ -12379,13 +12705,13 @@ export const MODELS = {
12379
12705
  reasoning: true,
12380
12706
  input: ["text"],
12381
12707
  cost: {
12382
- input: 0.06599999999999999,
12383
- output: 0.26,
12384
- cacheRead: 0.029,
12708
+ input: 0.063,
12709
+ output: 0.21,
12710
+ cacheRead: 0.020999999999999998,
12385
12711
  cacheWrite: 0,
12386
12712
  },
12387
12713
  contextWindow: 262144,
12388
- maxTokens: 262144,
12714
+ maxTokens: 4096,
12389
12715
  },
12390
12716
  "thedrummer/rocinante-12b": {
12391
12717
  id: "thedrummer/rocinante-12b",
@@ -12687,11 +13013,11 @@ export const MODELS = {
12687
13013
  cost: {
12688
13014
  input: 0.98,
12689
13015
  output: 3.08,
12690
- cacheRead: 0.182,
13016
+ cacheRead: 0.49,
12691
13017
  cacheWrite: 0,
12692
13018
  },
12693
13019
  contextWindow: 202752,
12694
- maxTokens: 4096,
13020
+ maxTokens: 65535,
12695
13021
  },
12696
13022
  "z-ai/glm-5.2": {
12697
13023
  id: "z-ai/glm-5.2",
@@ -12702,12 +13028,29 @@ export const MODELS = {
12702
13028
  reasoning: true,
12703
13029
  input: ["text"],
12704
13030
  cost: {
12705
- input: 1.2,
12706
- output: 4.1,
12707
- cacheRead: 0.19999999999999998,
13031
+ input: 0.98,
13032
+ output: 3.08,
13033
+ cacheRead: 0.182,
12708
13034
  cacheWrite: 0,
12709
13035
  },
12710
13036
  contextWindow: 1048576,
13037
+ maxTokens: 4096,
13038
+ },
13039
+ "z-ai/glm-5v-turbo": {
13040
+ id: "z-ai/glm-5v-turbo",
13041
+ name: "Z.ai: GLM 5V Turbo",
13042
+ api: "openai-completions",
13043
+ provider: "openrouter",
13044
+ baseUrl: "https://openrouter.ai/api/v1",
13045
+ reasoning: true,
13046
+ input: ["text", "image"],
13047
+ cost: {
13048
+ input: 1.2,
13049
+ output: 4,
13050
+ cacheRead: 0.24,
13051
+ cacheWrite: 0,
13052
+ },
13053
+ contextWindow: 202752,
12711
13054
  maxTokens: 131072,
12712
13055
  },
12713
13056
  "~anthropic/claude-fable-latest": {
@@ -12821,9 +13164,9 @@ export const MODELS = {
12821
13164
  reasoning: true,
12822
13165
  input: ["text", "image"],
12823
13166
  cost: {
12824
- input: 0.67,
12825
- output: 3.5,
12826
- cacheRead: 0.19999999999999998,
13167
+ input: 0.66,
13168
+ output: 3.41,
13169
+ cacheRead: 0.144,
12827
13170
  cacheWrite: 0,
12828
13171
  },
12829
13172
  contextWindow: 262144,
@@ -15588,6 +15931,23 @@ export const MODELS = {
15588
15931
  contextWindow: 200000,
15589
15932
  maxTokens: 8000,
15590
15933
  },
15934
+ "sakana/fugu-ultra": {
15935
+ id: "sakana/fugu-ultra",
15936
+ name: "Fugu Ultra",
15937
+ api: "anthropic-messages",
15938
+ provider: "vercel-ai-gateway",
15939
+ baseUrl: "https://ai-gateway.vercel.sh",
15940
+ reasoning: true,
15941
+ input: ["text", "image"],
15942
+ cost: {
15943
+ input: 5,
15944
+ output: 30,
15945
+ cacheRead: 0.5,
15946
+ cacheWrite: 0,
15947
+ },
15948
+ contextWindow: 1000000,
15949
+ maxTokens: 1000000,
15950
+ },
15591
15951
  "stepfun/step-3.5-flash": {
15592
15952
  id: "stepfun/step-3.5-flash",
15593
15953
  name: "StepFun 3.5 Flash",
@@ -15599,8 +15959,8 @@ export const MODELS = {
15599
15959
  cost: {
15600
15960
  input: 0.09,
15601
15961
  output: 0.3,
15602
- cacheRead: 0,
15603
- cacheWrite: 0.02,
15962
+ cacheRead: 0.02,
15963
+ cacheWrite: 0,
15604
15964
  },
15605
15965
  contextWindow: 262114,
15606
15966
  maxTokens: 262114,