@caupulican/pi-ai 0.80.34 → 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.
@@ -5132,6 +5132,24 @@ export const MODELS = {
5132
5132
  },
5133
5133
  },
5134
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
+ },
5135
5153
  "MiniMaxAI/MiniMax-M2.1": {
5136
5154
  id: "MiniMaxAI/MiniMax-M2.1",
5137
5155
  name: "MiniMax-M2.1",
@@ -5186,6 +5204,42 @@ export const MODELS = {
5186
5204
  contextWindow: 204800,
5187
5205
  maxTokens: 131072,
5188
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
+ },
5189
5243
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
5190
5244
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
5191
5245
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -5204,6 +5258,42 @@ export const MODELS = {
5204
5258
  contextWindow: 262144,
5205
5259
  maxTokens: 131072,
5206
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
+ },
5207
5297
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
5208
5298
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
5209
5299
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -5276,6 +5366,60 @@ export const MODELS = {
5276
5366
  contextWindow: 262144,
5277
5367
  maxTokens: 131072,
5278
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
+ },
5279
5423
  "Qwen/Qwen3.5-397B-A17B": {
5280
5424
  id: "Qwen/Qwen3.5-397B-A17B",
5281
5425
  name: "Qwen3.5-397B-A17B",
@@ -5294,6 +5438,42 @@ export const MODELS = {
5294
5438
  contextWindow: 262144,
5295
5439
  maxTokens: 32768,
5296
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
+ },
5297
5477
  "XiaomiMiMo/MiMo-V2-Flash": {
5298
5478
  id: "XiaomiMiMo/MiMo-V2-Flash",
5299
5479
  name: "MiMo-V2-Flash",
@@ -5312,6 +5492,24 @@ export const MODELS = {
5312
5492
  contextWindow: 262144,
5313
5493
  maxTokens: 4096,
5314
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
+ },
5315
5513
  "deepseek-ai/DeepSeek-R1-0528": {
5316
5514
  id: "deepseek-ai/DeepSeek-R1-0528",
5317
5515
  name: "DeepSeek-R1-0528",
@@ -5348,6 +5546,24 @@ export const MODELS = {
5348
5546
  contextWindow: 163840,
5349
5547
  maxTokens: 65536,
5350
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
+ },
5351
5567
  "deepseek-ai/DeepSeek-V4-Pro": {
5352
5568
  id: "deepseek-ai/DeepSeek-V4-Pro",
5353
5569
  name: "DeepSeek V4 Pro",
@@ -5366,6 +5582,60 @@ export const MODELS = {
5366
5582
  contextWindow: 1048576,
5367
5583
  maxTokens: 393216,
5368
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
+ },
5369
5639
  "moonshotai/Kimi-K2-Instruct": {
5370
5640
  id: "moonshotai/Kimi-K2-Instruct",
5371
5641
  name: "Kimi-K2-Instruct",
@@ -5456,6 +5726,114 @@ export const MODELS = {
5456
5726
  contextWindow: 262144,
5457
5727
  maxTokens: 262144,
5458
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
+ },
5459
5837
  "zai-org/GLM-4.7": {
5460
5838
  id: "zai-org/GLM-4.7",
5461
5839
  name: "GLM-4.7",
@@ -5528,6 +5906,24 @@ export const MODELS = {
5528
5906
  contextWindow: 202752,
5529
5907
  maxTokens: 131072,
5530
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
+ },
5531
5927
  },
5532
5928
  "kimi-coding": {
5533
5929
  "k2p7": {
@@ -7690,6 +8086,23 @@ export const MODELS = {
7690
8086
  contextWindow: 204800,
7691
8087
  maxTokens: 131072,
7692
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
+ },
7693
8106
  "gpt-5": {
7694
8107
  id: "gpt-5",
7695
8108
  name: "GPT-5",
@@ -8502,23 +8915,6 @@ export const MODELS = {
8502
8915
  contextWindow: 200000,
8503
8916
  maxTokens: 4096,
8504
8917
  },
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
8918
  "anthropic/claude-fable-5": {
8523
8919
  id: "anthropic/claude-fable-5",
8524
8920
  name: "Anthropic: Claude Fable 5",
@@ -10156,12 +10552,12 @@ export const MODELS = {
10156
10552
  input: ["text", "image"],
10157
10553
  cost: {
10158
10554
  input: 0.66,
10159
- output: 3.5,
10160
- cacheRead: 0.33,
10555
+ output: 3.41,
10556
+ cacheRead: 0.144,
10161
10557
  cacheWrite: 0,
10162
10558
  },
10163
10559
  contextWindow: 262144,
10164
- maxTokens: 65535,
10560
+ maxTokens: 262144,
10165
10561
  },
10166
10562
  "moonshotai/kimi-k2.7-code": {
10167
10563
  id: "moonshotai/kimi-k2.7-code",
@@ -10172,26 +10568,9 @@ export const MODELS = {
10172
10568
  reasoning: true,
10173
10569
  input: ["text", "image"],
10174
10570
  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,
10571
+ input: 0.6799999999999999,
10572
+ output: 3.41,
10573
+ cacheRead: 0.144,
10195
10574
  cacheWrite: 0,
10196
10575
  },
10197
10576
  contextWindow: 262144,
@@ -11433,23 +11812,6 @@ export const MODELS = {
11433
11812
  contextWindow: 262144,
11434
11813
  maxTokens: 32768,
11435
11814
  },
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
11815
  "qwen/qwen-2.5-72b-instruct": {
11454
11816
  id: "qwen/qwen-2.5-72b-instruct",
11455
11817
  name: "Qwen2.5 72B Instruct",
@@ -12666,13 +13028,30 @@ export const MODELS = {
12666
13028
  reasoning: true,
12667
13029
  input: ["text"],
12668
13030
  cost: {
12669
- input: 1,
12670
- output: 4,
12671
- cacheRead: 0.18,
13031
+ input: 0.98,
13032
+ output: 3.08,
13033
+ cacheRead: 0.182,
12672
13034
  cacheWrite: 0,
12673
13035
  },
12674
13036
  contextWindow: 1048576,
12675
- maxTokens: 32768,
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,
13054
+ maxTokens: 131072,
12676
13055
  },
12677
13056
  "~anthropic/claude-fable-latest": {
12678
13057
  id: "~anthropic/claude-fable-latest",
@@ -12786,12 +13165,12 @@ export const MODELS = {
12786
13165
  input: ["text", "image"],
12787
13166
  cost: {
12788
13167
  input: 0.66,
12789
- output: 3.5,
12790
- cacheRead: 0.33,
13168
+ output: 3.41,
13169
+ cacheRead: 0.144,
12791
13170
  cacheWrite: 0,
12792
13171
  },
12793
13172
  contextWindow: 262144,
12794
- maxTokens: 65535,
13173
+ maxTokens: 262144,
12795
13174
  },
12796
13175
  "~openai/gpt-latest": {
12797
13176
  id: "~openai/gpt-latest",
@@ -15552,6 +15931,23 @@ export const MODELS = {
15552
15931
  contextWindow: 200000,
15553
15932
  maxTokens: 8000,
15554
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
+ },
15555
15951
  "stepfun/step-3.5-flash": {
15556
15952
  id: "stepfun/step-3.5-flash",
15557
15953
  name: "StepFun 3.5 Flash",