@mariozechner/pi-ai 0.50.0 → 0.50.2

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.
@@ -3279,6 +3279,260 @@ export const MODELS = {
3279
3279
  maxTokens: 16384,
3280
3280
  },
3281
3281
  },
3282
+ "huggingface": {
3283
+ "MiniMaxAI/MiniMax-M2.1": {
3284
+ id: "MiniMaxAI/MiniMax-M2.1",
3285
+ name: "MiniMax-M2.1",
3286
+ api: "openai-completions",
3287
+ provider: "huggingface",
3288
+ baseUrl: "https://router.huggingface.co/v1",
3289
+ compat: { "supportsDeveloperRole": false },
3290
+ reasoning: true,
3291
+ input: ["text"],
3292
+ cost: {
3293
+ input: 0.3,
3294
+ output: 1.2,
3295
+ cacheRead: 0,
3296
+ cacheWrite: 0,
3297
+ },
3298
+ contextWindow: 204800,
3299
+ maxTokens: 131072,
3300
+ },
3301
+ "Qwen/Qwen3-235B-A22B-Thinking-2507": {
3302
+ id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
3303
+ name: "Qwen3-235B-A22B-Thinking-2507",
3304
+ api: "openai-completions",
3305
+ provider: "huggingface",
3306
+ baseUrl: "https://router.huggingface.co/v1",
3307
+ compat: { "supportsDeveloperRole": false },
3308
+ reasoning: true,
3309
+ input: ["text"],
3310
+ cost: {
3311
+ input: 0.3,
3312
+ output: 3,
3313
+ cacheRead: 0,
3314
+ cacheWrite: 0,
3315
+ },
3316
+ contextWindow: 262144,
3317
+ maxTokens: 131072,
3318
+ },
3319
+ "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
3320
+ id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
3321
+ name: "Qwen3-Coder-480B-A35B-Instruct",
3322
+ api: "openai-completions",
3323
+ provider: "huggingface",
3324
+ baseUrl: "https://router.huggingface.co/v1",
3325
+ compat: { "supportsDeveloperRole": false },
3326
+ reasoning: false,
3327
+ input: ["text"],
3328
+ cost: {
3329
+ input: 2,
3330
+ output: 2,
3331
+ cacheRead: 0,
3332
+ cacheWrite: 0,
3333
+ },
3334
+ contextWindow: 262144,
3335
+ maxTokens: 66536,
3336
+ },
3337
+ "Qwen/Qwen3-Next-80B-A3B-Instruct": {
3338
+ id: "Qwen/Qwen3-Next-80B-A3B-Instruct",
3339
+ name: "Qwen3-Next-80B-A3B-Instruct",
3340
+ api: "openai-completions",
3341
+ provider: "huggingface",
3342
+ baseUrl: "https://router.huggingface.co/v1",
3343
+ compat: { "supportsDeveloperRole": false },
3344
+ reasoning: false,
3345
+ input: ["text"],
3346
+ cost: {
3347
+ input: 0.25,
3348
+ output: 1,
3349
+ cacheRead: 0,
3350
+ cacheWrite: 0,
3351
+ },
3352
+ contextWindow: 262144,
3353
+ maxTokens: 66536,
3354
+ },
3355
+ "Qwen/Qwen3-Next-80B-A3B-Thinking": {
3356
+ id: "Qwen/Qwen3-Next-80B-A3B-Thinking",
3357
+ name: "Qwen3-Next-80B-A3B-Thinking",
3358
+ api: "openai-completions",
3359
+ provider: "huggingface",
3360
+ baseUrl: "https://router.huggingface.co/v1",
3361
+ compat: { "supportsDeveloperRole": false },
3362
+ reasoning: false,
3363
+ input: ["text"],
3364
+ cost: {
3365
+ input: 0.3,
3366
+ output: 2,
3367
+ cacheRead: 0,
3368
+ cacheWrite: 0,
3369
+ },
3370
+ contextWindow: 262144,
3371
+ maxTokens: 131072,
3372
+ },
3373
+ "XiaomiMiMo/MiMo-V2-Flash": {
3374
+ id: "XiaomiMiMo/MiMo-V2-Flash",
3375
+ name: "MiMo-V2-Flash",
3376
+ api: "openai-completions",
3377
+ provider: "huggingface",
3378
+ baseUrl: "https://router.huggingface.co/v1",
3379
+ compat: { "supportsDeveloperRole": false },
3380
+ reasoning: true,
3381
+ input: ["text"],
3382
+ cost: {
3383
+ input: 0.1,
3384
+ output: 0.3,
3385
+ cacheRead: 0,
3386
+ cacheWrite: 0,
3387
+ },
3388
+ contextWindow: 262144,
3389
+ maxTokens: 4096,
3390
+ },
3391
+ "deepseek-ai/DeepSeek-R1-0528": {
3392
+ id: "deepseek-ai/DeepSeek-R1-0528",
3393
+ name: "DeepSeek-R1-0528",
3394
+ api: "openai-completions",
3395
+ provider: "huggingface",
3396
+ baseUrl: "https://router.huggingface.co/v1",
3397
+ compat: { "supportsDeveloperRole": false },
3398
+ reasoning: true,
3399
+ input: ["text"],
3400
+ cost: {
3401
+ input: 3,
3402
+ output: 5,
3403
+ cacheRead: 0,
3404
+ cacheWrite: 0,
3405
+ },
3406
+ contextWindow: 163840,
3407
+ maxTokens: 163840,
3408
+ },
3409
+ "deepseek-ai/DeepSeek-V3.2": {
3410
+ id: "deepseek-ai/DeepSeek-V3.2",
3411
+ name: "DeepSeek-V3.2",
3412
+ api: "openai-completions",
3413
+ provider: "huggingface",
3414
+ baseUrl: "https://router.huggingface.co/v1",
3415
+ compat: { "supportsDeveloperRole": false },
3416
+ reasoning: true,
3417
+ input: ["text"],
3418
+ cost: {
3419
+ input: 0.28,
3420
+ output: 0.4,
3421
+ cacheRead: 0,
3422
+ cacheWrite: 0,
3423
+ },
3424
+ contextWindow: 163840,
3425
+ maxTokens: 65536,
3426
+ },
3427
+ "moonshotai/Kimi-K2-Instruct": {
3428
+ id: "moonshotai/Kimi-K2-Instruct",
3429
+ name: "Kimi-K2-Instruct",
3430
+ api: "openai-completions",
3431
+ provider: "huggingface",
3432
+ baseUrl: "https://router.huggingface.co/v1",
3433
+ compat: { "supportsDeveloperRole": false },
3434
+ reasoning: false,
3435
+ input: ["text"],
3436
+ cost: {
3437
+ input: 1,
3438
+ output: 3,
3439
+ cacheRead: 0,
3440
+ cacheWrite: 0,
3441
+ },
3442
+ contextWindow: 131072,
3443
+ maxTokens: 16384,
3444
+ },
3445
+ "moonshotai/Kimi-K2-Instruct-0905": {
3446
+ id: "moonshotai/Kimi-K2-Instruct-0905",
3447
+ name: "Kimi-K2-Instruct-0905",
3448
+ api: "openai-completions",
3449
+ provider: "huggingface",
3450
+ baseUrl: "https://router.huggingface.co/v1",
3451
+ compat: { "supportsDeveloperRole": false },
3452
+ reasoning: false,
3453
+ input: ["text"],
3454
+ cost: {
3455
+ input: 1,
3456
+ output: 3,
3457
+ cacheRead: 0,
3458
+ cacheWrite: 0,
3459
+ },
3460
+ contextWindow: 262144,
3461
+ maxTokens: 16384,
3462
+ },
3463
+ "moonshotai/Kimi-K2-Thinking": {
3464
+ id: "moonshotai/Kimi-K2-Thinking",
3465
+ name: "Kimi-K2-Thinking",
3466
+ api: "openai-completions",
3467
+ provider: "huggingface",
3468
+ baseUrl: "https://router.huggingface.co/v1",
3469
+ compat: { "supportsDeveloperRole": false },
3470
+ reasoning: true,
3471
+ input: ["text"],
3472
+ cost: {
3473
+ input: 0.6,
3474
+ output: 2.5,
3475
+ cacheRead: 0.15,
3476
+ cacheWrite: 0,
3477
+ },
3478
+ contextWindow: 262144,
3479
+ maxTokens: 262144,
3480
+ },
3481
+ "moonshotai/Kimi-K2.5": {
3482
+ id: "moonshotai/Kimi-K2.5",
3483
+ name: "Kimi-K2.5",
3484
+ api: "openai-completions",
3485
+ provider: "huggingface",
3486
+ baseUrl: "https://router.huggingface.co/v1",
3487
+ compat: { "supportsDeveloperRole": false },
3488
+ reasoning: true,
3489
+ input: ["text", "image"],
3490
+ cost: {
3491
+ input: 0.6,
3492
+ output: 3,
3493
+ cacheRead: 0.1,
3494
+ cacheWrite: 0,
3495
+ },
3496
+ contextWindow: 262144,
3497
+ maxTokens: 262144,
3498
+ },
3499
+ "zai-org/GLM-4.7": {
3500
+ id: "zai-org/GLM-4.7",
3501
+ name: "GLM-4.7",
3502
+ api: "openai-completions",
3503
+ provider: "huggingface",
3504
+ baseUrl: "https://router.huggingface.co/v1",
3505
+ compat: { "supportsDeveloperRole": false },
3506
+ reasoning: true,
3507
+ input: ["text"],
3508
+ cost: {
3509
+ input: 0.6,
3510
+ output: 2.2,
3511
+ cacheRead: 0.11,
3512
+ cacheWrite: 0,
3513
+ },
3514
+ contextWindow: 204800,
3515
+ maxTokens: 131072,
3516
+ },
3517
+ "zai-org/GLM-4.7-Flash": {
3518
+ id: "zai-org/GLM-4.7-Flash",
3519
+ name: "GLM-4.7-Flash",
3520
+ api: "openai-completions",
3521
+ provider: "huggingface",
3522
+ baseUrl: "https://router.huggingface.co/v1",
3523
+ compat: { "supportsDeveloperRole": false },
3524
+ reasoning: true,
3525
+ input: ["text"],
3526
+ cost: {
3527
+ input: 0,
3528
+ output: 0,
3529
+ cacheRead: 0,
3530
+ cacheWrite: 0,
3531
+ },
3532
+ contextWindow: 200000,
3533
+ maxTokens: 128000,
3534
+ },
3535
+ },
3282
3536
  "minimax": {
3283
3537
  "MiniMax-M2": {
3284
3538
  id: "MiniMax-M2",
@@ -4633,23 +4887,6 @@ export const MODELS = {
4633
4887
  contextWindow: 204800,
4634
4888
  maxTokens: 131072,
4635
4889
  },
4636
- "glm-4.7-free": {
4637
- id: "glm-4.7-free",
4638
- name: "GLM-4.7",
4639
- api: "openai-completions",
4640
- provider: "opencode",
4641
- baseUrl: "https://opencode.ai/zen/v1",
4642
- reasoning: true,
4643
- input: ["text"],
4644
- cost: {
4645
- input: 0,
4646
- output: 0,
4647
- cacheRead: 0,
4648
- cacheWrite: 0,
4649
- },
4650
- contextWindow: 204800,
4651
- maxTokens: 131072,
4652
- },
4653
4890
  "gpt-5": {
4654
4891
  id: "gpt-5",
4655
4892
  name: "GPT-5",
@@ -4803,23 +5040,6 @@ export const MODELS = {
4803
5040
  contextWindow: 400000,
4804
5041
  maxTokens: 128000,
4805
5042
  },
4806
- "grok-code": {
4807
- id: "grok-code",
4808
- name: "Grok Code Fast 1",
4809
- api: "openai-completions",
4810
- provider: "opencode",
4811
- baseUrl: "https://opencode.ai/zen/v1",
4812
- reasoning: true,
4813
- input: ["text"],
4814
- cost: {
4815
- input: 0,
4816
- output: 0,
4817
- cacheRead: 0,
4818
- cacheWrite: 0,
4819
- },
4820
- contextWindow: 256000,
4821
- maxTokens: 256000,
4822
- },
4823
5043
  "kimi-k2": {
4824
5044
  id: "kimi-k2",
4825
5045
  name: "Kimi K2",
@@ -4854,18 +5074,35 @@ export const MODELS = {
4854
5074
  contextWindow: 262144,
4855
5075
  maxTokens: 262144,
4856
5076
  },
4857
- "minimax-m2.1-free": {
4858
- id: "minimax-m2.1-free",
5077
+ "kimi-k2.5": {
5078
+ id: "kimi-k2.5",
5079
+ name: "Kimi K2.5",
5080
+ api: "openai-completions",
5081
+ provider: "opencode",
5082
+ baseUrl: "https://opencode.ai/zen/v1",
5083
+ reasoning: true,
5084
+ input: ["text", "image"],
5085
+ cost: {
5086
+ input: 0.6,
5087
+ output: 3,
5088
+ cacheRead: 0.1,
5089
+ cacheWrite: 0,
5090
+ },
5091
+ contextWindow: 262144,
5092
+ maxTokens: 262144,
5093
+ },
5094
+ "minimax-m2.1": {
5095
+ id: "minimax-m2.1",
4859
5096
  name: "MiniMax M2.1",
4860
- api: "anthropic-messages",
5097
+ api: "openai-completions",
4861
5098
  provider: "opencode",
4862
- baseUrl: "https://opencode.ai/zen",
5099
+ baseUrl: "https://opencode.ai/zen/v1",
4863
5100
  reasoning: true,
4864
5101
  input: ["text"],
4865
5102
  cost: {
4866
- input: 0,
4867
- output: 0,
4868
- cacheRead: 0,
5103
+ input: 0.3,
5104
+ output: 1.2,
5105
+ cacheRead: 0.1,
4869
5106
  cacheWrite: 0,
4870
5107
  },
4871
5108
  contextWindow: 204800,
@@ -5071,8 +5308,8 @@ export const MODELS = {
5071
5308
  cost: {
5072
5309
  input: 0.7999999999999999,
5073
5310
  output: 4,
5074
- cacheRead: 0,
5075
- cacheWrite: 0,
5311
+ cacheRead: 0.08,
5312
+ cacheWrite: 1,
5076
5313
  },
5077
5314
  contextWindow: 200000,
5078
5315
  maxTokens: 8192,
@@ -5230,6 +5467,23 @@ export const MODELS = {
5230
5467
  contextWindow: 1000000,
5231
5468
  maxTokens: 64000,
5232
5469
  },
5470
+ "arcee-ai/trinity-large-preview:free": {
5471
+ id: "arcee-ai/trinity-large-preview:free",
5472
+ name: "Arcee AI: Trinity Large Preview (free)",
5473
+ api: "openai-completions",
5474
+ provider: "openrouter",
5475
+ baseUrl: "https://openrouter.ai/api/v1",
5476
+ reasoning: false,
5477
+ input: ["text"],
5478
+ cost: {
5479
+ input: 0,
5480
+ output: 0,
5481
+ cacheRead: 0,
5482
+ cacheWrite: 0,
5483
+ },
5484
+ contextWindow: 131000,
5485
+ maxTokens: 4096,
5486
+ },
5233
5487
  "arcee-ai/trinity-mini": {
5234
5488
  id: "arcee-ai/trinity-mini",
5235
5489
  name: "Arcee AI: Trinity Mini",
@@ -5598,7 +5852,7 @@ export const MODELS = {
5598
5852
  cost: {
5599
5853
  input: 0.21,
5600
5854
  output: 0.32,
5601
- cacheRead: 0,
5855
+ cacheRead: 0.21,
5602
5856
  cacheWrite: 0,
5603
5857
  },
5604
5858
  contextWindow: 163840,
@@ -5721,7 +5975,7 @@ export const MODELS = {
5721
5975
  cacheWrite: 0.08333333333333334,
5722
5976
  },
5723
5977
  contextWindow: 1048576,
5724
- maxTokens: 65535,
5978
+ maxTokens: 65536,
5725
5979
  },
5726
5980
  "google/gemini-2.5-pro": {
5727
5981
  id: "google/gemini-2.5-pro",
@@ -6114,23 +6368,6 @@ export const MODELS = {
6114
6368
  contextWindow: 262144,
6115
6369
  maxTokens: 65536,
6116
6370
  },
6117
- "mistralai/devstral-2512:free": {
6118
- id: "mistralai/devstral-2512:free",
6119
- name: "Mistral: Devstral 2 2512 (free)",
6120
- api: "openai-completions",
6121
- provider: "openrouter",
6122
- baseUrl: "https://openrouter.ai/api/v1",
6123
- reasoning: false,
6124
- input: ["text"],
6125
- cost: {
6126
- input: 0,
6127
- output: 0,
6128
- cacheRead: 0,
6129
- cacheWrite: 0,
6130
- },
6131
- contextWindow: 262144,
6132
- maxTokens: 4096,
6133
- },
6134
6371
  "mistralai/devstral-medium": {
6135
6372
  id: "mistralai/devstral-medium",
6136
6373
  name: "Mistral: Devstral Medium",
@@ -6641,6 +6878,23 @@ export const MODELS = {
6641
6878
  contextWindow: 262144,
6642
6879
  maxTokens: 65535,
6643
6880
  },
6881
+ "moonshotai/kimi-k2.5": {
6882
+ id: "moonshotai/kimi-k2.5",
6883
+ name: "MoonshotAI: Kimi K2.5",
6884
+ api: "openai-completions",
6885
+ provider: "openrouter",
6886
+ baseUrl: "https://openrouter.ai/api/v1",
6887
+ reasoning: true,
6888
+ input: ["text", "image"],
6889
+ cost: {
6890
+ input: 0.5700000000000001,
6891
+ output: 2.8499999999999996,
6892
+ cacheRead: 0,
6893
+ cacheWrite: 0,
6894
+ },
6895
+ contextWindow: 262144,
6896
+ maxTokens: 262144,
6897
+ },
6644
6898
  "nex-agi/deepseek-v3.1-nex-n1": {
6645
6899
  id: "nex-agi/deepseek-v3.1-nex-n1",
6646
6900
  name: "Nex AGI: DeepSeek V3.1 Nex N1",
@@ -6735,13 +6989,13 @@ export const MODELS = {
6735
6989
  reasoning: true,
6736
6990
  input: ["text"],
6737
6991
  cost: {
6738
- input: 0.06,
6739
- output: 0.24,
6992
+ input: 0.049999999999999996,
6993
+ output: 0.19999999999999998,
6740
6994
  cacheRead: 0,
6741
6995
  cacheWrite: 0,
6742
6996
  },
6743
6997
  contextWindow: 262144,
6744
- maxTokens: 262144,
6998
+ maxTokens: 4096,
6745
6999
  },
6746
7000
  "nvidia/nemotron-3-nano-30b-a3b:free": {
6747
7001
  id: "nvidia/nemotron-3-nano-30b-a3b:free",
@@ -7995,7 +8249,7 @@ export const MODELS = {
7995
8249
  cost: {
7996
8250
  input: 0.049999999999999996,
7997
8251
  output: 0.25,
7998
- cacheRead: 0,
8252
+ cacheRead: 0.049999999999999996,
7999
8253
  cacheWrite: 0,
8000
8254
  },
8001
8255
  contextWindow: 32000,
@@ -8443,6 +8697,23 @@ export const MODELS = {
8443
8697
  contextWindow: 163840,
8444
8698
  maxTokens: 65536,
8445
8699
  },
8700
+ "upstage/solar-pro-3:free": {
8701
+ id: "upstage/solar-pro-3:free",
8702
+ name: "Upstage: Solar Pro 3 (free)",
8703
+ api: "openai-completions",
8704
+ provider: "openrouter",
8705
+ baseUrl: "https://openrouter.ai/api/v1",
8706
+ reasoning: true,
8707
+ input: ["text"],
8708
+ cost: {
8709
+ input: 0,
8710
+ output: 0,
8711
+ cacheRead: 0,
8712
+ cacheWrite: 0,
8713
+ },
8714
+ contextWindow: 128000,
8715
+ maxTokens: 4096,
8716
+ },
8446
8717
  "x-ai/grok-3": {
8447
8718
  id: "x-ai/grok-3",
8448
8719
  name: "xAI: Grok 3",
@@ -8938,6 +9209,23 @@ export const MODELS = {
8938
9209
  contextWindow: 262144,
8939
9210
  maxTokens: 32768,
8940
9211
  },
9212
+ "alibaba/qwen3-max-thinking": {
9213
+ id: "alibaba/qwen3-max-thinking",
9214
+ name: "Qwen 3 Max Thinking",
9215
+ api: "anthropic-messages",
9216
+ provider: "vercel-ai-gateway",
9217
+ baseUrl: "https://ai-gateway.vercel.sh",
9218
+ reasoning: true,
9219
+ input: ["text"],
9220
+ cost: {
9221
+ input: 1.2,
9222
+ output: 6,
9223
+ cacheRead: 0.24,
9224
+ cacheWrite: 0,
9225
+ },
9226
+ contextWindow: 256000,
9227
+ maxTokens: 256000,
9228
+ },
8941
9229
  "anthropic/claude-3-haiku": {
8942
9230
  id: "anthropic/claude-3-haiku",
8943
9231
  name: "Claude 3 Haiku",
@@ -9125,6 +9413,23 @@ export const MODELS = {
9125
9413
  contextWindow: 1000000,
9126
9414
  maxTokens: 64000,
9127
9415
  },
9416
+ "arcee-ai/trinity-large-preview": {
9417
+ id: "arcee-ai/trinity-large-preview",
9418
+ name: "Trinity Large Preview",
9419
+ api: "anthropic-messages",
9420
+ provider: "vercel-ai-gateway",
9421
+ baseUrl: "https://ai-gateway.vercel.sh",
9422
+ reasoning: false,
9423
+ input: ["text"],
9424
+ cost: {
9425
+ input: 0.25,
9426
+ output: 1,
9427
+ cacheRead: 0,
9428
+ cacheWrite: 0,
9429
+ },
9430
+ contextWindow: 131000,
9431
+ maxTokens: 131000,
9432
+ },
9128
9433
  "bytedance/seed-1.6": {
9129
9434
  id: "bytedance/seed-1.6",
9130
9435
  name: "Seed 1.6",
@@ -9822,6 +10127,23 @@ export const MODELS = {
9822
10127
  contextWindow: 256000,
9823
10128
  maxTokens: 16384,
9824
10129
  },
10130
+ "moonshotai/kimi-k2.5": {
10131
+ id: "moonshotai/kimi-k2.5",
10132
+ name: "Kimi K2.5",
10133
+ api: "anthropic-messages",
10134
+ provider: "vercel-ai-gateway",
10135
+ baseUrl: "https://ai-gateway.vercel.sh",
10136
+ reasoning: true,
10137
+ input: ["text", "image"],
10138
+ cost: {
10139
+ input: 1.2,
10140
+ output: 1.2,
10141
+ cacheRead: 0.6,
10142
+ cacheWrite: 0,
10143
+ },
10144
+ contextWindow: 256000,
10145
+ maxTokens: 256000,
10146
+ },
9825
10147
  "nvidia/nemotron-nano-12b-v2-vl": {
9826
10148
  id: "nvidia/nemotron-nano-12b-v2-vl",
9827
10149
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -11346,6 +11668,24 @@ export const MODELS = {
11346
11668
  contextWindow: 204800,
11347
11669
  maxTokens: 131072,
11348
11670
  },
11671
+ "glm-4.7-flash": {
11672
+ id: "glm-4.7-flash",
11673
+ name: "GLM-4.7-Flash",
11674
+ api: "openai-completions",
11675
+ provider: "zai",
11676
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
11677
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
11678
+ reasoning: true,
11679
+ input: ["text"],
11680
+ cost: {
11681
+ input: 0,
11682
+ output: 0,
11683
+ cacheRead: 0,
11684
+ cacheWrite: 0,
11685
+ },
11686
+ contextWindow: 200000,
11687
+ maxTokens: 131072,
11688
+ },
11349
11689
  },
11350
11690
  };
11351
11691
  //# sourceMappingURL=models.generated.js.map