@mariozechner/pi-ai 0.64.0 → 0.65.1

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.
@@ -1551,23 +1551,6 @@ export const MODELS = {
1551
1551
  contextWindow: 200000,
1552
1552
  maxTokens: 64000,
1553
1553
  },
1554
- "claude-3-7-sonnet-latest": {
1555
- id: "claude-3-7-sonnet-latest",
1556
- name: "Claude Sonnet 3.7 (latest)",
1557
- api: "anthropic-messages",
1558
- provider: "anthropic",
1559
- baseUrl: "https://api.anthropic.com",
1560
- reasoning: true,
1561
- input: ["text", "image"],
1562
- cost: {
1563
- input: 3,
1564
- output: 15,
1565
- cacheRead: 0.3,
1566
- cacheWrite: 3.75,
1567
- },
1568
- contextWindow: 200000,
1569
- maxTokens: 64000,
1570
- },
1571
1554
  "claude-3-haiku-20240307": {
1572
1555
  id: "claude-3-haiku-20240307",
1573
1556
  name: "Claude Haiku 3",
@@ -2301,6 +2284,23 @@ export const MODELS = {
2301
2284
  contextWindow: 400000,
2302
2285
  maxTokens: 128000,
2303
2286
  },
2287
+ "gpt-5.3-chat-latest": {
2288
+ id: "gpt-5.3-chat-latest",
2289
+ name: "GPT-5.3 Chat (latest)",
2290
+ api: "azure-openai-responses",
2291
+ provider: "azure-openai-responses",
2292
+ baseUrl: "",
2293
+ reasoning: false,
2294
+ input: ["text", "image"],
2295
+ cost: {
2296
+ input: 1.75,
2297
+ output: 14,
2298
+ cacheRead: 0.175,
2299
+ cacheWrite: 0,
2300
+ },
2301
+ contextWindow: 128000,
2302
+ maxTokens: 16384,
2303
+ },
2304
2304
  "gpt-5.3-codex": {
2305
2305
  id: "gpt-5.3-codex",
2306
2306
  name: "GPT-5.3 Codex",
@@ -3477,6 +3477,40 @@ export const MODELS = {
3477
3477
  contextWindow: 131072,
3478
3478
  maxTokens: 8192,
3479
3479
  },
3480
+ "gemma-4-26b": {
3481
+ id: "gemma-4-26b",
3482
+ name: "Gemma 4 26B",
3483
+ api: "google-generative-ai",
3484
+ provider: "google",
3485
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
3486
+ reasoning: true,
3487
+ input: ["text", "image"],
3488
+ cost: {
3489
+ input: 0,
3490
+ output: 0,
3491
+ cacheRead: 0,
3492
+ cacheWrite: 0,
3493
+ },
3494
+ contextWindow: 256000,
3495
+ maxTokens: 8192,
3496
+ },
3497
+ "gemma-4-31b": {
3498
+ id: "gemma-4-31b",
3499
+ name: "Gemma 4 31B",
3500
+ api: "google-generative-ai",
3501
+ provider: "google",
3502
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
3503
+ reasoning: true,
3504
+ input: ["text", "image"],
3505
+ cost: {
3506
+ input: 0,
3507
+ output: 0,
3508
+ cacheRead: 0,
3509
+ cacheWrite: 0,
3510
+ },
3511
+ contextWindow: 256000,
3512
+ maxTokens: 8192,
3513
+ },
3480
3514
  },
3481
3515
  "google-antigravity": {
3482
3516
  "claude-opus-4-5-thinking": {
@@ -5026,22 +5060,39 @@ export const MODELS = {
5026
5060
  contextWindow: 128000,
5027
5061
  maxTokens: 16384,
5028
5062
  },
5063
+ "mistral-small-2603": {
5064
+ id: "mistral-small-2603",
5065
+ name: "Mistral Small 4",
5066
+ api: "mistral-conversations",
5067
+ provider: "mistral",
5068
+ baseUrl: "https://api.mistral.ai",
5069
+ reasoning: true,
5070
+ input: ["text", "image"],
5071
+ cost: {
5072
+ input: 0.15,
5073
+ output: 0.6,
5074
+ cacheRead: 0,
5075
+ cacheWrite: 0,
5076
+ },
5077
+ contextWindow: 256000,
5078
+ maxTokens: 256000,
5079
+ },
5029
5080
  "mistral-small-latest": {
5030
5081
  id: "mistral-small-latest",
5031
5082
  name: "Mistral Small (latest)",
5032
5083
  api: "mistral-conversations",
5033
5084
  provider: "mistral",
5034
5085
  baseUrl: "https://api.mistral.ai",
5035
- reasoning: false,
5086
+ reasoning: true,
5036
5087
  input: ["text", "image"],
5037
5088
  cost: {
5038
- input: 0.1,
5039
- output: 0.3,
5089
+ input: 0.15,
5090
+ output: 0.6,
5040
5091
  cacheRead: 0,
5041
5092
  cacheWrite: 0,
5042
5093
  },
5043
- contextWindow: 128000,
5044
- maxTokens: 16384,
5094
+ contextWindow: 256000,
5095
+ maxTokens: 256000,
5045
5096
  },
5046
5097
  "open-mistral-7b": {
5047
5098
  id: "open-mistral-7b",
@@ -5572,6 +5623,23 @@ export const MODELS = {
5572
5623
  contextWindow: 400000,
5573
5624
  maxTokens: 128000,
5574
5625
  },
5626
+ "gpt-5.3-chat-latest": {
5627
+ id: "gpt-5.3-chat-latest",
5628
+ name: "GPT-5.3 Chat (latest)",
5629
+ api: "openai-responses",
5630
+ provider: "openai",
5631
+ baseUrl: "https://api.openai.com/v1",
5632
+ reasoning: false,
5633
+ input: ["text", "image"],
5634
+ cost: {
5635
+ input: 1.75,
5636
+ output: 14,
5637
+ cacheRead: 0.175,
5638
+ cacheWrite: 0,
5639
+ },
5640
+ contextWindow: 128000,
5641
+ maxTokens: 16384,
5642
+ },
5575
5643
  "gpt-5.3-codex": {
5576
5644
  id: "gpt-5.3-codex",
5577
5645
  name: "GPT-5.3 Codex",
@@ -6426,40 +6494,6 @@ export const MODELS = {
6426
6494
  contextWindow: 262144,
6427
6495
  maxTokens: 65536,
6428
6496
  },
6429
- "mimo-v2-omni-free": {
6430
- id: "mimo-v2-omni-free",
6431
- name: "MiMo V2 Omni Free",
6432
- api: "openai-completions",
6433
- provider: "opencode",
6434
- baseUrl: "https://opencode.ai/zen/v1",
6435
- reasoning: true,
6436
- input: ["text", "image"],
6437
- cost: {
6438
- input: 0,
6439
- output: 0,
6440
- cacheRead: 0,
6441
- cacheWrite: 0,
6442
- },
6443
- contextWindow: 262144,
6444
- maxTokens: 64000,
6445
- },
6446
- "mimo-v2-pro-free": {
6447
- id: "mimo-v2-pro-free",
6448
- name: "MiMo V2 Pro Free",
6449
- api: "openai-completions",
6450
- provider: "opencode",
6451
- baseUrl: "https://opencode.ai/zen/v1",
6452
- reasoning: true,
6453
- input: ["text"],
6454
- cost: {
6455
- input: 0,
6456
- output: 0,
6457
- cacheRead: 0,
6458
- cacheWrite: 0,
6459
- },
6460
- contextWindow: 1048576,
6461
- maxTokens: 64000,
6462
- },
6463
6497
  "minimax-m2.5": {
6464
6498
  id: "minimax-m2.5",
6465
6499
  name: "MiniMax M2.5",
@@ -6508,9 +6542,26 @@ export const MODELS = {
6508
6542
  cacheRead: 0,
6509
6543
  cacheWrite: 0,
6510
6544
  },
6511
- contextWindow: 1000000,
6545
+ contextWindow: 204800,
6512
6546
  maxTokens: 128000,
6513
6547
  },
6548
+ "qwen3.6-plus-free": {
6549
+ id: "qwen3.6-plus-free",
6550
+ name: "Qwen3.6 Plus Free",
6551
+ api: "openai-completions",
6552
+ provider: "opencode",
6553
+ baseUrl: "https://opencode.ai/zen/v1",
6554
+ reasoning: true,
6555
+ input: ["text"],
6556
+ cost: {
6557
+ input: 0,
6558
+ output: 0,
6559
+ cacheRead: 0,
6560
+ cacheWrite: 0,
6561
+ },
6562
+ contextWindow: 1048576,
6563
+ maxTokens: 64000,
6564
+ },
6514
6565
  },
6515
6566
  "opencode-go": {
6516
6567
  "glm-5": {
@@ -6547,12 +6598,46 @@ export const MODELS = {
6547
6598
  contextWindow: 262144,
6548
6599
  maxTokens: 65536,
6549
6600
  },
6601
+ "mimo-v2-omni": {
6602
+ id: "mimo-v2-omni",
6603
+ name: "MiMo V2 Omni",
6604
+ api: "openai-completions",
6605
+ provider: "opencode-go",
6606
+ baseUrl: "https://opencode.ai/zen/go/v1",
6607
+ reasoning: true,
6608
+ input: ["text", "image"],
6609
+ cost: {
6610
+ input: 0.4,
6611
+ output: 2,
6612
+ cacheRead: 0.08,
6613
+ cacheWrite: 0,
6614
+ },
6615
+ contextWindow: 262144,
6616
+ maxTokens: 64000,
6617
+ },
6618
+ "mimo-v2-pro": {
6619
+ id: "mimo-v2-pro",
6620
+ name: "MiMo V2 Pro",
6621
+ api: "openai-completions",
6622
+ provider: "opencode-go",
6623
+ baseUrl: "https://opencode.ai/zen/go/v1",
6624
+ reasoning: true,
6625
+ input: ["text"],
6626
+ cost: {
6627
+ input: 1,
6628
+ output: 3,
6629
+ cacheRead: 0.2,
6630
+ cacheWrite: 0,
6631
+ },
6632
+ contextWindow: 1048576,
6633
+ maxTokens: 64000,
6634
+ },
6550
6635
  "minimax-m2.5": {
6551
6636
  id: "minimax-m2.5",
6552
6637
  name: "MiniMax M2.5",
6553
- api: "anthropic-messages",
6638
+ api: "openai-completions",
6554
6639
  provider: "opencode-go",
6555
- baseUrl: "https://opencode.ai/zen/go",
6640
+ baseUrl: "https://opencode.ai/zen/go/v1",
6556
6641
  reasoning: true,
6557
6642
  input: ["text"],
6558
6643
  cost: {
@@ -6753,23 +6838,6 @@ export const MODELS = {
6753
6838
  contextWindow: 200000,
6754
6839
  maxTokens: 8192,
6755
6840
  },
6756
- "anthropic/claude-3.5-sonnet": {
6757
- id: "anthropic/claude-3.5-sonnet",
6758
- name: "Anthropic: Claude 3.5 Sonnet",
6759
- api: "openai-completions",
6760
- provider: "openrouter",
6761
- baseUrl: "https://openrouter.ai/api/v1",
6762
- reasoning: false,
6763
- input: ["text", "image"],
6764
- cost: {
6765
- input: 6,
6766
- output: 30,
6767
- cacheRead: 0.6,
6768
- cacheWrite: 7.5,
6769
- },
6770
- contextWindow: 200000,
6771
- maxTokens: 8192,
6772
- },
6773
6841
  "anthropic/claude-3.7-sonnet": {
6774
6842
  id: "anthropic/claude-3.7-sonnet",
6775
6843
  name: "Anthropic: Claude 3.7 Sonnet",
@@ -6957,6 +7025,23 @@ export const MODELS = {
6957
7025
  contextWindow: 131000,
6958
7026
  maxTokens: 4096,
6959
7027
  },
7028
+ "arcee-ai/trinity-large-thinking": {
7029
+ id: "arcee-ai/trinity-large-thinking",
7030
+ name: "Arcee AI: Trinity Large Thinking",
7031
+ api: "openai-completions",
7032
+ provider: "openrouter",
7033
+ baseUrl: "https://openrouter.ai/api/v1",
7034
+ reasoning: true,
7035
+ input: ["text"],
7036
+ cost: {
7037
+ input: 0.22,
7038
+ output: 0.85,
7039
+ cacheRead: 0,
7040
+ cacheWrite: 0,
7041
+ },
7042
+ contextWindow: 262144,
7043
+ maxTokens: 262144,
7044
+ },
6960
7045
  "arcee-ai/trinity-mini": {
6961
7046
  id: "arcee-ai/trinity-mini",
6962
7047
  name: "Arcee AI: Trinity Mini",
@@ -7397,7 +7482,7 @@ export const MODELS = {
7397
7482
  cacheWrite: 0.08333333333333334,
7398
7483
  },
7399
7484
  contextWindow: 1048576,
7400
- maxTokens: 65536,
7485
+ maxTokens: 65535,
7401
7486
  },
7402
7487
  "google/gemini-2.5-pro": {
7403
7488
  id: "google/gemini-2.5-pro",
@@ -7518,6 +7603,40 @@ export const MODELS = {
7518
7603
  contextWindow: 1048576,
7519
7604
  maxTokens: 65536,
7520
7605
  },
7606
+ "google/gemma-4-26b-a4b-it": {
7607
+ id: "google/gemma-4-26b-a4b-it",
7608
+ name: "Google: Gemma 4 26B A4B ",
7609
+ api: "openai-completions",
7610
+ provider: "openrouter",
7611
+ baseUrl: "https://openrouter.ai/api/v1",
7612
+ reasoning: true,
7613
+ input: ["text", "image"],
7614
+ cost: {
7615
+ input: 0.13,
7616
+ output: 0.39999999999999997,
7617
+ cacheRead: 0,
7618
+ cacheWrite: 0,
7619
+ },
7620
+ contextWindow: 262144,
7621
+ maxTokens: 262144,
7622
+ },
7623
+ "google/gemma-4-31b-it": {
7624
+ id: "google/gemma-4-31b-it",
7625
+ name: "Google: Gemma 4 31B",
7626
+ api: "openai-completions",
7627
+ provider: "openrouter",
7628
+ baseUrl: "https://openrouter.ai/api/v1",
7629
+ reasoning: true,
7630
+ input: ["text", "image"],
7631
+ cost: {
7632
+ input: 0.14,
7633
+ output: 0.39999999999999997,
7634
+ cacheRead: 0,
7635
+ cacheWrite: 0,
7636
+ },
7637
+ contextWindow: 262144,
7638
+ maxTokens: 131072,
7639
+ },
7521
7640
  "inception/mercury": {
7522
7641
  id: "inception/mercury",
7523
7642
  name: "Inception: Mercury",
@@ -7569,23 +7688,6 @@ export const MODELS = {
7569
7688
  contextWindow: 128000,
7570
7689
  maxTokens: 32000,
7571
7690
  },
7572
- "kwaipilot/kat-coder-pro": {
7573
- id: "kwaipilot/kat-coder-pro",
7574
- name: "Kwaipilot: KAT-Coder-Pro V1",
7575
- api: "openai-completions",
7576
- provider: "openrouter",
7577
- baseUrl: "https://openrouter.ai/api/v1",
7578
- reasoning: false,
7579
- input: ["text"],
7580
- cost: {
7581
- input: 0.207,
7582
- output: 0.828,
7583
- cacheRead: 0.0414,
7584
- cacheWrite: 0,
7585
- },
7586
- contextWindow: 256000,
7587
- maxTokens: 128000,
7588
- },
7589
7691
  "kwaipilot/kat-coder-pro-v2": {
7590
7692
  id: "kwaipilot/kat-coder-pro-v2",
7591
7693
  name: "Kwaipilot: KAT-Coder-Pro V2",
@@ -7799,9 +7901,9 @@ export const MODELS = {
7799
7901
  reasoning: true,
7800
7902
  input: ["text"],
7801
7903
  cost: {
7802
- input: 0.19,
7803
- output: 1.15,
7804
- cacheRead: 0.095,
7904
+ input: 0.118,
7905
+ output: 0.9900000000000001,
7906
+ cacheRead: 0.059,
7805
7907
  cacheWrite: 0,
7806
7908
  },
7807
7909
  contextWindow: 196608,
@@ -8096,23 +8198,6 @@ export const MODELS = {
8096
8198
  contextWindow: 32768,
8097
8199
  maxTokens: 4096,
8098
8200
  },
8099
- "mistralai/mistral-small-24b-instruct-2501": {
8100
- id: "mistralai/mistral-small-24b-instruct-2501",
8101
- name: "Mistral: Mistral Small 3",
8102
- api: "openai-completions",
8103
- provider: "openrouter",
8104
- baseUrl: "https://openrouter.ai/api/v1",
8105
- reasoning: false,
8106
- input: ["text"],
8107
- cost: {
8108
- input: 0.049999999999999996,
8109
- output: 0.08,
8110
- cacheRead: 0,
8111
- cacheWrite: 0,
8112
- },
8113
- contextWindow: 32768,
8114
- maxTokens: 16384,
8115
- },
8116
8201
  "mistralai/mistral-small-2603": {
8117
8202
  id: "mistralai/mistral-small-2603",
8118
8203
  name: "Mistral: Mistral Small 4",
@@ -9167,13 +9252,47 @@ export const MODELS = {
9167
9252
  contextWindow: 1050000,
9168
9253
  maxTokens: 128000,
9169
9254
  },
9170
- "openai/gpt-oss-120b": {
9171
- id: "openai/gpt-oss-120b",
9172
- name: "OpenAI: gpt-oss-120b",
9255
+ "openai/gpt-audio": {
9256
+ id: "openai/gpt-audio",
9257
+ name: "OpenAI: GPT Audio",
9173
9258
  api: "openai-completions",
9174
9259
  provider: "openrouter",
9175
9260
  baseUrl: "https://openrouter.ai/api/v1",
9176
- reasoning: true,
9261
+ reasoning: false,
9262
+ input: ["text"],
9263
+ cost: {
9264
+ input: 2.5,
9265
+ output: 10,
9266
+ cacheRead: 0,
9267
+ cacheWrite: 0,
9268
+ },
9269
+ contextWindow: 128000,
9270
+ maxTokens: 16384,
9271
+ },
9272
+ "openai/gpt-audio-mini": {
9273
+ id: "openai/gpt-audio-mini",
9274
+ name: "OpenAI: GPT Audio Mini",
9275
+ api: "openai-completions",
9276
+ provider: "openrouter",
9277
+ baseUrl: "https://openrouter.ai/api/v1",
9278
+ reasoning: false,
9279
+ input: ["text"],
9280
+ cost: {
9281
+ input: 0.6,
9282
+ output: 2.4,
9283
+ cacheRead: 0,
9284
+ cacheWrite: 0,
9285
+ },
9286
+ contextWindow: 128000,
9287
+ maxTokens: 16384,
9288
+ },
9289
+ "openai/gpt-oss-120b": {
9290
+ id: "openai/gpt-oss-120b",
9291
+ name: "OpenAI: gpt-oss-120b",
9292
+ api: "openai-completions",
9293
+ provider: "openrouter",
9294
+ baseUrl: "https://openrouter.ai/api/v1",
9295
+ reasoning: true,
9177
9296
  input: ["text"],
9178
9297
  cost: {
9179
9298
  input: 0.039,
@@ -10134,7 +10253,7 @@ export const MODELS = {
10134
10253
  cacheWrite: 0,
10135
10254
  },
10136
10255
  contextWindow: 256000,
10137
- maxTokens: 65536,
10256
+ maxTokens: 32768,
10138
10257
  },
10139
10258
  "qwen/qwen3.5-flash-02-23": {
10140
10259
  id: "qwen/qwen3.5-flash-02-23",
@@ -10170,6 +10289,23 @@ export const MODELS = {
10170
10289
  contextWindow: 1000000,
10171
10290
  maxTokens: 65536,
10172
10291
  },
10292
+ "qwen/qwen3.6-plus:free": {
10293
+ id: "qwen/qwen3.6-plus:free",
10294
+ name: "Qwen: Qwen3.6 Plus (free)",
10295
+ api: "openai-completions",
10296
+ provider: "openrouter",
10297
+ baseUrl: "https://openrouter.ai/api/v1",
10298
+ reasoning: true,
10299
+ input: ["text", "image"],
10300
+ cost: {
10301
+ input: 0,
10302
+ output: 0,
10303
+ cacheRead: 0,
10304
+ cacheWrite: 0,
10305
+ },
10306
+ contextWindow: 1000000,
10307
+ maxTokens: 65536,
10308
+ },
10173
10309
  "qwen/qwq-32b": {
10174
10310
  id: "qwen/qwq-32b",
10175
10311
  name: "Qwen: QwQ 32B",
@@ -10187,8 +10323,8 @@ export const MODELS = {
10187
10323
  contextWindow: 131072,
10188
10324
  maxTokens: 131072,
10189
10325
  },
10190
- "reka/reka-edge": {
10191
- id: "reka/reka-edge",
10326
+ "rekaai/reka-edge": {
10327
+ id: "rekaai/reka-edge",
10192
10328
  name: "Reka Edge",
10193
10329
  api: "openai-completions",
10194
10330
  provider: "openrouter",
@@ -10266,11 +10402,11 @@ export const MODELS = {
10266
10402
  cost: {
10267
10403
  input: 0.09999999999999999,
10268
10404
  output: 0.3,
10269
- cacheRead: 0.02,
10405
+ cacheRead: 0,
10270
10406
  cacheWrite: 0,
10271
10407
  },
10272
10408
  contextWindow: 262144,
10273
- maxTokens: 4096,
10409
+ maxTokens: 65536,
10274
10410
  },
10275
10411
  "stepfun/step-3.5-flash:free": {
10276
10412
  id: "stepfun/step-3.5-flash:free",
@@ -10476,9 +10612,9 @@ export const MODELS = {
10476
10612
  contextWindow: 2000000,
10477
10613
  maxTokens: 30000,
10478
10614
  },
10479
- "x-ai/grok-4.20-beta": {
10480
- id: "x-ai/grok-4.20-beta",
10481
- name: "xAI: Grok 4.20 Beta",
10615
+ "x-ai/grok-4.20": {
10616
+ id: "x-ai/grok-4.20",
10617
+ name: "xAI: Grok 4.20",
10482
10618
  api: "openai-completions",
10483
10619
  provider: "openrouter",
10484
10620
  baseUrl: "https://openrouter.ai/api/v1",
@@ -10748,6 +10884,23 @@ export const MODELS = {
10748
10884
  contextWindow: 202752,
10749
10885
  maxTokens: 131072,
10750
10886
  },
10887
+ "z-ai/glm-5v-turbo": {
10888
+ id: "z-ai/glm-5v-turbo",
10889
+ name: "Z.ai: GLM 5V Turbo",
10890
+ api: "openai-completions",
10891
+ provider: "openrouter",
10892
+ baseUrl: "https://openrouter.ai/api/v1",
10893
+ reasoning: true,
10894
+ input: ["text", "image"],
10895
+ cost: {
10896
+ input: 1.2,
10897
+ output: 4,
10898
+ cacheRead: 0.24,
10899
+ cacheWrite: 0,
10900
+ },
10901
+ contextWindow: 202752,
10902
+ maxTokens: 131072,
10903
+ },
10751
10904
  },
10752
10905
  "vercel-ai-gateway": {
10753
10906
  "alibaba/qwen-3-14b": {
@@ -10769,19 +10922,19 @@ export const MODELS = {
10769
10922
  },
10770
10923
  "alibaba/qwen-3-235b": {
10771
10924
  id: "alibaba/qwen-3-235b",
10772
- name: "Qwen3-235B-A22B",
10925
+ name: "Qwen3 235B A22b Instruct 2507",
10773
10926
  api: "anthropic-messages",
10774
10927
  provider: "vercel-ai-gateway",
10775
10928
  baseUrl: "https://ai-gateway.vercel.sh",
10776
10929
  reasoning: false,
10777
10930
  input: ["text"],
10778
10931
  cost: {
10779
- input: 0.071,
10780
- output: 0.463,
10781
- cacheRead: 0,
10932
+ input: 0.22,
10933
+ output: 0.88,
10934
+ cacheRead: 0.11,
10782
10935
  cacheWrite: 0,
10783
10936
  },
10784
- contextWindow: 40960,
10937
+ contextWindow: 32768,
10785
10938
  maxTokens: 16384,
10786
10939
  },
10787
10940
  "alibaba/qwen-3-30b": {
@@ -10810,13 +10963,13 @@ export const MODELS = {
10810
10963
  reasoning: true,
10811
10964
  input: ["text"],
10812
10965
  cost: {
10813
- input: 0.29,
10814
- output: 0.59,
10815
- cacheRead: 0.145,
10966
+ input: 0.16,
10967
+ output: 0.64,
10968
+ cacheRead: 0,
10816
10969
  cacheWrite: 0,
10817
10970
  },
10818
- contextWindow: 131072,
10819
- maxTokens: 40960,
10971
+ contextWindow: 128000,
10972
+ maxTokens: 8192,
10820
10973
  },
10821
10974
  "alibaba/qwen3-235b-a22b-thinking": {
10822
10975
  id: "alibaba/qwen3-235b-a22b-thinking",
@@ -10844,13 +10997,13 @@ export const MODELS = {
10844
10997
  reasoning: false,
10845
10998
  input: ["text"],
10846
10999
  cost: {
10847
- input: 0.39999999999999997,
10848
- output: 1.5999999999999999,
10849
- cacheRead: 0.022,
11000
+ input: 1.5,
11001
+ output: 7.5,
11002
+ cacheRead: 0.3,
10850
11003
  cacheWrite: 0,
10851
11004
  },
10852
11005
  contextWindow: 262144,
10853
- maxTokens: 66536,
11006
+ maxTokens: 65536,
10854
11007
  },
10855
11008
  "alibaba/qwen3-coder-30b-a3b": {
10856
11009
  id: "alibaba/qwen3-coder-30b-a3b",
@@ -10963,13 +11116,13 @@ export const MODELS = {
10963
11116
  reasoning: true,
10964
11117
  input: ["text", "image"],
10965
11118
  cost: {
10966
- input: 0.22,
10967
- output: 0.88,
11119
+ input: 0.39999999999999997,
11120
+ output: 4,
10968
11121
  cacheRead: 0,
10969
11122
  cacheWrite: 0,
10970
11123
  },
10971
- contextWindow: 256000,
10972
- maxTokens: 256000,
11124
+ contextWindow: 131072,
11125
+ maxTokens: 32768,
10973
11126
  },
10974
11127
  "alibaba/qwen3.5-flash": {
10975
11128
  id: "alibaba/qwen3.5-flash",
@@ -11005,6 +11158,23 @@ export const MODELS = {
11005
11158
  contextWindow: 1000000,
11006
11159
  maxTokens: 64000,
11007
11160
  },
11161
+ "alibaba/qwen3.6-plus": {
11162
+ id: "alibaba/qwen3.6-plus",
11163
+ name: "Qwen 3.6 Plus",
11164
+ api: "anthropic-messages",
11165
+ provider: "vercel-ai-gateway",
11166
+ baseUrl: "https://ai-gateway.vercel.sh",
11167
+ reasoning: true,
11168
+ input: ["text", "image"],
11169
+ cost: {
11170
+ input: 0.5,
11171
+ output: 3,
11172
+ cacheRead: 0.09999999999999999,
11173
+ cacheWrite: 0.625,
11174
+ },
11175
+ contextWindow: 1000000,
11176
+ maxTokens: 64000,
11177
+ },
11008
11178
  "anthropic/claude-3-haiku": {
11009
11179
  id: "anthropic/claude-3-haiku",
11010
11180
  name: "Claude 3 Haiku",
@@ -11039,40 +11209,6 @@ export const MODELS = {
11039
11209
  contextWindow: 200000,
11040
11210
  maxTokens: 8192,
11041
11211
  },
11042
- "anthropic/claude-3.5-sonnet": {
11043
- id: "anthropic/claude-3.5-sonnet",
11044
- name: "Claude 3.5 Sonnet",
11045
- api: "anthropic-messages",
11046
- provider: "vercel-ai-gateway",
11047
- baseUrl: "https://ai-gateway.vercel.sh",
11048
- reasoning: false,
11049
- input: ["text", "image"],
11050
- cost: {
11051
- input: 3,
11052
- output: 15,
11053
- cacheRead: 0.3,
11054
- cacheWrite: 3.75,
11055
- },
11056
- contextWindow: 200000,
11057
- maxTokens: 8192,
11058
- },
11059
- "anthropic/claude-3.5-sonnet-20240620": {
11060
- id: "anthropic/claude-3.5-sonnet-20240620",
11061
- name: "Claude 3.5 Sonnet (2024-06-20)",
11062
- api: "anthropic-messages",
11063
- provider: "vercel-ai-gateway",
11064
- baseUrl: "https://ai-gateway.vercel.sh",
11065
- reasoning: false,
11066
- input: ["text", "image"],
11067
- cost: {
11068
- input: 3,
11069
- output: 15,
11070
- cacheRead: 0.3,
11071
- cacheWrite: 3.75,
11072
- },
11073
- contextWindow: 200000,
11074
- maxTokens: 8192,
11075
- },
11076
11212
  "anthropic/claude-3.7-sonnet": {
11077
11213
  id: "anthropic/claude-3.7-sonnet",
11078
11214
  name: "Claude 3.7 Sonnet",
@@ -11243,6 +11379,23 @@ export const MODELS = {
11243
11379
  contextWindow: 131000,
11244
11380
  maxTokens: 131000,
11245
11381
  },
11382
+ "arcee-ai/trinity-large-thinking": {
11383
+ id: "arcee-ai/trinity-large-thinking",
11384
+ name: "Trinity Large Thinking",
11385
+ api: "anthropic-messages",
11386
+ provider: "vercel-ai-gateway",
11387
+ baseUrl: "https://ai-gateway.vercel.sh",
11388
+ reasoning: true,
11389
+ input: ["text"],
11390
+ cost: {
11391
+ input: 0.25,
11392
+ output: 0.8999999999999999,
11393
+ cacheRead: 0,
11394
+ cacheWrite: 0,
11395
+ },
11396
+ contextWindow: 262100,
11397
+ maxTokens: 80000,
11398
+ },
11246
11399
  "bytedance/seed-1.6": {
11247
11400
  id: "bytedance/seed-1.6",
11248
11401
  name: "Seed 1.6",
@@ -11320,13 +11473,13 @@ export const MODELS = {
11320
11473
  reasoning: true,
11321
11474
  input: ["text"],
11322
11475
  cost: {
11323
- input: 0.5,
11324
- output: 1.5,
11325
- cacheRead: 0,
11476
+ input: 0.56,
11477
+ output: 1.68,
11478
+ cacheRead: 0.28,
11326
11479
  cacheWrite: 0,
11327
11480
  },
11328
11481
  contextWindow: 163840,
11329
- maxTokens: 16384,
11482
+ maxTokens: 8192,
11330
11483
  },
11331
11484
  "deepseek/deepseek-v3.1-terminus": {
11332
11485
  id: "deepseek/deepseek-v3.1-terminus",
@@ -11532,6 +11685,40 @@ export const MODELS = {
11532
11685
  contextWindow: 1000000,
11533
11686
  maxTokens: 64000,
11534
11687
  },
11688
+ "google/gemma-4-26b-a4b-it": {
11689
+ id: "google/gemma-4-26b-a4b-it",
11690
+ name: "Gemma 4 26B A4B IT",
11691
+ api: "anthropic-messages",
11692
+ provider: "vercel-ai-gateway",
11693
+ baseUrl: "https://ai-gateway.vercel.sh",
11694
+ reasoning: false,
11695
+ input: ["text", "image"],
11696
+ cost: {
11697
+ input: 0.13,
11698
+ output: 0.39999999999999997,
11699
+ cacheRead: 0,
11700
+ cacheWrite: 0,
11701
+ },
11702
+ contextWindow: 262144,
11703
+ maxTokens: 131072,
11704
+ },
11705
+ "google/gemma-4-31b-it": {
11706
+ id: "google/gemma-4-31b-it",
11707
+ name: "Gemma 4 31B IT",
11708
+ api: "anthropic-messages",
11709
+ provider: "vercel-ai-gateway",
11710
+ baseUrl: "https://ai-gateway.vercel.sh",
11711
+ reasoning: false,
11712
+ input: ["text", "image"],
11713
+ cost: {
11714
+ input: 0.14,
11715
+ output: 0.39999999999999997,
11716
+ cacheRead: 0,
11717
+ cacheWrite: 0,
11718
+ },
11719
+ contextWindow: 262144,
11720
+ maxTokens: 131072,
11721
+ },
11535
11722
  "inception/mercury-2": {
11536
11723
  id: "inception/mercury-2",
11537
11724
  name: "Mercury 2",
@@ -11598,23 +11785,6 @@ export const MODELS = {
11598
11785
  cacheWrite: 0,
11599
11786
  },
11600
11787
  contextWindow: 128000,
11601
- maxTokens: 100000,
11602
- },
11603
- "meituan/longcat-flash-thinking": {
11604
- id: "meituan/longcat-flash-thinking",
11605
- name: "LongCat Flash Thinking",
11606
- api: "anthropic-messages",
11607
- provider: "vercel-ai-gateway",
11608
- baseUrl: "https://ai-gateway.vercel.sh",
11609
- reasoning: true,
11610
- input: ["text"],
11611
- cost: {
11612
- input: 0.15,
11613
- output: 1.5,
11614
- cacheRead: 0,
11615
- cacheWrite: 0,
11616
- },
11617
- contextWindow: 128000,
11618
11788
  maxTokens: 8192,
11619
11789
  },
11620
11790
  "meta/llama-3.1-70b": {
@@ -11643,13 +11813,13 @@ export const MODELS = {
11643
11813
  reasoning: false,
11644
11814
  input: ["text"],
11645
11815
  cost: {
11646
- input: 0.09999999999999999,
11647
- output: 0.09999999999999999,
11648
- cacheRead: 0.09999999999999999,
11816
+ input: 0.22,
11817
+ output: 0.22,
11818
+ cacheRead: 0,
11649
11819
  cacheWrite: 0,
11650
11820
  },
11651
11821
  contextWindow: 128000,
11652
- maxTokens: 16384,
11822
+ maxTokens: 8192,
11653
11823
  },
11654
11824
  "meta/llama-3.2-11b": {
11655
11825
  id: "meta/llama-3.2-11b",
@@ -12027,20 +12197,20 @@ export const MODELS = {
12027
12197
  },
12028
12198
  "moonshotai/kimi-k2": {
12029
12199
  id: "moonshotai/kimi-k2",
12030
- name: "Kimi K2",
12200
+ name: "Kimi K2 Instruct",
12031
12201
  api: "anthropic-messages",
12032
12202
  provider: "vercel-ai-gateway",
12033
12203
  baseUrl: "https://ai-gateway.vercel.sh",
12034
12204
  reasoning: false,
12035
12205
  input: ["text"],
12036
12206
  cost: {
12037
- input: 0.6,
12038
- output: 2.5,
12039
- cacheRead: 0.15,
12207
+ input: 0.5700000000000001,
12208
+ output: 2.3,
12209
+ cacheRead: 0,
12040
12210
  cacheWrite: 0,
12041
12211
  },
12042
12212
  contextWindow: 131072,
12043
- maxTokens: 16384,
12213
+ maxTokens: 131072,
12044
12214
  },
12045
12215
  "moonshotai/kimi-k2-0905": {
12046
12216
  id: "moonshotai/kimi-k2-0905",
@@ -12053,11 +12223,11 @@ export const MODELS = {
12053
12223
  cost: {
12054
12224
  input: 0.6,
12055
12225
  output: 2.5,
12056
- cacheRead: 0.15,
12226
+ cacheRead: 0.3,
12057
12227
  cacheWrite: 0,
12058
12228
  },
12059
12229
  contextWindow: 256000,
12060
- maxTokens: 16384,
12230
+ maxTokens: 128000,
12061
12231
  },
12062
12232
  "moonshotai/kimi-k2-thinking": {
12063
12233
  id: "moonshotai/kimi-k2-thinking",
@@ -12620,6 +12790,23 @@ export const MODELS = {
12620
12790
  contextWindow: 1050000,
12621
12791
  maxTokens: 128000,
12622
12792
  },
12793
+ "openai/gpt-oss-120b": {
12794
+ id: "openai/gpt-oss-120b",
12795
+ name: "gpt-oss-120b",
12796
+ api: "anthropic-messages",
12797
+ provider: "vercel-ai-gateway",
12798
+ baseUrl: "https://ai-gateway.vercel.sh",
12799
+ reasoning: true,
12800
+ input: ["text"],
12801
+ cost: {
12802
+ input: 0.15,
12803
+ output: 0.6,
12804
+ cacheRead: 0,
12805
+ cacheWrite: 0,
12806
+ },
12807
+ contextWindow: 128000,
12808
+ maxTokens: 8192,
12809
+ },
12623
12810
  "openai/gpt-oss-20b": {
12624
12811
  id: "openai/gpt-oss-20b",
12625
12812
  name: "gpt-oss-20b",
@@ -12629,12 +12816,12 @@ export const MODELS = {
12629
12816
  reasoning: true,
12630
12817
  input: ["text"],
12631
12818
  cost: {
12632
- input: 0.07,
12633
- output: 0.3,
12819
+ input: 0.049999999999999996,
12820
+ output: 0.19999999999999998,
12634
12821
  cacheRead: 0,
12635
12822
  cacheWrite: 0,
12636
12823
  },
12637
- contextWindow: 128000,
12824
+ contextWindow: 131072,
12638
12825
  maxTokens: 8192,
12639
12826
  },
12640
12827
  "openai/gpt-oss-safeguard-20b": {
@@ -13105,9 +13292,9 @@ export const MODELS = {
13105
13292
  reasoning: true,
13106
13293
  input: ["text"],
13107
13294
  cost: {
13108
- input: 0.09999999999999999,
13109
- output: 0.3,
13110
- cacheRead: 0.02,
13295
+ input: 0.09,
13296
+ output: 0.29,
13297
+ cacheRead: 0.045,
13111
13298
  cacheWrite: 0,
13112
13299
  },
13113
13300
  contextWindow: 262144,
@@ -13243,7 +13430,7 @@ export const MODELS = {
13243
13430
  cost: {
13244
13431
  input: 0.6,
13245
13432
  output: 2.2,
13246
- cacheRead: 0,
13433
+ cacheRead: 0.11,
13247
13434
  cacheWrite: 0,
13248
13435
  },
13249
13436
  contextWindow: 200000,
@@ -13283,6 +13470,23 @@ export const MODELS = {
13283
13470
  contextWindow: 200000,
13284
13471
  maxTokens: 128000,
13285
13472
  },
13473
+ "zai/glm-5": {
13474
+ id: "zai/glm-5",
13475
+ name: "GLM 5",
13476
+ api: "anthropic-messages",
13477
+ provider: "vercel-ai-gateway",
13478
+ baseUrl: "https://ai-gateway.vercel.sh",
13479
+ reasoning: true,
13480
+ input: ["text"],
13481
+ cost: {
13482
+ input: 1,
13483
+ output: 3.1999999999999997,
13484
+ cacheRead: 0.19999999999999998,
13485
+ cacheWrite: 0,
13486
+ },
13487
+ contextWindow: 202800,
13488
+ maxTokens: 131100,
13489
+ },
13286
13490
  "zai/glm-5-turbo": {
13287
13491
  id: "zai/glm-5-turbo",
13288
13492
  name: "GLM 5 Turbo",
@@ -13300,6 +13504,23 @@ export const MODELS = {
13300
13504
  contextWindow: 202800,
13301
13505
  maxTokens: 131100,
13302
13506
  },
13507
+ "zai/glm-5v-turbo": {
13508
+ id: "zai/glm-5v-turbo",
13509
+ name: "GLM 5V Turbo",
13510
+ api: "anthropic-messages",
13511
+ provider: "vercel-ai-gateway",
13512
+ baseUrl: "https://ai-gateway.vercel.sh",
13513
+ reasoning: true,
13514
+ input: ["text", "image"],
13515
+ cost: {
13516
+ input: 1.2,
13517
+ output: 4,
13518
+ cacheRead: 0.24,
13519
+ cacheWrite: 0,
13520
+ },
13521
+ contextWindow: 200000,
13522
+ maxTokens: 128000,
13523
+ },
13303
13524
  },
13304
13525
  "xai": {
13305
13526
  "grok-2": {
@@ -13722,9 +13943,9 @@ export const MODELS = {
13722
13943
  reasoning: true,
13723
13944
  input: ["text"],
13724
13945
  cost: {
13725
- input: 0.6,
13726
- output: 2.2,
13727
- cacheRead: 0.11,
13946
+ input: 0,
13947
+ output: 0,
13948
+ cacheRead: 0,
13728
13949
  cacheWrite: 0,
13729
13950
  },
13730
13951
  contextWindow: 131072,
@@ -13740,9 +13961,9 @@ export const MODELS = {
13740
13961
  reasoning: true,
13741
13962
  input: ["text"],
13742
13963
  cost: {
13743
- input: 0.2,
13744
- output: 1.1,
13745
- cacheRead: 0.03,
13964
+ input: 0,
13965
+ output: 0,
13966
+ cacheRead: 0,
13746
13967
  cacheWrite: 0,
13747
13968
  },
13748
13969
  contextWindow: 131072,
@@ -13776,8 +13997,8 @@ export const MODELS = {
13776
13997
  reasoning: true,
13777
13998
  input: ["text", "image"],
13778
13999
  cost: {
13779
- input: 0.6,
13780
- output: 1.8,
14000
+ input: 0,
14001
+ output: 0,
13781
14002
  cacheRead: 0,
13782
14003
  cacheWrite: 0,
13783
14004
  },
@@ -13790,13 +14011,13 @@ export const MODELS = {
13790
14011
  api: "openai-completions",
13791
14012
  provider: "zai",
13792
14013
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13793
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14014
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13794
14015
  reasoning: true,
13795
14016
  input: ["text"],
13796
14017
  cost: {
13797
- input: 0.6,
13798
- output: 2.2,
13799
- cacheRead: 0.11,
14018
+ input: 0,
14019
+ output: 0,
14020
+ cacheRead: 0,
13800
14021
  cacheWrite: 0,
13801
14022
  },
13802
14023
  contextWindow: 204800,
@@ -13808,12 +14029,12 @@ export const MODELS = {
13808
14029
  api: "openai-completions",
13809
14030
  provider: "zai",
13810
14031
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13811
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14032
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13812
14033
  reasoning: true,
13813
14034
  input: ["text", "image"],
13814
14035
  cost: {
13815
- input: 0.3,
13816
- output: 0.9,
14036
+ input: 0,
14037
+ output: 0,
13817
14038
  cacheRead: 0,
13818
14039
  cacheWrite: 0,
13819
14040
  },
@@ -13826,13 +14047,13 @@ export const MODELS = {
13826
14047
  api: "openai-completions",
13827
14048
  provider: "zai",
13828
14049
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13829
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14050
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13830
14051
  reasoning: true,
13831
14052
  input: ["text"],
13832
14053
  cost: {
13833
- input: 0.6,
13834
- output: 2.2,
13835
- cacheRead: 0.11,
14054
+ input: 0,
14055
+ output: 0,
14056
+ cacheRead: 0,
13836
14057
  cacheWrite: 0,
13837
14058
  },
13838
14059
  contextWindow: 204800,
@@ -13844,7 +14065,7 @@ export const MODELS = {
13844
14065
  api: "openai-completions",
13845
14066
  provider: "zai",
13846
14067
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13847
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14068
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13848
14069
  reasoning: true,
13849
14070
  input: ["text"],
13850
14071
  cost: {
@@ -13862,7 +14083,7 @@ export const MODELS = {
13862
14083
  api: "openai-completions",
13863
14084
  provider: "zai",
13864
14085
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13865
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14086
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13866
14087
  reasoning: true,
13867
14088
  input: ["text"],
13868
14089
  cost: {
@@ -13880,13 +14101,13 @@ export const MODELS = {
13880
14101
  api: "openai-completions",
13881
14102
  provider: "zai",
13882
14103
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13883
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14104
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13884
14105
  reasoning: true,
13885
14106
  input: ["text"],
13886
14107
  cost: {
13887
- input: 1,
13888
- output: 3.2,
13889
- cacheRead: 0.2,
14108
+ input: 0,
14109
+ output: 0,
14110
+ cacheRead: 0,
13890
14111
  cacheWrite: 0,
13891
14112
  },
13892
14113
  contextWindow: 204800,
@@ -13898,13 +14119,49 @@ export const MODELS = {
13898
14119
  api: "openai-completions",
13899
14120
  provider: "zai",
13900
14121
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
13901
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14122
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
13902
14123
  reasoning: true,
13903
14124
  input: ["text"],
13904
14125
  cost: {
13905
- input: 1.2,
13906
- output: 4,
13907
- cacheRead: 0.24,
14126
+ input: 0,
14127
+ output: 0,
14128
+ cacheRead: 0,
14129
+ cacheWrite: 0,
14130
+ },
14131
+ contextWindow: 200000,
14132
+ maxTokens: 131072,
14133
+ },
14134
+ "glm-5.1": {
14135
+ id: "glm-5.1",
14136
+ name: "GLM-5.1",
14137
+ api: "openai-completions",
14138
+ provider: "zai",
14139
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
14140
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14141
+ reasoning: true,
14142
+ input: ["text"],
14143
+ cost: {
14144
+ input: 0,
14145
+ output: 0,
14146
+ cacheRead: 0,
14147
+ cacheWrite: 0,
14148
+ },
14149
+ contextWindow: 200000,
14150
+ maxTokens: 131072,
14151
+ },
14152
+ "glm-5v-turbo": {
14153
+ id: "glm-5v-turbo",
14154
+ name: "glm-5v-turbo",
14155
+ api: "openai-completions",
14156
+ provider: "zai",
14157
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
14158
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14159
+ reasoning: true,
14160
+ input: ["text", "image"],
14161
+ cost: {
14162
+ input: 0,
14163
+ output: 0,
14164
+ cacheRead: 0,
13908
14165
  cacheWrite: 0,
13909
14166
  },
13910
14167
  contextWindow: 200000,