@avadisabelle/ava-pi-ai 0.64.0 → 0.64.2

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.
@@ -257,6 +257,23 @@ export const MODELS = {
257
257
  contextWindow: 1000000,
258
258
  maxTokens: 128000,
259
259
  },
260
+ "anthropic.claude-opus-4-7": {
261
+ id: "anthropic.claude-opus-4-7",
262
+ name: "Claude Opus 4.7",
263
+ api: "bedrock-converse-stream",
264
+ provider: "amazon-bedrock",
265
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
266
+ reasoning: true,
267
+ input: ["text", "image"],
268
+ cost: {
269
+ input: 5,
270
+ output: 25,
271
+ cacheRead: 0.5,
272
+ cacheWrite: 6.25,
273
+ },
274
+ contextWindow: 1000000,
275
+ maxTokens: 128000,
276
+ },
260
277
  "anthropic.claude-sonnet-4-20250514-v1:0": {
261
278
  id: "anthropic.claude-sonnet-4-20250514-v1:0",
262
279
  name: "Claude Sonnet 4",
@@ -410,6 +427,23 @@ export const MODELS = {
410
427
  contextWindow: 1000000,
411
428
  maxTokens: 128000,
412
429
  },
430
+ "eu.anthropic.claude-opus-4-7": {
431
+ id: "eu.anthropic.claude-opus-4-7",
432
+ name: "Claude Opus 4.7 (EU)",
433
+ api: "bedrock-converse-stream",
434
+ provider: "amazon-bedrock",
435
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
436
+ reasoning: true,
437
+ input: ["text", "image"],
438
+ cost: {
439
+ input: 5,
440
+ output: 25,
441
+ cacheRead: 0.5,
442
+ cacheWrite: 6.25,
443
+ },
444
+ contextWindow: 1000000,
445
+ maxTokens: 128000,
446
+ },
413
447
  "eu.anthropic.claude-sonnet-4-20250514-v1:0": {
414
448
  id: "eu.anthropic.claude-sonnet-4-20250514-v1:0",
415
449
  name: "Claude Sonnet 4 (EU)",
@@ -512,6 +546,23 @@ export const MODELS = {
512
546
  contextWindow: 1000000,
513
547
  maxTokens: 128000,
514
548
  },
549
+ "global.anthropic.claude-opus-4-7": {
550
+ id: "global.anthropic.claude-opus-4-7",
551
+ name: "Claude Opus 4.7 (Global)",
552
+ api: "bedrock-converse-stream",
553
+ provider: "amazon-bedrock",
554
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
555
+ reasoning: true,
556
+ input: ["text", "image"],
557
+ cost: {
558
+ input: 5,
559
+ output: 25,
560
+ cacheRead: 0.5,
561
+ cacheWrite: 6.25,
562
+ },
563
+ contextWindow: 1000000,
564
+ maxTokens: 128000,
565
+ },
515
566
  "global.anthropic.claude-sonnet-4-20250514-v1:0": {
516
567
  id: "global.anthropic.claude-sonnet-4-20250514-v1:0",
517
568
  name: "Claude Sonnet 4 (Global)",
@@ -1209,6 +1260,23 @@ export const MODELS = {
1209
1260
  contextWindow: 131072,
1210
1261
  maxTokens: 65536,
1211
1262
  },
1263
+ "qwen.qwen3-coder-next": {
1264
+ id: "qwen.qwen3-coder-next",
1265
+ name: "Qwen3 Coder Next",
1266
+ api: "bedrock-converse-stream",
1267
+ provider: "amazon-bedrock",
1268
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1269
+ reasoning: true,
1270
+ input: ["text"],
1271
+ cost: {
1272
+ input: 0.22,
1273
+ output: 1.8,
1274
+ cacheRead: 0,
1275
+ cacheWrite: 0,
1276
+ },
1277
+ contextWindow: 131072,
1278
+ maxTokens: 65536,
1279
+ },
1212
1280
  "qwen.qwen3-next-80b-a3b": {
1213
1281
  id: "qwen.qwen3-next-80b-a3b",
1214
1282
  name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
@@ -1328,6 +1396,23 @@ export const MODELS = {
1328
1396
  contextWindow: 1000000,
1329
1397
  maxTokens: 128000,
1330
1398
  },
1399
+ "us.anthropic.claude-opus-4-7": {
1400
+ id: "us.anthropic.claude-opus-4-7",
1401
+ name: "Claude Opus 4.7 (US)",
1402
+ api: "bedrock-converse-stream",
1403
+ provider: "amazon-bedrock",
1404
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1405
+ reasoning: true,
1406
+ input: ["text", "image"],
1407
+ cost: {
1408
+ input: 5,
1409
+ output: 25,
1410
+ cacheRead: 0.5,
1411
+ cacheWrite: 6.25,
1412
+ },
1413
+ contextWindow: 1000000,
1414
+ maxTokens: 128000,
1415
+ },
1331
1416
  "us.anthropic.claude-sonnet-4-20250514-v1:0": {
1332
1417
  id: "us.anthropic.claude-sonnet-4-20250514-v1:0",
1333
1418
  name: "Claude Sonnet 4 (US)",
@@ -1755,6 +1840,23 @@ export const MODELS = {
1755
1840
  contextWindow: 1000000,
1756
1841
  maxTokens: 128000,
1757
1842
  },
1843
+ "claude-opus-4-7": {
1844
+ id: "claude-opus-4-7",
1845
+ name: "Claude Opus 4.7",
1846
+ api: "anthropic-messages",
1847
+ provider: "anthropic",
1848
+ baseUrl: "https://api.anthropic.com",
1849
+ reasoning: true,
1850
+ input: ["text", "image"],
1851
+ cost: {
1852
+ input: 5,
1853
+ output: 25,
1854
+ cacheRead: 0.5,
1855
+ cacheWrite: 6.25,
1856
+ },
1857
+ contextWindow: 1000000,
1858
+ maxTokens: 128000,
1859
+ },
1758
1860
  "claude-sonnet-4-0": {
1759
1861
  id: "claude-sonnet-4-0",
1760
1862
  name: "Claude Sonnet 4 (latest)",
@@ -3015,7 +3117,7 @@ export const MODELS = {
3015
3117
  },
3016
3118
  "gpt-5.4-mini": {
3017
3119
  id: "gpt-5.4-mini",
3018
- name: "GPT-5.4 mini",
3120
+ name: "GPT-5.4 Mini",
3019
3121
  api: "openai-responses",
3020
3122
  provider: "github-copilot",
3021
3123
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3477,8 +3579,8 @@ export const MODELS = {
3477
3579
  contextWindow: 131072,
3478
3580
  maxTokens: 8192,
3479
3581
  },
3480
- "gemma-4-26b": {
3481
- id: "gemma-4-26b",
3582
+ "gemma-4-26b-it": {
3583
+ id: "gemma-4-26b-it",
3482
3584
  name: "Gemma 4 26B",
3483
3585
  api: "google-generative-ai",
3484
3586
  provider: "google",
@@ -3494,8 +3596,8 @@ export const MODELS = {
3494
3596
  contextWindow: 256000,
3495
3597
  maxTokens: 8192,
3496
3598
  },
3497
- "gemma-4-31b": {
3498
- id: "gemma-4-31b",
3599
+ "gemma-4-31b-it": {
3600
+ id: "gemma-4-31b-it",
3499
3601
  name: "Gemma 4 31B",
3500
3602
  api: "google-generative-ai",
3501
3603
  provider: "google",
@@ -4322,6 +4424,24 @@ export const MODELS = {
4322
4424
  contextWindow: 204800,
4323
4425
  maxTokens: 131072,
4324
4426
  },
4427
+ "MiniMaxAI/MiniMax-M2.7": {
4428
+ id: "MiniMaxAI/MiniMax-M2.7",
4429
+ name: "MiniMax-M2.7",
4430
+ api: "openai-completions",
4431
+ provider: "huggingface",
4432
+ baseUrl: "https://router.huggingface.co/v1",
4433
+ compat: { "supportsDeveloperRole": false },
4434
+ reasoning: true,
4435
+ input: ["text"],
4436
+ cost: {
4437
+ input: 0.3,
4438
+ output: 1.2,
4439
+ cacheRead: 0.06,
4440
+ cacheWrite: 0,
4441
+ },
4442
+ contextWindow: 204800,
4443
+ maxTokens: 131072,
4444
+ },
4325
4445
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
4326
4446
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
4327
4447
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -4610,6 +4730,24 @@ export const MODELS = {
4610
4730
  contextWindow: 202752,
4611
4731
  maxTokens: 131072,
4612
4732
  },
4733
+ "zai-org/GLM-5.1": {
4734
+ id: "zai-org/GLM-5.1",
4735
+ name: "GLM-5.1",
4736
+ api: "openai-completions",
4737
+ provider: "huggingface",
4738
+ baseUrl: "https://router.huggingface.co/v1",
4739
+ compat: { "supportsDeveloperRole": false },
4740
+ reasoning: true,
4741
+ input: ["text"],
4742
+ cost: {
4743
+ input: 1,
4744
+ output: 3.2,
4745
+ cacheRead: 0.2,
4746
+ cacheWrite: 0,
4747
+ },
4748
+ contextWindow: 202752,
4749
+ maxTokens: 131072,
4750
+ },
4613
4751
  },
4614
4752
  "kimi-coding": {
4615
4753
  "k2p5": {
@@ -6290,6 +6428,23 @@ export const MODELS = {
6290
6428
  contextWindow: 1000000,
6291
6429
  maxTokens: 128000,
6292
6430
  },
6431
+ "claude-opus-4-7": {
6432
+ id: "claude-opus-4-7",
6433
+ name: "Claude Opus 4.7",
6434
+ api: "anthropic-messages",
6435
+ provider: "opencode",
6436
+ baseUrl: "https://opencode.ai/zen",
6437
+ reasoning: true,
6438
+ input: ["text", "image"],
6439
+ cost: {
6440
+ input: 5,
6441
+ output: 25,
6442
+ cacheRead: 0.5,
6443
+ cacheWrite: 6.25,
6444
+ },
6445
+ contextWindow: 1000000,
6446
+ maxTokens: 128000,
6447
+ },
6293
6448
  "claude-sonnet-4": {
6294
6449
  id: "claude-sonnet-4",
6295
6450
  name: "Claude Sonnet 4",
@@ -6392,6 +6547,23 @@ export const MODELS = {
6392
6547
  contextWindow: 204800,
6393
6548
  maxTokens: 131072,
6394
6549
  },
6550
+ "glm-5.1": {
6551
+ id: "glm-5.1",
6552
+ name: "GLM-5.1",
6553
+ api: "openai-completions",
6554
+ provider: "opencode",
6555
+ baseUrl: "https://opencode.ai/zen/v1",
6556
+ reasoning: true,
6557
+ input: ["text"],
6558
+ cost: {
6559
+ input: 1.4,
6560
+ output: 4.4,
6561
+ cacheRead: 0.26,
6562
+ cacheWrite: 0,
6563
+ },
6564
+ contextWindow: 204800,
6565
+ maxTokens: 131072,
6566
+ },
6395
6567
  "gpt-5": {
6396
6568
  id: "gpt-5",
6397
6569
  name: "GPT-5",
@@ -6698,22 +6870,39 @@ export const MODELS = {
6698
6870
  contextWindow: 204800,
6699
6871
  maxTokens: 128000,
6700
6872
  },
6701
- "qwen3.6-plus-free": {
6702
- id: "qwen3.6-plus-free",
6703
- name: "Qwen3.6 Plus Free",
6873
+ "qwen3.5-plus": {
6874
+ id: "qwen3.5-plus",
6875
+ name: "Qwen3.5 Plus",
6704
6876
  api: "openai-completions",
6705
6877
  provider: "opencode",
6706
6878
  baseUrl: "https://opencode.ai/zen/v1",
6707
6879
  reasoning: true,
6708
- input: ["text"],
6880
+ input: ["text", "image"],
6709
6881
  cost: {
6710
- input: 0,
6711
- output: 0,
6712
- cacheRead: 0,
6713
- cacheWrite: 0,
6882
+ input: 0.2,
6883
+ output: 1.2,
6884
+ cacheRead: 0.02,
6885
+ cacheWrite: 0.25,
6714
6886
  },
6715
- contextWindow: 1048576,
6716
- maxTokens: 64000,
6887
+ contextWindow: 262144,
6888
+ maxTokens: 65536,
6889
+ },
6890
+ "qwen3.6-plus": {
6891
+ id: "qwen3.6-plus",
6892
+ name: "Qwen3.6 Plus",
6893
+ api: "openai-completions",
6894
+ provider: "opencode",
6895
+ baseUrl: "https://opencode.ai/zen/v1",
6896
+ reasoning: true,
6897
+ input: ["text", "image"],
6898
+ cost: {
6899
+ input: 0.5,
6900
+ output: 3,
6901
+ cacheRead: 0.05,
6902
+ cacheWrite: 0.625,
6903
+ },
6904
+ contextWindow: 262144,
6905
+ maxTokens: 65536,
6717
6906
  },
6718
6907
  },
6719
6908
  "opencode-go": {
@@ -6734,6 +6923,23 @@ export const MODELS = {
6734
6923
  contextWindow: 204800,
6735
6924
  maxTokens: 131072,
6736
6925
  },
6926
+ "glm-5.1": {
6927
+ id: "glm-5.1",
6928
+ name: "GLM-5.1",
6929
+ api: "openai-completions",
6930
+ provider: "opencode-go",
6931
+ baseUrl: "https://opencode.ai/zen/go/v1",
6932
+ reasoning: true,
6933
+ input: ["text"],
6934
+ cost: {
6935
+ input: 1.4,
6936
+ output: 4.4,
6937
+ cacheRead: 0.26,
6938
+ cacheWrite: 0,
6939
+ },
6940
+ contextWindow: 204800,
6941
+ maxTokens: 131072,
6942
+ },
6737
6943
  "kimi-k2.5": {
6738
6944
  id: "kimi-k2.5",
6739
6945
  name: "Kimi K2.5",
@@ -6788,9 +6994,9 @@ export const MODELS = {
6788
6994
  "minimax-m2.5": {
6789
6995
  id: "minimax-m2.5",
6790
6996
  name: "MiniMax M2.5",
6791
- api: "openai-completions",
6997
+ api: "anthropic-messages",
6792
6998
  provider: "opencode-go",
6793
- baseUrl: "https://opencode.ai/zen/go/v1",
6999
+ baseUrl: "https://opencode.ai/zen/go",
6794
7000
  reasoning: true,
6795
7001
  input: ["text"],
6796
7002
  cost: {
@@ -6819,6 +7025,40 @@ export const MODELS = {
6819
7025
  contextWindow: 204800,
6820
7026
  maxTokens: 131072,
6821
7027
  },
7028
+ "qwen3.5-plus": {
7029
+ id: "qwen3.5-plus",
7030
+ name: "Qwen3.5 Plus",
7031
+ api: "openai-completions",
7032
+ provider: "opencode-go",
7033
+ baseUrl: "https://opencode.ai/zen/go/v1",
7034
+ reasoning: true,
7035
+ input: ["text", "image"],
7036
+ cost: {
7037
+ input: 0.2,
7038
+ output: 1.2,
7039
+ cacheRead: 0.02,
7040
+ cacheWrite: 0.25,
7041
+ },
7042
+ contextWindow: 262144,
7043
+ maxTokens: 65536,
7044
+ },
7045
+ "qwen3.6-plus": {
7046
+ id: "qwen3.6-plus",
7047
+ name: "Qwen3.6 Plus",
7048
+ api: "openai-completions",
7049
+ provider: "opencode-go",
7050
+ baseUrl: "https://opencode.ai/zen/go/v1",
7051
+ reasoning: true,
7052
+ input: ["text", "image"],
7053
+ cost: {
7054
+ input: 0.5,
7055
+ output: 3,
7056
+ cacheRead: 0.05,
7057
+ cacheWrite: 0.625,
7058
+ },
7059
+ contextWindow: 262144,
7060
+ maxTokens: 65536,
7061
+ },
6822
7062
  },
6823
7063
  "openrouter": {
6824
7064
  "ai21/jamba-large-1.7": {
@@ -7006,7 +7246,7 @@ export const MODELS = {
7006
7246
  cacheWrite: 3.75,
7007
7247
  },
7008
7248
  contextWindow: 200000,
7009
- maxTokens: 64000,
7249
+ maxTokens: 128000,
7010
7250
  },
7011
7251
  "anthropic/claude-3.7-sonnet:thinking": {
7012
7252
  id: "anthropic/claude-3.7-sonnet:thinking",
@@ -7110,6 +7350,40 @@ export const MODELS = {
7110
7350
  contextWindow: 1000000,
7111
7351
  maxTokens: 128000,
7112
7352
  },
7353
+ "anthropic/claude-opus-4.6-fast": {
7354
+ id: "anthropic/claude-opus-4.6-fast",
7355
+ name: "Anthropic: Claude Opus 4.6 (Fast)",
7356
+ api: "openai-completions",
7357
+ provider: "openrouter",
7358
+ baseUrl: "https://openrouter.ai/api/v1",
7359
+ reasoning: true,
7360
+ input: ["text", "image"],
7361
+ cost: {
7362
+ input: 30,
7363
+ output: 150,
7364
+ cacheRead: 3,
7365
+ cacheWrite: 37.5,
7366
+ },
7367
+ contextWindow: 1000000,
7368
+ maxTokens: 128000,
7369
+ },
7370
+ "anthropic/claude-opus-4.7": {
7371
+ id: "anthropic/claude-opus-4.7",
7372
+ name: "Anthropic: Claude Opus 4.7",
7373
+ api: "openai-completions",
7374
+ provider: "openrouter",
7375
+ baseUrl: "https://openrouter.ai/api/v1",
7376
+ reasoning: true,
7377
+ input: ["text", "image"],
7378
+ cost: {
7379
+ input: 5,
7380
+ output: 25,
7381
+ cacheRead: 0.5,
7382
+ cacheWrite: 6.25,
7383
+ },
7384
+ contextWindow: 1000000,
7385
+ maxTokens: 128000,
7386
+ },
7113
7387
  "anthropic/claude-sonnet-4": {
7114
7388
  id: "anthropic/claude-sonnet-4",
7115
7389
  name: "Anthropic: Claude Sonnet 4",
@@ -7124,7 +7398,7 @@ export const MODELS = {
7124
7398
  cacheRead: 0.3,
7125
7399
  cacheWrite: 3.75,
7126
7400
  },
7127
- contextWindow: 200000,
7401
+ contextWindow: 1000000,
7128
7402
  maxTokens: 64000,
7129
7403
  },
7130
7404
  "anthropic/claude-sonnet-4.5": {
@@ -7189,7 +7463,7 @@ export const MODELS = {
7189
7463
  cost: {
7190
7464
  input: 0.22,
7191
7465
  output: 0.85,
7192
- cacheRead: 0,
7466
+ cacheRead: 0.06,
7193
7467
  cacheWrite: 0,
7194
7468
  },
7195
7469
  contextWindow: 262144,
@@ -7212,39 +7486,22 @@ export const MODELS = {
7212
7486
  contextWindow: 131072,
7213
7487
  maxTokens: 131072,
7214
7488
  },
7215
- "arcee-ai/trinity-mini:free": {
7216
- id: "arcee-ai/trinity-mini:free",
7217
- name: "Arcee AI: Trinity Mini (free)",
7489
+ "arcee-ai/virtuoso-large": {
7490
+ id: "arcee-ai/virtuoso-large",
7491
+ name: "Arcee AI: Virtuoso Large",
7218
7492
  api: "openai-completions",
7219
7493
  provider: "openrouter",
7220
7494
  baseUrl: "https://openrouter.ai/api/v1",
7221
- reasoning: true,
7495
+ reasoning: false,
7222
7496
  input: ["text"],
7223
7497
  cost: {
7224
- input: 0,
7225
- output: 0,
7498
+ input: 0.75,
7499
+ output: 1.2,
7226
7500
  cacheRead: 0,
7227
7501
  cacheWrite: 0,
7228
7502
  },
7229
7503
  contextWindow: 131072,
7230
- maxTokens: 4096,
7231
- },
7232
- "arcee-ai/virtuoso-large": {
7233
- id: "arcee-ai/virtuoso-large",
7234
- name: "Arcee AI: Virtuoso Large",
7235
- api: "openai-completions",
7236
- provider: "openrouter",
7237
- baseUrl: "https://openrouter.ai/api/v1",
7238
- reasoning: false,
7239
- input: ["text"],
7240
- cost: {
7241
- input: 0.75,
7242
- output: 1.2,
7243
- cacheRead: 0,
7244
- cacheWrite: 0,
7245
- },
7246
- contextWindow: 131072,
7247
- maxTokens: 64000,
7504
+ maxTokens: 64000,
7248
7505
  },
7249
7506
  "auto": {
7250
7507
  id: "auto",
@@ -7476,13 +7733,13 @@ export const MODELS = {
7476
7733
  reasoning: true,
7477
7734
  input: ["text"],
7478
7735
  cost: {
7479
- input: 0.44999999999999996,
7736
+ input: 0.5,
7480
7737
  output: 2.1500000000000004,
7481
- cacheRead: 0.22499999999999998,
7738
+ cacheRead: 0.35,
7482
7739
  cacheWrite: 0,
7483
7740
  },
7484
7741
  contextWindow: 163840,
7485
- maxTokens: 65536,
7742
+ maxTokens: 4096,
7486
7743
  },
7487
7744
  "deepseek/deepseek-v3.1-terminus": {
7488
7745
  id: "deepseek/deepseek-v3.1-terminus",
@@ -7765,72 +8022,72 @@ export const MODELS = {
7765
8022
  reasoning: true,
7766
8023
  input: ["text", "image"],
7767
8024
  cost: {
7768
- input: 0.13,
8025
+ input: 0.07,
7769
8026
  output: 0.39999999999999997,
7770
- cacheRead: 0,
8027
+ cacheRead: 0.04,
7771
8028
  cacheWrite: 0,
7772
8029
  },
7773
8030
  contextWindow: 262144,
7774
8031
  maxTokens: 262144,
7775
8032
  },
7776
- "google/gemma-4-31b-it": {
7777
- id: "google/gemma-4-31b-it",
7778
- name: "Google: Gemma 4 31B",
8033
+ "google/gemma-4-26b-a4b-it:free": {
8034
+ id: "google/gemma-4-26b-a4b-it:free",
8035
+ name: "Google: Gemma 4 26B A4B (free)",
7779
8036
  api: "openai-completions",
7780
8037
  provider: "openrouter",
7781
8038
  baseUrl: "https://openrouter.ai/api/v1",
7782
8039
  reasoning: true,
7783
8040
  input: ["text", "image"],
7784
8041
  cost: {
7785
- input: 0.14,
7786
- output: 0.39999999999999997,
8042
+ input: 0,
8043
+ output: 0,
7787
8044
  cacheRead: 0,
7788
8045
  cacheWrite: 0,
7789
8046
  },
7790
8047
  contextWindow: 262144,
7791
- maxTokens: 131072,
8048
+ maxTokens: 32768,
7792
8049
  },
7793
- "inception/mercury": {
7794
- id: "inception/mercury",
7795
- name: "Inception: Mercury",
8050
+ "google/gemma-4-31b-it": {
8051
+ id: "google/gemma-4-31b-it",
8052
+ name: "Google: Gemma 4 31B",
7796
8053
  api: "openai-completions",
7797
8054
  provider: "openrouter",
7798
8055
  baseUrl: "https://openrouter.ai/api/v1",
7799
- reasoning: false,
7800
- input: ["text"],
8056
+ reasoning: true,
8057
+ input: ["text", "image"],
7801
8058
  cost: {
7802
- input: 0.25,
7803
- output: 0.75,
7804
- cacheRead: 0.024999999999999998,
8059
+ input: 0.13,
8060
+ output: 0.38,
8061
+ cacheRead: 0.019999999499999997,
7805
8062
  cacheWrite: 0,
7806
8063
  },
7807
- contextWindow: 128000,
7808
- maxTokens: 32000,
8064
+ contextWindow: 262144,
8065
+ maxTokens: 4096,
7809
8066
  },
7810
- "inception/mercury-2": {
7811
- id: "inception/mercury-2",
7812
- name: "Inception: Mercury 2",
8067
+ "google/gemma-4-31b-it:free": {
8068
+ id: "google/gemma-4-31b-it:free",
8069
+ name: "Google: Gemma 4 31B (free)",
7813
8070
  api: "openai-completions",
7814
8071
  provider: "openrouter",
7815
8072
  baseUrl: "https://openrouter.ai/api/v1",
7816
8073
  reasoning: true,
7817
- input: ["text"],
8074
+ input: ["text", "image"],
7818
8075
  cost: {
7819
- input: 0.25,
7820
- output: 0.75,
7821
- cacheRead: 0.024999999999999998,
8076
+ input: 0,
8077
+ output: 0,
8078
+ cacheRead: 0,
7822
8079
  cacheWrite: 0,
7823
8080
  },
7824
- contextWindow: 128000,
7825
- maxTokens: 50000,
8081
+ contextWindow: 262144,
8082
+ maxTokens: 32768,
7826
8083
  },
7827
- "inception/mercury-coder": {
7828
- id: "inception/mercury-coder",
7829
- name: "Inception: Mercury Coder",
8084
+ "inception/mercury-2": {
8085
+ id: "inception/mercury-2",
8086
+ name: "Inception: Mercury 2",
7830
8087
  api: "openai-completions",
7831
8088
  provider: "openrouter",
7832
8089
  baseUrl: "https://openrouter.ai/api/v1",
7833
- reasoning: false,
8090
+ reasoning: true,
7834
8091
  input: ["text"],
7835
8092
  cost: {
7836
8093
  input: 0.25,
@@ -7839,7 +8096,7 @@ export const MODELS = {
7839
8096
  cacheWrite: 0,
7840
8097
  },
7841
8098
  contextWindow: 128000,
7842
- maxTokens: 32000,
8099
+ maxTokens: 50000,
7843
8100
  },
7844
8101
  "kwaipilot/kat-coder-pro-v2": {
7845
8102
  id: "kwaipilot/kat-coder-pro-v2",
@@ -7858,23 +8115,6 @@ export const MODELS = {
7858
8115
  contextWindow: 256000,
7859
8116
  maxTokens: 80000,
7860
8117
  },
7861
- "meituan/longcat-flash-chat": {
7862
- id: "meituan/longcat-flash-chat",
7863
- name: "Meituan: LongCat Flash Chat",
7864
- api: "openai-completions",
7865
- provider: "openrouter",
7866
- baseUrl: "https://openrouter.ai/api/v1",
7867
- reasoning: false,
7868
- input: ["text"],
7869
- cost: {
7870
- input: 0.19999999999999998,
7871
- output: 0.7999999999999999,
7872
- cacheRead: 0.19999999999999998,
7873
- cacheWrite: 0,
7874
- },
7875
- contextWindow: 131072,
7876
- maxTokens: 131072,
7877
- },
7878
8118
  "meta-llama/llama-3-8b-instruct": {
7879
8119
  id: "meta-llama/llama-3-8b-instruct",
7880
8120
  name: "Meta: Llama 3 8B Instruct",
@@ -8037,13 +8277,13 @@ export const MODELS = {
8037
8277
  reasoning: true,
8038
8278
  input: ["text"],
8039
8279
  cost: {
8040
- input: 0.27,
8280
+ input: 0.29,
8041
8281
  output: 0.95,
8042
- cacheRead: 0.0290000007,
8282
+ cacheRead: 0.03,
8043
8283
  cacheWrite: 0,
8044
8284
  },
8045
8285
  contextWindow: 196608,
8046
- maxTokens: 4096,
8286
+ maxTokens: 196608,
8047
8287
  },
8048
8288
  "minimax/minimax-m2.5": {
8049
8289
  id: "minimax/minimax-m2.5",
@@ -8077,7 +8317,7 @@ export const MODELS = {
8077
8317
  cacheWrite: 0,
8078
8318
  },
8079
8319
  contextWindow: 196608,
8080
- maxTokens: 196608,
8320
+ maxTokens: 8192,
8081
8321
  },
8082
8322
  "minimax/minimax-m2.7": {
8083
8323
  id: "minimax/minimax-m2.7",
@@ -8090,11 +8330,11 @@ export const MODELS = {
8090
8330
  cost: {
8091
8331
  input: 0.3,
8092
8332
  output: 1.2,
8093
- cacheRead: 0.06,
8333
+ cacheRead: 0.059,
8094
8334
  cacheWrite: 0,
8095
8335
  },
8096
- contextWindow: 204800,
8097
- maxTokens: 131072,
8336
+ contextWindow: 196608,
8337
+ maxTokens: 4096,
8098
8338
  },
8099
8339
  "mistralai/codestral-2508": {
8100
8340
  id: "mistralai/codestral-2508",
@@ -8485,7 +8725,7 @@ export const MODELS = {
8485
8725
  cacheWrite: 0,
8486
8726
  },
8487
8727
  contextWindow: 131072,
8488
- maxTokens: 131072,
8728
+ maxTokens: 32768,
8489
8729
  },
8490
8730
  "moonshotai/kimi-k2-0905": {
8491
8731
  id: "moonshotai/kimi-k2-0905",
@@ -8498,11 +8738,11 @@ export const MODELS = {
8498
8738
  cost: {
8499
8739
  input: 0.39999999999999997,
8500
8740
  output: 2,
8501
- cacheRead: 0.15,
8741
+ cacheRead: 0,
8502
8742
  cacheWrite: 0,
8503
8743
  },
8504
- contextWindow: 131072,
8505
- maxTokens: 4096,
8744
+ contextWindow: 262144,
8745
+ maxTokens: 262144,
8506
8746
  },
8507
8747
  "moonshotai/kimi-k2-thinking": {
8508
8748
  id: "moonshotai/kimi-k2-thinking",
@@ -8513,13 +8753,13 @@ export const MODELS = {
8513
8753
  reasoning: true,
8514
8754
  input: ["text"],
8515
8755
  cost: {
8516
- input: 0.47,
8517
- output: 2,
8518
- cacheRead: 0.14100000000000001,
8756
+ input: 0.6,
8757
+ output: 2.5,
8758
+ cacheRead: 0.15,
8519
8759
  cacheWrite: 0,
8520
8760
  },
8521
- contextWindow: 131072,
8522
- maxTokens: 4096,
8761
+ contextWindow: 262144,
8762
+ maxTokens: 262144,
8523
8763
  },
8524
8764
  "moonshotai/kimi-k2.5": {
8525
8765
  id: "moonshotai/kimi-k2.5",
@@ -8632,9 +8872,9 @@ export const MODELS = {
8632
8872
  reasoning: true,
8633
8873
  input: ["text"],
8634
8874
  cost: {
8635
- input: 0.09999999999999999,
8636
- output: 0.5,
8637
- cacheRead: 0.09999999999999999,
8875
+ input: 0.09,
8876
+ output: 0.44999999999999996,
8877
+ cacheRead: 0,
8638
8878
  cacheWrite: 0,
8639
8879
  },
8640
8880
  contextWindow: 262144,
@@ -8859,7 +9099,7 @@ export const MODELS = {
8859
9099
  cacheWrite: 0,
8860
9100
  },
8861
9101
  contextWindow: 1047576,
8862
- maxTokens: 32768,
9102
+ maxTokens: 4096,
8863
9103
  },
8864
9104
  "openai/gpt-4.1-mini": {
8865
9105
  id: "openai/gpt-4.1-mini",
@@ -8906,7 +9146,7 @@ export const MODELS = {
8906
9146
  cost: {
8907
9147
  input: 2.5,
8908
9148
  output: 10,
8909
- cacheRead: 1.25,
9149
+ cacheRead: 0,
8910
9150
  cacheWrite: 0,
8911
9151
  },
8912
9152
  contextWindow: 128000,
@@ -9127,11 +9367,11 @@ export const MODELS = {
9127
9367
  cost: {
9128
9368
  input: 0.049999999999999996,
9129
9369
  output: 0.39999999999999997,
9130
- cacheRead: 0.005,
9370
+ cacheRead: 0.01,
9131
9371
  cacheWrite: 0,
9132
9372
  },
9133
9373
  contextWindow: 400000,
9134
- maxTokens: 128000,
9374
+ maxTokens: 4096,
9135
9375
  },
9136
9376
  "openai/gpt-5-pro": {
9137
9377
  id: "openai/gpt-5-pro",
@@ -9161,7 +9401,7 @@ export const MODELS = {
9161
9401
  cost: {
9162
9402
  input: 1.25,
9163
9403
  output: 10,
9164
- cacheRead: 0.125,
9404
+ cacheRead: 0.13,
9165
9405
  cacheWrite: 0,
9166
9406
  },
9167
9407
  contextWindow: 400000,
@@ -9229,11 +9469,11 @@ export const MODELS = {
9229
9469
  cost: {
9230
9470
  input: 0.25,
9231
9471
  output: 2,
9232
- cacheRead: 0.024999999999999998,
9472
+ cacheRead: 0.03,
9233
9473
  cacheWrite: 0,
9234
9474
  },
9235
9475
  contextWindow: 400000,
9236
- maxTokens: 100000,
9476
+ maxTokens: 128000,
9237
9477
  },
9238
9478
  "openai/gpt-5.2": {
9239
9479
  id: "openai/gpt-5.2",
@@ -9267,7 +9507,7 @@ export const MODELS = {
9267
9507
  cacheWrite: 0,
9268
9508
  },
9269
9509
  contextWindow: 128000,
9270
- maxTokens: 16384,
9510
+ maxTokens: 32000,
9271
9511
  },
9272
9512
  "openai/gpt-5.2-codex": {
9273
9513
  id: "openai/gpt-5.2-codex",
@@ -9483,12 +9723,12 @@ export const MODELS = {
9483
9723
  input: ["text"],
9484
9724
  cost: {
9485
9725
  input: 0.03,
9486
- output: 0.11,
9487
- cacheRead: 0.015,
9726
+ output: 0.14,
9727
+ cacheRead: 0,
9488
9728
  cacheWrite: 0,
9489
9729
  },
9490
9730
  contextWindow: 131072,
9491
- maxTokens: 131072,
9731
+ maxTokens: 4096,
9492
9732
  },
9493
9733
  "openai/gpt-oss-20b:free": {
9494
9734
  id: "openai/gpt-oss-20b:free",
@@ -9505,7 +9745,7 @@ export const MODELS = {
9505
9745
  cacheWrite: 0,
9506
9746
  },
9507
9747
  contextWindow: 131072,
9508
- maxTokens: 131072,
9748
+ maxTokens: 8192,
9509
9749
  },
9510
9750
  "openai/gpt-oss-safeguard-20b": {
9511
9751
  id: "openai/gpt-oss-safeguard-20b",
@@ -9694,6 +9934,23 @@ export const MODELS = {
9694
9934
  contextWindow: 2000000,
9695
9935
  maxTokens: 4096,
9696
9936
  },
9937
+ "openrouter/elephant-alpha": {
9938
+ id: "openrouter/elephant-alpha",
9939
+ name: "Elephant",
9940
+ api: "openai-completions",
9941
+ provider: "openrouter",
9942
+ baseUrl: "https://openrouter.ai/api/v1",
9943
+ reasoning: false,
9944
+ input: ["text"],
9945
+ cost: {
9946
+ input: 0,
9947
+ output: 0,
9948
+ cacheRead: 0,
9949
+ cacheWrite: 0,
9950
+ },
9951
+ contextWindow: 262144,
9952
+ maxTokens: 32768,
9953
+ },
9697
9954
  "openrouter/free": {
9698
9955
  id: "openrouter/free",
9699
9956
  name: "Free Models Router",
@@ -9791,7 +10048,7 @@ export const MODELS = {
9791
10048
  input: 0.26,
9792
10049
  output: 0.78,
9793
10050
  cacheRead: 0.052000000000000005,
9794
- cacheWrite: 0,
10051
+ cacheWrite: 0.325,
9795
10052
  },
9796
10053
  contextWindow: 1000000,
9797
10054
  maxTokens: 32768,
@@ -9808,7 +10065,7 @@ export const MODELS = {
9808
10065
  input: 0.26,
9809
10066
  output: 0.78,
9810
10067
  cacheRead: 0,
9811
- cacheWrite: 0,
10068
+ cacheWrite: 0.325,
9812
10069
  },
9813
10070
  contextWindow: 1000000,
9814
10071
  maxTokens: 32768,
@@ -9825,7 +10082,7 @@ export const MODELS = {
9825
10082
  input: 0.26,
9826
10083
  output: 0.78,
9827
10084
  cacheRead: 0,
9828
- cacheWrite: 0,
10085
+ cacheWrite: 0.325,
9829
10086
  },
9830
10087
  contextWindow: 1000000,
9831
10088
  maxTokens: 32768,
@@ -9924,13 +10181,13 @@ export const MODELS = {
9924
10181
  reasoning: true,
9925
10182
  input: ["text"],
9926
10183
  cost: {
9927
- input: 0.14950000000000002,
9928
- output: 1.495,
10184
+ input: 0.13,
10185
+ output: 0.6,
9929
10186
  cacheRead: 0,
9930
10187
  cacheWrite: 0,
9931
10188
  },
9932
- contextWindow: 131072,
9933
- maxTokens: 4096,
10189
+ contextWindow: 262144,
10190
+ maxTokens: 262144,
9934
10191
  },
9935
10192
  "qwen/qwen3-30b-a3b": {
9936
10193
  id: "qwen/qwen3-30b-a3b",
@@ -10063,7 +10320,7 @@ export const MODELS = {
10063
10320
  input: 0.195,
10064
10321
  output: 0.975,
10065
10322
  cacheRead: 0.039,
10066
- cacheWrite: 0,
10323
+ cacheWrite: 0.24375,
10067
10324
  },
10068
10325
  contextWindow: 1000000,
10069
10326
  maxTokens: 65536,
@@ -10077,13 +10334,13 @@ export const MODELS = {
10077
10334
  reasoning: false,
10078
10335
  input: ["text"],
10079
10336
  cost: {
10080
- input: 0.12,
10081
- output: 0.75,
10082
- cacheRead: 0.06,
10337
+ input: 0.15,
10338
+ output: 0.7999999999999999,
10339
+ cacheRead: 0.12,
10083
10340
  cacheWrite: 0,
10084
10341
  },
10085
10342
  contextWindow: 262144,
10086
- maxTokens: 65536,
10343
+ maxTokens: 262144,
10087
10344
  },
10088
10345
  "qwen/qwen3-coder-plus": {
10089
10346
  id: "qwen/qwen3-coder-plus",
@@ -10097,7 +10354,7 @@ export const MODELS = {
10097
10354
  input: 0.65,
10098
10355
  output: 3.25,
10099
10356
  cacheRead: 0.13,
10100
- cacheWrite: 0,
10357
+ cacheWrite: 0.8125,
10101
10358
  },
10102
10359
  contextWindow: 1000000,
10103
10360
  maxTokens: 65536,
@@ -10131,7 +10388,7 @@ export const MODELS = {
10131
10388
  input: 0.78,
10132
10389
  output: 3.9,
10133
10390
  cacheRead: 0.156,
10134
- cacheWrite: 0,
10391
+ cacheWrite: 0.975,
10135
10392
  },
10136
10393
  contextWindow: 262144,
10137
10394
  maxTokens: 32768,
@@ -10385,7 +10642,7 @@ export const MODELS = {
10385
10642
  cost: {
10386
10643
  input: 0.39,
10387
10644
  output: 2.34,
10388
- cacheRead: 0,
10645
+ cacheRead: 0.195,
10389
10646
  cacheWrite: 0,
10390
10647
  },
10391
10648
  contextWindow: 262144,
@@ -10420,7 +10677,7 @@ export const MODELS = {
10420
10677
  input: 0.065,
10421
10678
  output: 0.26,
10422
10679
  cacheRead: 0,
10423
- cacheWrite: 0,
10680
+ cacheWrite: 0.08125,
10424
10681
  },
10425
10682
  contextWindow: 1000000,
10426
10683
  maxTokens: 65536,
@@ -10437,24 +10694,24 @@ export const MODELS = {
10437
10694
  input: 0.26,
10438
10695
  output: 1.56,
10439
10696
  cacheRead: 0,
10440
- cacheWrite: 0,
10697
+ cacheWrite: 0.325,
10441
10698
  },
10442
10699
  contextWindow: 1000000,
10443
10700
  maxTokens: 65536,
10444
10701
  },
10445
- "qwen/qwen3.6-plus:free": {
10446
- id: "qwen/qwen3.6-plus:free",
10447
- name: "Qwen: Qwen3.6 Plus (free)",
10702
+ "qwen/qwen3.6-plus": {
10703
+ id: "qwen/qwen3.6-plus",
10704
+ name: "Qwen: Qwen3.6 Plus",
10448
10705
  api: "openai-completions",
10449
10706
  provider: "openrouter",
10450
10707
  baseUrl: "https://openrouter.ai/api/v1",
10451
10708
  reasoning: true,
10452
10709
  input: ["text", "image"],
10453
10710
  cost: {
10454
- input: 0,
10455
- output: 0,
10711
+ input: 0.325,
10712
+ output: 1.95,
10456
10713
  cacheRead: 0,
10457
- cacheWrite: 0,
10714
+ cacheWrite: 0.40625,
10458
10715
  },
10459
10716
  contextWindow: 1000000,
10460
10717
  maxTokens: 65536,
@@ -10561,23 +10818,6 @@ export const MODELS = {
10561
10818
  contextWindow: 262144,
10562
10819
  maxTokens: 65536,
10563
10820
  },
10564
- "stepfun/step-3.5-flash:free": {
10565
- id: "stepfun/step-3.5-flash:free",
10566
- name: "StepFun: Step 3.5 Flash (free)",
10567
- api: "openai-completions",
10568
- provider: "openrouter",
10569
- baseUrl: "https://openrouter.ai/api/v1",
10570
- reasoning: true,
10571
- input: ["text"],
10572
- cost: {
10573
- input: 0,
10574
- output: 0,
10575
- cacheRead: 0,
10576
- cacheWrite: 0,
10577
- },
10578
- contextWindow: 256000,
10579
- maxTokens: 256000,
10580
- },
10581
10821
  "thedrummer/rocinante-12b": {
10582
10822
  id: "thedrummer/rocinante-12b",
10583
10823
  name: "TheDrummer: Rocinante 12B",
@@ -11037,6 +11277,23 @@ export const MODELS = {
11037
11277
  contextWindow: 202752,
11038
11278
  maxTokens: 131072,
11039
11279
  },
11280
+ "z-ai/glm-5.1": {
11281
+ id: "z-ai/glm-5.1",
11282
+ name: "Z.ai: GLM 5.1",
11283
+ api: "openai-completions",
11284
+ provider: "openrouter",
11285
+ baseUrl: "https://openrouter.ai/api/v1",
11286
+ reasoning: true,
11287
+ input: ["text"],
11288
+ cost: {
11289
+ input: 0.95,
11290
+ output: 3.15,
11291
+ cacheRead: 0.475,
11292
+ cacheWrite: 0,
11293
+ },
11294
+ contextWindow: 202752,
11295
+ maxTokens: 65535,
11296
+ },
11040
11297
  "z-ai/glm-5v-turbo": {
11041
11298
  id: "z-ai/glm-5v-turbo",
11042
11299
  name: "Z.ai: GLM 5V Turbo",
@@ -11082,13 +11339,13 @@ export const MODELS = {
11082
11339
  reasoning: false,
11083
11340
  input: ["text"],
11084
11341
  cost: {
11085
- input: 0.22,
11086
- output: 0.88,
11087
- cacheRead: 0.11,
11342
+ input: 0.6,
11343
+ output: 1.2,
11344
+ cacheRead: 0.6,
11088
11345
  cacheWrite: 0,
11089
11346
  },
11090
- contextWindow: 32768,
11091
- maxTokens: 16384,
11347
+ contextWindow: 131000,
11348
+ maxTokens: 40000,
11092
11349
  },
11093
11350
  "alibaba/qwen-3-30b": {
11094
11351
  id: "alibaba/qwen-3-30b",
@@ -11323,7 +11580,7 @@ export const MODELS = {
11323
11580
  input: 0.5,
11324
11581
  output: 3,
11325
11582
  cacheRead: 0.09999999999999999,
11326
- cacheWrite: 0.625,
11583
+ cacheWrite: 0,
11327
11584
  },
11328
11585
  contextWindow: 1000000,
11329
11586
  maxTokens: 64000,
@@ -11464,6 +11721,23 @@ export const MODELS = {
11464
11721
  contextWindow: 1000000,
11465
11722
  maxTokens: 128000,
11466
11723
  },
11724
+ "anthropic/claude-opus-4.7": {
11725
+ id: "anthropic/claude-opus-4.7",
11726
+ name: "Claude Opus 4.7",
11727
+ api: "anthropic-messages",
11728
+ provider: "vercel-ai-gateway",
11729
+ baseUrl: "https://ai-gateway.vercel.sh",
11730
+ reasoning: true,
11731
+ input: ["text", "image"],
11732
+ cost: {
11733
+ input: 5,
11734
+ output: 25,
11735
+ cacheRead: 0.5,
11736
+ cacheWrite: 6.25,
11737
+ },
11738
+ contextWindow: 1000000,
11739
+ maxTokens: 128000,
11740
+ },
11467
11741
  "anthropic/claude-sonnet-4": {
11468
11742
  id: "anthropic/claude-sonnet-4",
11469
11743
  name: "Claude Sonnet 4",
@@ -11815,7 +12089,7 @@ export const MODELS = {
11815
12089
  cost: {
11816
12090
  input: 0.25,
11817
12091
  output: 1.5,
11818
- cacheRead: 0,
12092
+ cacheRead: 0.03,
11819
12093
  cacheWrite: 0,
11820
12094
  },
11821
12095
  contextWindow: 1000000,
@@ -11938,7 +12212,7 @@ export const MODELS = {
11938
12212
  cacheWrite: 0,
11939
12213
  },
11940
12214
  contextWindow: 128000,
11941
- maxTokens: 8192,
12215
+ maxTokens: 100000,
11942
12216
  },
11943
12217
  "meta/llama-3.1-70b": {
11944
12218
  id: "meta/llama-3.1-70b",
@@ -12943,26 +13217,9 @@ export const MODELS = {
12943
13217
  contextWindow: 1050000,
12944
13218
  maxTokens: 128000,
12945
13219
  },
12946
- "openai/gpt-oss-120b": {
12947
- id: "openai/gpt-oss-120b",
12948
- name: "gpt-oss-120b",
12949
- api: "anthropic-messages",
12950
- provider: "vercel-ai-gateway",
12951
- baseUrl: "https://ai-gateway.vercel.sh",
12952
- reasoning: true,
12953
- input: ["text"],
12954
- cost: {
12955
- input: 0.15,
12956
- output: 0.6,
12957
- cacheRead: 0,
12958
- cacheWrite: 0,
12959
- },
12960
- contextWindow: 128000,
12961
- maxTokens: 8192,
12962
- },
12963
13220
  "openai/gpt-oss-20b": {
12964
13221
  id: "openai/gpt-oss-20b",
12965
- name: "gpt-oss-20b",
13222
+ name: "GPT OSS 120B",
12966
13223
  api: "anthropic-messages",
12967
13224
  provider: "vercel-ai-gateway",
12968
13225
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -12979,7 +13236,7 @@ export const MODELS = {
12979
13236
  },
12980
13237
  "openai/gpt-oss-safeguard-20b": {
12981
13238
  id: "openai/gpt-oss-safeguard-20b",
12982
- name: "gpt-oss-safeguard-20b",
13239
+ name: "GPT OSS Safeguard 20B",
12983
13240
  api: "anthropic-messages",
12984
13241
  provider: "vercel-ai-gateway",
12985
13242
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -13147,23 +13404,6 @@ export const MODELS = {
13147
13404
  contextWindow: 131072,
13148
13405
  maxTokens: 131072,
13149
13406
  },
13150
- "xai/grok-2-vision": {
13151
- id: "xai/grok-2-vision",
13152
- name: "Grok 2 Vision",
13153
- api: "anthropic-messages",
13154
- provider: "vercel-ai-gateway",
13155
- baseUrl: "https://ai-gateway.vercel.sh",
13156
- reasoning: false,
13157
- input: ["text", "image"],
13158
- cost: {
13159
- input: 2,
13160
- output: 10,
13161
- cacheRead: 0,
13162
- cacheWrite: 0,
13163
- },
13164
- contextWindow: 32768,
13165
- maxTokens: 32768,
13166
- },
13167
13407
  "xai/grok-3": {
13168
13408
  id: "xai/grok-3",
13169
13409
  name: "Grok 3 Beta",
@@ -13256,7 +13496,7 @@ export const MODELS = {
13256
13496
  provider: "vercel-ai-gateway",
13257
13497
  baseUrl: "https://ai-gateway.vercel.sh",
13258
13498
  reasoning: false,
13259
- input: ["text"],
13499
+ input: ["text", "image"],
13260
13500
  cost: {
13261
13501
  input: 0.19999999999999998,
13262
13502
  output: 0.5,
@@ -13273,7 +13513,7 @@ export const MODELS = {
13273
13513
  provider: "vercel-ai-gateway",
13274
13514
  baseUrl: "https://ai-gateway.vercel.sh",
13275
13515
  reasoning: true,
13276
- input: ["text"],
13516
+ input: ["text", "image"],
13277
13517
  cost: {
13278
13518
  input: 0.19999999999999998,
13279
13519
  output: 0.5,
@@ -13290,7 +13530,7 @@ export const MODELS = {
13290
13530
  provider: "vercel-ai-gateway",
13291
13531
  baseUrl: "https://ai-gateway.vercel.sh",
13292
13532
  reasoning: false,
13293
- input: ["text"],
13533
+ input: ["text", "image"],
13294
13534
  cost: {
13295
13535
  input: 0.19999999999999998,
13296
13536
  output: 0.5,
@@ -13307,7 +13547,7 @@ export const MODELS = {
13307
13547
  provider: "vercel-ai-gateway",
13308
13548
  baseUrl: "https://ai-gateway.vercel.sh",
13309
13549
  reasoning: true,
13310
- input: ["text"],
13550
+ input: ["text", "image"],
13311
13551
  cost: {
13312
13552
  input: 0.19999999999999998,
13313
13553
  output: 0.5,
@@ -13324,7 +13564,7 @@ export const MODELS = {
13324
13564
  provider: "vercel-ai-gateway",
13325
13565
  baseUrl: "https://ai-gateway.vercel.sh",
13326
13566
  reasoning: true,
13327
- input: ["text"],
13567
+ input: ["text", "image"],
13328
13568
  cost: {
13329
13569
  input: 2,
13330
13570
  output: 6,
@@ -13341,7 +13581,7 @@ export const MODELS = {
13341
13581
  provider: "vercel-ai-gateway",
13342
13582
  baseUrl: "https://ai-gateway.vercel.sh",
13343
13583
  reasoning: true,
13344
- input: ["text"],
13584
+ input: ["text", "image"],
13345
13585
  cost: {
13346
13586
  input: 2,
13347
13587
  output: 6,
@@ -13581,13 +13821,13 @@ export const MODELS = {
13581
13821
  reasoning: true,
13582
13822
  input: ["text"],
13583
13823
  cost: {
13584
- input: 0.6,
13585
- output: 2.2,
13586
- cacheRead: 0.11,
13824
+ input: 2.25,
13825
+ output: 2.75,
13826
+ cacheRead: 2.25,
13587
13827
  cacheWrite: 0,
13588
13828
  },
13589
- contextWindow: 200000,
13590
- maxTokens: 120000,
13829
+ contextWindow: 131000,
13830
+ maxTokens: 40000,
13591
13831
  },
13592
13832
  "zai/glm-4.7-flash": {
13593
13833
  id: "zai/glm-4.7-flash",
@@ -13657,6 +13897,23 @@ export const MODELS = {
13657
13897
  contextWindow: 202800,
13658
13898
  maxTokens: 131100,
13659
13899
  },
13900
+ "zai/glm-5.1": {
13901
+ id: "zai/glm-5.1",
13902
+ name: "GLM 5.1",
13903
+ api: "anthropic-messages",
13904
+ provider: "vercel-ai-gateway",
13905
+ baseUrl: "https://ai-gateway.vercel.sh",
13906
+ reasoning: true,
13907
+ input: ["text", "image"],
13908
+ cost: {
13909
+ input: 1.4,
13910
+ output: 4.4,
13911
+ cacheRead: 0.26,
13912
+ cacheWrite: 0,
13913
+ },
13914
+ contextWindow: 202752,
13915
+ maxTokens: 202752,
13916
+ },
13660
13917
  "zai/glm-5v-turbo": {
13661
13918
  id: "zai/glm-5v-turbo",
13662
13919
  name: "GLM 5V Turbo",
@@ -14284,6 +14541,24 @@ export const MODELS = {
14284
14541
  contextWindow: 200000,
14285
14542
  maxTokens: 131072,
14286
14543
  },
14544
+ "glm-5.1": {
14545
+ id: "glm-5.1",
14546
+ name: "GLM-5.1",
14547
+ api: "openai-completions",
14548
+ provider: "zai",
14549
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
14550
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14551
+ reasoning: true,
14552
+ input: ["text"],
14553
+ cost: {
14554
+ input: 1.4,
14555
+ output: 4.4,
14556
+ cacheRead: 0.26,
14557
+ cacheWrite: 0,
14558
+ },
14559
+ contextWindow: 200000,
14560
+ maxTokens: 131072,
14561
+ },
14287
14562
  "glm-5v-turbo": {
14288
14563
  id: "glm-5v-turbo",
14289
14564
  name: "glm-5v-turbo",