@elyracode/ai 0.9.47 → 0.9.48

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.
@@ -126,6 +126,24 @@ export const MODELS = {
126
126
  contextWindow: 200000,
127
127
  maxTokens: 64000,
128
128
  },
129
+ "anthropic.claude-haiku-5-5": {
130
+ id: "anthropic.claude-haiku-5-5",
131
+ name: "Claude Haiku 5.5",
132
+ api: "bedrock-converse-stream",
133
+ provider: "amazon-bedrock",
134
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
135
+ reasoning: true,
136
+ thinkingType: "adaptive",
137
+ input: ["text", "image"],
138
+ cost: {
139
+ input: 0.1,
140
+ output: 0.5,
141
+ cacheRead: 0.01,
142
+ cacheWrite: 0.125,
143
+ },
144
+ contextWindow: 1000000,
145
+ maxTokens: 128000,
146
+ },
129
147
  "anthropic.claude-opus-4-1-20250805-v1:0": {
130
148
  id: "anthropic.claude-opus-4-1-20250805-v1:0",
131
149
  name: "Claude Opus 4.1",
@@ -321,7 +339,7 @@ export const MODELS = {
321
339
  cost: {
322
340
  input: 2,
323
341
  output: 10,
324
- cacheRead: 0.2,
342
+ cacheRead: 0.1,
325
343
  cacheWrite: 2.5,
326
344
  },
327
345
  contextWindow: 1000000,
@@ -414,6 +432,24 @@ export const MODELS = {
414
432
  contextWindow: 200000,
415
433
  maxTokens: 64000,
416
434
  },
435
+ "au.anthropic.claude-haiku-5-5": {
436
+ id: "au.anthropic.claude-haiku-5-5",
437
+ name: "Claude Haiku 5.5 (AU)",
438
+ api: "bedrock-converse-stream",
439
+ provider: "amazon-bedrock",
440
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
441
+ reasoning: true,
442
+ thinkingType: "adaptive",
443
+ input: ["text", "image"],
444
+ cost: {
445
+ input: 0.11,
446
+ output: 0.55,
447
+ cacheRead: 0.011,
448
+ cacheWrite: 0.1375,
449
+ },
450
+ contextWindow: 1000000,
451
+ maxTokens: 128000,
452
+ },
417
453
  "au.anthropic.claude-opus-4-6-v1": {
418
454
  id: "au.anthropic.claude-opus-4-6-v1",
419
455
  name: "AU Anthropic Claude Opus 4.6",
@@ -717,6 +753,24 @@ export const MODELS = {
717
753
  contextWindow: 200000,
718
754
  maxTokens: 64000,
719
755
  },
756
+ "eu.anthropic.claude-haiku-5-5": {
757
+ id: "eu.anthropic.claude-haiku-5-5",
758
+ name: "Claude Haiku 5.5 (EU)",
759
+ api: "bedrock-converse-stream",
760
+ provider: "amazon-bedrock",
761
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
762
+ reasoning: true,
763
+ thinkingType: "adaptive",
764
+ input: ["text", "image"],
765
+ cost: {
766
+ input: 0.11,
767
+ output: 0.55,
768
+ cacheRead: 0.011,
769
+ cacheWrite: 0.1375,
770
+ },
771
+ contextWindow: 1000000,
772
+ maxTokens: 128000,
773
+ },
720
774
  "eu.anthropic.claude-opus-4-5-20251101-v1:0": {
721
775
  id: "eu.anthropic.claude-opus-4-5-20251101-v1:0",
722
776
  name: "Claude Opus 4.5 (EU)",
@@ -900,6 +954,24 @@ export const MODELS = {
900
954
  contextWindow: 1000000,
901
955
  maxTokens: 128000,
902
956
  },
957
+ "eu.anthropic.claude-sonnet-5-5": {
958
+ id: "eu.anthropic.claude-sonnet-5-5",
959
+ name: "Claude Sonnet 5.5 (EU)",
960
+ api: "bedrock-converse-stream",
961
+ provider: "amazon-bedrock",
962
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
963
+ reasoning: true,
964
+ thinkingType: "adaptive",
965
+ input: ["text", "image"],
966
+ cost: {
967
+ input: 2.2,
968
+ output: 11,
969
+ cacheRead: 0.11,
970
+ cacheWrite: 2.75,
971
+ },
972
+ contextWindow: 1000000,
973
+ maxTokens: 128000,
974
+ },
903
975
  "eu.mistral.pixtral-large-2502-v1:0": {
904
976
  id: "eu.mistral.pixtral-large-2502-v1:0",
905
977
  name: "Pixtral Large (25.02) (EU)",
@@ -990,6 +1062,24 @@ export const MODELS = {
990
1062
  contextWindow: 200000,
991
1063
  maxTokens: 64000,
992
1064
  },
1065
+ "global.anthropic.claude-haiku-5-5": {
1066
+ id: "global.anthropic.claude-haiku-5-5",
1067
+ name: "Claude Haiku 5.5 (Global)",
1068
+ api: "bedrock-converse-stream",
1069
+ provider: "amazon-bedrock",
1070
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1071
+ reasoning: true,
1072
+ thinkingType: "adaptive",
1073
+ input: ["text", "image"],
1074
+ cost: {
1075
+ input: 0.1,
1076
+ output: 0.5,
1077
+ cacheRead: 0.01,
1078
+ cacheWrite: 0.125,
1079
+ },
1080
+ contextWindow: 1000000,
1081
+ maxTokens: 128000,
1082
+ },
993
1083
  "global.anthropic.claude-opus-4-5-20251101-v1:0": {
994
1084
  id: "global.anthropic.claude-opus-4-5-20251101-v1:0",
995
1085
  name: "Claude Opus 4.5 (Global)",
@@ -1185,7 +1275,7 @@ export const MODELS = {
1185
1275
  cost: {
1186
1276
  input: 2,
1187
1277
  output: 10,
1188
- cacheRead: 0.2,
1278
+ cacheRead: 0.1,
1189
1279
  cacheWrite: 2.5,
1190
1280
  },
1191
1281
  contextWindow: 1000000,
@@ -1361,6 +1451,23 @@ export const MODELS = {
1361
1451
  contextWindow: 500000,
1362
1452
  maxTokens: 500000,
1363
1453
  },
1454
+ "global.zai.glm-5.3": {
1455
+ id: "global.zai.glm-5.3",
1456
+ name: "GLM-5.3 (Global)",
1457
+ api: "bedrock-converse-stream",
1458
+ provider: "amazon-bedrock",
1459
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1460
+ reasoning: true,
1461
+ input: ["text"],
1462
+ cost: {
1463
+ input: 1.68,
1464
+ output: 5.28,
1465
+ cacheRead: 0.312,
1466
+ cacheWrite: 2.1,
1467
+ },
1468
+ contextWindow: 1000000,
1469
+ maxTokens: 128000,
1470
+ },
1364
1471
  "google.gemma-4-26b-a4b": {
1365
1472
  id: "google.gemma-4-26b-a4b",
1366
1473
  name: "Gemma 4 26B A4B IT",
@@ -1535,6 +1642,24 @@ export const MODELS = {
1535
1642
  contextWindow: 200000,
1536
1643
  maxTokens: 64000,
1537
1644
  },
1645
+ "jp.anthropic.claude-haiku-5-5": {
1646
+ id: "jp.anthropic.claude-haiku-5-5",
1647
+ name: "Claude Haiku 5.5 (JP)",
1648
+ api: "bedrock-converse-stream",
1649
+ provider: "amazon-bedrock",
1650
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1651
+ reasoning: true,
1652
+ thinkingType: "adaptive",
1653
+ input: ["text", "image"],
1654
+ cost: {
1655
+ input: 0.11,
1656
+ output: 0.55,
1657
+ cacheRead: 0.011,
1658
+ cacheWrite: 0.1375,
1659
+ },
1660
+ contextWindow: 1000000,
1661
+ maxTokens: 128000,
1662
+ },
1538
1663
  "jp.anthropic.claude-opus-4-7": {
1539
1664
  id: "jp.anthropic.claude-opus-4-7",
1540
1665
  name: "Claude Opus 4.7 (JP)",
@@ -2605,6 +2730,24 @@ export const MODELS = {
2605
2730
  contextWindow: 200000,
2606
2731
  maxTokens: 64000,
2607
2732
  },
2733
+ "us.anthropic.claude-haiku-5-5": {
2734
+ id: "us.anthropic.claude-haiku-5-5",
2735
+ name: "Claude Haiku 5.5 (US)",
2736
+ api: "bedrock-converse-stream",
2737
+ provider: "amazon-bedrock",
2738
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2739
+ reasoning: true,
2740
+ thinkingType: "adaptive",
2741
+ input: ["text", "image"],
2742
+ cost: {
2743
+ input: 0.11,
2744
+ output: 0.55,
2745
+ cacheRead: 0.011,
2746
+ cacheWrite: 0.1375,
2747
+ },
2748
+ contextWindow: 1000000,
2749
+ maxTokens: 128000,
2750
+ },
2608
2751
  "us.anthropic.claude-opus-4-1-20250805-v1:0": {
2609
2752
  id: "us.anthropic.claude-opus-4-1-20250805-v1:0",
2610
2753
  name: "Claude Opus 4.1 (US)",
@@ -2806,6 +2949,24 @@ export const MODELS = {
2806
2949
  contextWindow: 1000000,
2807
2950
  maxTokens: 128000,
2808
2951
  },
2952
+ "us.anthropic.claude-sonnet-5-5": {
2953
+ id: "us.anthropic.claude-sonnet-5-5",
2954
+ name: "Claude Sonnet 5.5 (US)",
2955
+ api: "bedrock-converse-stream",
2956
+ provider: "amazon-bedrock",
2957
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2958
+ reasoning: true,
2959
+ thinkingType: "adaptive",
2960
+ input: ["text", "image"],
2961
+ cost: {
2962
+ input: 2.2,
2963
+ output: 11,
2964
+ cacheRead: 0.11,
2965
+ cacheWrite: 2.75,
2966
+ },
2967
+ contextWindow: 1000000,
2968
+ maxTokens: 128000,
2969
+ },
2809
2970
  "us.meta.llama3-1-70b-instruct-v1:0": {
2810
2971
  id: "us.meta.llama3-1-70b-instruct-v1:0",
2811
2972
  name: "Llama 3.1 70B Instruct (US)",
@@ -3112,6 +3273,23 @@ export const MODELS = {
3112
3273
  contextWindow: 500000,
3113
3274
  maxTokens: 500000,
3114
3275
  },
3276
+ "us.zai.glm-5.3": {
3277
+ id: "us.zai.glm-5.3",
3278
+ name: "GLM-5.3 (US)",
3279
+ api: "bedrock-converse-stream",
3280
+ provider: "amazon-bedrock",
3281
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
3282
+ reasoning: true,
3283
+ input: ["text"],
3284
+ cost: {
3285
+ input: 1.848,
3286
+ output: 5.808,
3287
+ cacheRead: 0.3432,
3288
+ cacheWrite: 2.31,
3289
+ },
3290
+ contextWindow: 1000000,
3291
+ maxTokens: 128000,
3292
+ },
3115
3293
  "writer.palmyra-x4-v1:0": {
3116
3294
  id: "writer.palmyra-x4-v1:0",
3117
3295
  name: "Palmyra X4",
@@ -3307,6 +3485,24 @@ export const MODELS = {
3307
3485
  contextWindow: 200000,
3308
3486
  maxTokens: 64000,
3309
3487
  },
3488
+ "claude-haiku-5-5": {
3489
+ id: "claude-haiku-5-5",
3490
+ name: "Claude Haiku 5.5",
3491
+ api: "anthropic-messages",
3492
+ provider: "anthropic",
3493
+ baseUrl: "https://api.anthropic.com",
3494
+ reasoning: true,
3495
+ thinkingType: "adaptive",
3496
+ input: ["text", "image"],
3497
+ cost: {
3498
+ input: 0.1,
3499
+ output: 0.5,
3500
+ cacheRead: 0.01,
3501
+ cacheWrite: 0.125,
3502
+ },
3503
+ contextWindow: 1000000,
3504
+ maxTokens: 128000,
3505
+ },
3310
3506
  "claude-opus-4-5": {
3311
3507
  id: "claude-opus-4-5",
3312
3508
  name: "Claude Opus 4.5 (latest)",
@@ -3451,7 +3647,7 @@ export const MODELS = {
3451
3647
  cacheRead: 0.3,
3452
3648
  cacheWrite: 3.75,
3453
3649
  },
3454
- contextWindow: 1000000,
3650
+ contextWindow: 200000,
3455
3651
  maxTokens: 64000,
3456
3652
  },
3457
3653
  "claude-sonnet-4-5-20250929": {
@@ -3469,7 +3665,7 @@ export const MODELS = {
3469
3665
  cacheRead: 0.3,
3470
3666
  cacheWrite: 3.75,
3471
3667
  },
3472
- contextWindow: 1000000,
3668
+ contextWindow: 200000,
3473
3669
  maxTokens: 64000,
3474
3670
  },
3475
3671
  "claude-sonnet-4-6": {
@@ -3520,7 +3716,7 @@ export const MODELS = {
3520
3716
  cost: {
3521
3717
  input: 2,
3522
3718
  output: 10,
3523
- cacheRead: 0.2,
3719
+ cacheRead: 0.1,
3524
3720
  cacheWrite: 2.5,
3525
3721
  },
3526
3722
  contextWindow: 1000000,
@@ -4530,6 +4726,23 @@ export const MODELS = {
4530
4726
  contextWindow: 1000000,
4531
4727
  maxTokens: 128000,
4532
4728
  },
4729
+ "claude-sonnet-5.5": {
4730
+ id: "claude-sonnet-5.5",
4731
+ name: "Claude Sonnet 5.5",
4732
+ api: "anthropic-messages",
4733
+ provider: "cloudflare-ai-gateway",
4734
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
4735
+ reasoning: true,
4736
+ input: ["text", "image"],
4737
+ cost: {
4738
+ input: 2,
4739
+ output: 10,
4740
+ cacheRead: 0.2,
4741
+ cacheWrite: 2.5,
4742
+ },
4743
+ contextWindow: 1000000,
4744
+ maxTokens: 128000,
4745
+ },
4533
4746
  "gpt-4.1": {
4534
4747
  id: "gpt-4.1",
4535
4748
  name: "GPT-4.1",
@@ -4900,6 +5113,23 @@ export const MODELS = {
4900
5113
  contextWindow: 1050000,
4901
5114
  maxTokens: 128000,
4902
5115
  },
5116
+ "gpt-6.1-sol": {
5117
+ id: "gpt-6.1-sol",
5118
+ name: "GPT-6.1 Sol",
5119
+ api: "openai-responses",
5120
+ provider: "cloudflare-ai-gateway",
5121
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
5122
+ reasoning: true,
5123
+ input: ["text", "image"],
5124
+ cost: {
5125
+ input: 2,
5126
+ output: 10,
5127
+ cacheRead: 0.1,
5128
+ cacheWrite: 2.5,
5129
+ },
5130
+ contextWindow: 1050000,
5131
+ maxTokens: 128000,
5132
+ },
4903
5133
  "o3": {
4904
5134
  id: "o3",
4905
5135
  name: "o3",
@@ -5756,6 +5986,25 @@ export const MODELS = {
5756
5986
  contextWindow: 200000,
5757
5987
  maxTokens: 64000,
5758
5988
  },
5989
+ "claude-haiku-5.5": {
5990
+ id: "claude-haiku-5.5",
5991
+ name: "Claude Haiku 5.5",
5992
+ api: "openai-completions",
5993
+ provider: "github-copilot",
5994
+ baseUrl: "https://api.individual.githubcopilot.com",
5995
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
5996
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
5997
+ reasoning: true,
5998
+ input: ["text", "image"],
5999
+ cost: {
6000
+ input: 0.1,
6001
+ output: 0.5,
6002
+ cacheRead: 0.01,
6003
+ cacheWrite: 0.125,
6004
+ },
6005
+ contextWindow: 1000000,
6006
+ maxTokens: 128000,
6007
+ },
5759
6008
  "claude-opus-4.7": {
5760
6009
  id: "claude-opus-4.7",
5761
6010
  name: "Claude Opus 4.7",
@@ -5882,7 +6131,7 @@ export const MODELS = {
5882
6131
  cost: {
5883
6132
  input: 2,
5884
6133
  output: 10,
5885
- cacheRead: 0.2,
6134
+ cacheRead: 0.1,
5886
6135
  cacheWrite: 2.5,
5887
6136
  },
5888
6137
  contextWindow: 1000000,
@@ -8616,6 +8865,23 @@ export const MODELS = {
8616
8865
  },
8617
8866
  },
8618
8867
  "mistral": {
8868
+ "codestral-2508": {
8869
+ id: "codestral-2508",
8870
+ name: "Codestral 25.08",
8871
+ api: "mistral-conversations",
8872
+ provider: "mistral",
8873
+ baseUrl: "https://api.mistral.ai",
8874
+ reasoning: false,
8875
+ input: ["text"],
8876
+ cost: {
8877
+ input: 0.3,
8878
+ output: 0.9,
8879
+ cacheRead: 0.03,
8880
+ cacheWrite: 0,
8881
+ },
8882
+ contextWindow: 256000,
8883
+ maxTokens: 8192,
8884
+ },
8619
8885
  "codestral-latest": {
8620
8886
  id: "codestral-latest",
8621
8887
  name: "Codestral (latest)",
@@ -8735,6 +9001,23 @@ export const MODELS = {
8735
9001
  contextWindow: 128000,
8736
9002
  maxTokens: 128000,
8737
9003
  },
9004
+ "glm-5-2": {
9005
+ id: "glm-5-2",
9006
+ name: "GLM-5.2",
9007
+ api: "mistral-conversations",
9008
+ provider: "mistral",
9009
+ baseUrl: "https://api.mistral.ai",
9010
+ reasoning: true,
9011
+ input: ["text"],
9012
+ cost: {
9013
+ input: 1.4,
9014
+ output: 4.4,
9015
+ cacheRead: 0.14,
9016
+ cacheWrite: 0,
9017
+ },
9018
+ contextWindow: 1048576,
9019
+ maxTokens: 131072,
9020
+ },
8738
9021
  "labs-devstral-small-2512": {
8739
9022
  id: "labs-devstral-small-2512",
8740
9023
  name: "Devstral Small 2",
@@ -8752,23 +9035,74 @@ export const MODELS = {
8752
9035
  contextWindow: 256000,
8753
9036
  maxTokens: 256000,
8754
9037
  },
8755
- "magistral-medium-latest": {
8756
- id: "magistral-medium-latest",
8757
- name: "Magistral Medium (latest)",
9038
+ "labs-leanstral-1-5-1": {
9039
+ id: "labs-leanstral-1-5-1",
9040
+ name: "Leanstral 1.5",
8758
9041
  api: "mistral-conversations",
8759
9042
  provider: "mistral",
8760
9043
  baseUrl: "https://api.mistral.ai",
8761
9044
  reasoning: true,
8762
- input: ["text"],
9045
+ input: ["text", "image"],
8763
9046
  cost: {
8764
- input: 2,
8765
- output: 5,
9047
+ input: 0,
9048
+ output: 0,
8766
9049
  cacheRead: 0,
8767
9050
  cacheWrite: 0,
8768
9051
  },
8769
- contextWindow: 128000,
9052
+ contextWindow: 262144,
9053
+ maxTokens: 128000,
9054
+ },
9055
+ "magistral-medium-latest": {
9056
+ id: "magistral-medium-latest",
9057
+ name: "Magistral Medium (latest)",
9058
+ api: "mistral-conversations",
9059
+ provider: "mistral",
9060
+ baseUrl: "https://api.mistral.ai",
9061
+ reasoning: true,
9062
+ input: ["text", "image"],
9063
+ cost: {
9064
+ input: 2,
9065
+ output: 5,
9066
+ cacheRead: 0,
9067
+ cacheWrite: 0,
9068
+ },
9069
+ contextWindow: 262144,
8770
9070
  maxTokens: 16384,
8771
9071
  },
9072
+ "ministral-14b-2512": {
9073
+ id: "ministral-14b-2512",
9074
+ name: "Ministral 3 14B",
9075
+ api: "mistral-conversations",
9076
+ provider: "mistral",
9077
+ baseUrl: "https://api.mistral.ai",
9078
+ reasoning: false,
9079
+ input: ["text", "image"],
9080
+ cost: {
9081
+ input: 0.2,
9082
+ output: 0.2,
9083
+ cacheRead: 0.02,
9084
+ cacheWrite: 0,
9085
+ },
9086
+ contextWindow: 262144,
9087
+ maxTokens: 262144,
9088
+ },
9089
+ "ministral-3b-2512": {
9090
+ id: "ministral-3b-2512",
9091
+ name: "Ministral 3 3B",
9092
+ api: "mistral-conversations",
9093
+ provider: "mistral",
9094
+ baseUrl: "https://api.mistral.ai",
9095
+ reasoning: false,
9096
+ input: ["text", "image"],
9097
+ cost: {
9098
+ input: 0.1,
9099
+ output: 0.1,
9100
+ cacheRead: 0.01,
9101
+ cacheWrite: 0,
9102
+ },
9103
+ contextWindow: 131072,
9104
+ maxTokens: 262144,
9105
+ },
8772
9106
  "ministral-3b-latest": {
8773
9107
  id: "ministral-3b-latest",
8774
9108
  name: "Ministral 3B (latest)",
@@ -8776,16 +9110,33 @@ export const MODELS = {
8776
9110
  provider: "mistral",
8777
9111
  baseUrl: "https://api.mistral.ai",
8778
9112
  reasoning: false,
8779
- input: ["text"],
9113
+ input: ["text", "image"],
8780
9114
  cost: {
8781
9115
  input: 0.04,
8782
9116
  output: 0.04,
8783
9117
  cacheRead: 0,
8784
9118
  cacheWrite: 0,
8785
9119
  },
8786
- contextWindow: 128000,
9120
+ contextWindow: 131072,
8787
9121
  maxTokens: 128000,
8788
9122
  },
9123
+ "ministral-8b-2512": {
9124
+ id: "ministral-8b-2512",
9125
+ name: "Ministral 3 8B",
9126
+ api: "mistral-conversations",
9127
+ provider: "mistral",
9128
+ baseUrl: "https://api.mistral.ai",
9129
+ reasoning: false,
9130
+ input: ["text", "image"],
9131
+ cost: {
9132
+ input: 0.15,
9133
+ output: 0.15,
9134
+ cacheRead: 0.015,
9135
+ cacheWrite: 0,
9136
+ },
9137
+ contextWindow: 262144,
9138
+ maxTokens: 262144,
9139
+ },
8789
9140
  "ministral-8b-latest": {
8790
9141
  id: "ministral-8b-latest",
8791
9142
  name: "Ministral 8B (latest)",
@@ -8793,14 +9144,14 @@ export const MODELS = {
8793
9144
  provider: "mistral",
8794
9145
  baseUrl: "https://api.mistral.ai",
8795
9146
  reasoning: false,
8796
- input: ["text"],
9147
+ input: ["text", "image"],
8797
9148
  cost: {
8798
9149
  input: 0.1,
8799
9150
  output: 0.1,
8800
9151
  cacheRead: 0,
8801
9152
  cacheWrite: 0,
8802
9153
  },
8803
- contextWindow: 128000,
9154
+ contextWindow: 262144,
8804
9155
  maxTokens: 128000,
8805
9156
  },
8806
9157
  "mistral-large-2411": {
@@ -8837,6 +9188,23 @@ export const MODELS = {
8837
9188
  contextWindow: 262144,
8838
9189
  maxTokens: 262144,
8839
9190
  },
9191
+ "mistral-large-4": {
9192
+ id: "mistral-large-4",
9193
+ name: "Mistral Large 4",
9194
+ api: "mistral-conversations",
9195
+ provider: "mistral",
9196
+ baseUrl: "https://api.mistral.ai",
9197
+ reasoning: true,
9198
+ input: ["text", "image"],
9199
+ cost: {
9200
+ input: 0.68,
9201
+ output: 2.09,
9202
+ cacheRead: 0.07,
9203
+ cacheWrite: 0,
9204
+ },
9205
+ contextWindow: 524288,
9206
+ maxTokens: 262144,
9207
+ },
8840
9208
  "mistral-large-latest": {
8841
9209
  id: "mistral-large-latest",
8842
9210
  name: "Mistral Large (latest)",
@@ -8987,7 +9355,7 @@ export const MODELS = {
8987
9355
  cacheRead: 0.015,
8988
9356
  cacheWrite: 0,
8989
9357
  },
8990
- contextWindow: 256000,
9358
+ contextWindow: 262144,
8991
9359
  maxTokens: 256000,
8992
9360
  },
8993
9361
  "mistral-small-latest": {
@@ -9004,7 +9372,7 @@ export const MODELS = {
9004
9372
  cacheRead: 0.015,
9005
9373
  cacheWrite: 0,
9006
9374
  },
9007
- contextWindow: 256000,
9375
+ contextWindow: 262144,
9008
9376
  maxTokens: 256000,
9009
9377
  },
9010
9378
  "open-mistral-7b": {
@@ -9109,6 +9477,23 @@ export const MODELS = {
9109
9477
  contextWindow: 128000,
9110
9478
  maxTokens: 128000,
9111
9479
  },
9480
+ "voxtral-small-2507": {
9481
+ id: "voxtral-small-2507",
9482
+ name: "Voxtral Small",
9483
+ api: "mistral-conversations",
9484
+ provider: "mistral",
9485
+ baseUrl: "https://api.mistral.ai",
9486
+ reasoning: false,
9487
+ input: ["text"],
9488
+ cost: {
9489
+ input: 0.1,
9490
+ output: 0.4,
9491
+ cacheRead: 0,
9492
+ cacheWrite: 0,
9493
+ },
9494
+ contextWindow: 32768,
9495
+ maxTokens: 32000,
9496
+ },
9112
9497
  "voxtral-small-latest": {
9113
9498
  id: "voxtral-small-latest",
9114
9499
  name: "Voxtral Small (latest)",
@@ -9123,7 +9508,7 @@ export const MODELS = {
9123
9508
  cacheRead: 0,
9124
9509
  cacheWrite: 0,
9125
9510
  },
9126
- contextWindow: 32000,
9511
+ contextWindow: 32768,
9127
9512
  maxTokens: 32000,
9128
9513
  },
9129
9514
  "zai-glm-5-2": {
@@ -9140,7 +9525,7 @@ export const MODELS = {
9140
9525
  cacheRead: 0.14,
9141
9526
  cacheWrite: 0,
9142
9527
  },
9143
- contextWindow: 1000000,
9528
+ contextWindow: 1048576,
9144
9529
  maxTokens: 131072,
9145
9530
  },
9146
9531
  "zai-glm-5-3": {
@@ -9157,7 +9542,7 @@ export const MODELS = {
9157
9542
  cacheRead: 0.14,
9158
9543
  cacheWrite: 0,
9159
9544
  },
9160
- contextWindow: 1000000,
9545
+ contextWindow: 1048576,
9161
9546
  maxTokens: 131072,
9162
9547
  },
9163
9548
  },
@@ -9229,7 +9614,7 @@ export const MODELS = {
9229
9614
  input: 3,
9230
9615
  output: 15,
9231
9616
  cacheRead: 0.3,
9232
- cacheWrite: 0,
9617
+ cacheWrite: 3,
9233
9618
  },
9234
9619
  contextWindow: 1048576,
9235
9620
  maxTokens: 1048576,
@@ -9303,7 +9688,7 @@ export const MODELS = {
9303
9688
  input: 3,
9304
9689
  output: 15,
9305
9690
  cacheRead: 0.3,
9306
- cacheWrite: 0,
9691
+ cacheWrite: 3,
9307
9692
  },
9308
9693
  contextWindow: 1048576,
9309
9694
  maxTokens: 1048576,
@@ -10317,6 +10702,23 @@ export const MODELS = {
10317
10702
  contextWindow: 200000,
10318
10703
  maxTokens: 64000,
10319
10704
  },
10705
+ "claude-haiku-5-5": {
10706
+ id: "claude-haiku-5-5",
10707
+ name: "Claude Haiku 5.5",
10708
+ api: "anthropic-messages",
10709
+ provider: "opencode",
10710
+ baseUrl: "https://opencode.ai/zen",
10711
+ reasoning: true,
10712
+ input: ["text", "image"],
10713
+ cost: {
10714
+ input: 0.1,
10715
+ output: 0.5,
10716
+ cacheRead: 0.01,
10717
+ cacheWrite: 0.125,
10718
+ },
10719
+ contextWindow: 1000000,
10720
+ maxTokens: 128000,
10721
+ },
10320
10722
  "claude-opus-4-5": {
10321
10723
  id: "claude-opus-4-5",
10322
10724
  name: "Claude Opus 4.5",
@@ -10583,6 +10985,23 @@ export const MODELS = {
10583
10985
  contextWindow: 1000000,
10584
10986
  maxTokens: 384000,
10585
10987
  },
10988
+ "exo-free": {
10989
+ id: "exo-free",
10990
+ name: "Exo Free",
10991
+ api: "openai-completions",
10992
+ provider: "opencode",
10993
+ baseUrl: "https://opencode.ai/zen/v1",
10994
+ reasoning: true,
10995
+ input: ["text", "image"],
10996
+ cost: {
10997
+ input: 0,
10998
+ output: 0,
10999
+ cacheRead: 0,
11000
+ cacheWrite: 0,
11001
+ },
11002
+ contextWindow: 1048576,
11003
+ maxTokens: 131072,
11004
+ },
10586
11005
  "fledge-alpha-free": {
10587
11006
  id: "fledge-alpha-free",
10588
11007
  name: "Fledge Alpha Free",
@@ -11476,6 +11895,23 @@ export const MODELS = {
11476
11895
  contextWindow: 512000,
11477
11896
  maxTokens: 128000,
11478
11897
  },
11898
+ "mistral-large-4": {
11899
+ id: "mistral-large-4",
11900
+ name: "Mistral Large 4",
11901
+ api: "openai-completions",
11902
+ provider: "opencode",
11903
+ baseUrl: "https://opencode.ai/zen/v1",
11904
+ reasoning: true,
11905
+ input: ["text", "image"],
11906
+ cost: {
11907
+ input: 0.68,
11908
+ output: 2.09,
11909
+ cacheRead: 0.07,
11910
+ cacheWrite: 0,
11911
+ },
11912
+ contextWindow: 524288,
11913
+ maxTokens: 262144,
11914
+ },
11479
11915
  "muse-spark-1.2": {
11480
11916
  id: "muse-spark-1.2",
11481
11917
  name: "Muse Spark 1.2",
@@ -11648,6 +12084,23 @@ export const MODELS = {
11648
12084
  },
11649
12085
  },
11650
12086
  "opencode-go": {
12087
+ "claude-haiku-5-5": {
12088
+ id: "claude-haiku-5-5",
12089
+ name: "Claude Haiku 5.5",
12090
+ api: "anthropic-messages",
12091
+ provider: "opencode-go",
12092
+ baseUrl: "https://opencode.ai/zen/go",
12093
+ reasoning: true,
12094
+ input: ["text", "image"],
12095
+ cost: {
12096
+ input: 0.1,
12097
+ output: 0.5,
12098
+ cacheRead: 0.01,
12099
+ cacheWrite: 0.125,
12100
+ },
12101
+ contextWindow: 1000000,
12102
+ maxTokens: 128000,
12103
+ },
11651
12104
  "deepseek-v4-flash": {
11652
12105
  id: "deepseek-v4-flash",
11653
12106
  name: "DeepSeek V4 Flash",
@@ -12085,9 +12538,9 @@ export const MODELS = {
12085
12538
  "qwen3.7-plus": {
12086
12539
  id: "qwen3.7-plus",
12087
12540
  name: "Qwen3.7 Plus",
12088
- api: "openai-completions",
12541
+ api: "anthropic-messages",
12089
12542
  provider: "opencode-go",
12090
- baseUrl: "https://opencode.ai/zen/go/v1",
12543
+ baseUrl: "https://opencode.ai/zen/go",
12091
12544
  reasoning: true,
12092
12545
  input: ["text", "image"],
12093
12546
  cost: {
@@ -12119,9 +12572,9 @@ export const MODELS = {
12119
12572
  "qwen3.8-max": {
12120
12573
  id: "qwen3.8-max",
12121
12574
  name: "Qwen3.8 Max",
12122
- api: "openai-completions",
12575
+ api: "anthropic-messages",
12123
12576
  provider: "opencode-go",
12124
- baseUrl: "https://opencode.ai/zen/go/v1",
12577
+ baseUrl: "https://opencode.ai/zen/go",
12125
12578
  reasoning: true,
12126
12579
  input: ["text", "image"],
12127
12580
  cost: {
@@ -12133,18 +12586,18 @@ export const MODELS = {
12133
12586
  contextWindow: 1000000,
12134
12587
  maxTokens: 131072,
12135
12588
  },
12136
- "space-bunny-free": {
12137
- id: "space-bunny-free",
12138
- name: "Space Bunny Free",
12589
+ "space-bunny": {
12590
+ id: "space-bunny",
12591
+ name: "Space Bunny",
12139
12592
  api: "openai-completions",
12140
12593
  provider: "opencode-go",
12141
12594
  baseUrl: "https://opencode.ai/zen/go/v1",
12142
12595
  reasoning: true,
12143
12596
  input: ["text", "image"],
12144
12597
  cost: {
12145
- input: 0,
12146
- output: 0,
12147
- cacheRead: 0,
12598
+ input: 0.15,
12599
+ output: 0.6,
12600
+ cacheRead: 0.03,
12148
12601
  cacheWrite: 0,
12149
12602
  },
12150
12603
  contextWindow: 1048576,
@@ -12428,6 +12881,40 @@ export const MODELS = {
12428
12881
  contextWindow: 200000,
12429
12882
  maxTokens: 64000,
12430
12883
  },
12884
+ "anthropic/claude-haiku-5.5": {
12885
+ id: "anthropic/claude-haiku-5.5",
12886
+ name: "Anthropic: Claude Haiku 5.5",
12887
+ api: "openai-completions",
12888
+ provider: "openrouter",
12889
+ baseUrl: "https://openrouter.ai/api/v1",
12890
+ reasoning: true,
12891
+ input: ["text", "image"],
12892
+ cost: {
12893
+ input: 0.09999999999999999,
12894
+ output: 0.5,
12895
+ cacheRead: 0.01,
12896
+ cacheWrite: 0.125,
12897
+ },
12898
+ contextWindow: 1000000,
12899
+ maxTokens: 128000,
12900
+ },
12901
+ "anthropic/claude-haiku-5.5:batch": {
12902
+ id: "anthropic/claude-haiku-5.5:batch",
12903
+ name: "Anthropic: Claude Haiku 5.5 (batch)",
12904
+ api: "openai-completions",
12905
+ provider: "openrouter",
12906
+ baseUrl: "https://openrouter.ai/api/v1",
12907
+ reasoning: true,
12908
+ input: ["text", "image"],
12909
+ cost: {
12910
+ input: 0.049999999999999996,
12911
+ output: 0.25,
12912
+ cacheRead: 0.005,
12913
+ cacheWrite: 0.0625,
12914
+ },
12915
+ contextWindow: 1000000,
12916
+ maxTokens: 128000,
12917
+ },
12431
12918
  "anthropic/claude-opus-4.1": {
12432
12919
  id: "anthropic/claude-opus-4.1",
12433
12920
  name: "Anthropic: Claude Opus 4.1",
@@ -12785,7 +13272,7 @@ export const MODELS = {
12785
13272
  cost: {
12786
13273
  input: 2,
12787
13274
  output: 10,
12788
- cacheRead: 0.19999999999999998,
13275
+ cacheRead: 0.09999999999999999,
12789
13276
  cacheWrite: 2.5,
12790
13277
  },
12791
13278
  contextWindow: 1000000,
@@ -12802,7 +13289,7 @@ export const MODELS = {
12802
13289
  cost: {
12803
13290
  input: 1,
12804
13291
  output: 5,
12805
- cacheRead: 0.09999999999999999,
13292
+ cacheRead: 0.049999999999999996,
12806
13293
  cacheWrite: 1.25,
12807
13294
  },
12808
13295
  contextWindow: 1000000,
@@ -13157,13 +13644,13 @@ export const MODELS = {
13157
13644
  reasoning: true,
13158
13645
  input: ["text"],
13159
13646
  cost: {
13160
- input: 0.28,
13647
+ input: 0.25899999999999995,
13161
13648
  output: 0.42,
13162
- cacheRead: 0.028,
13649
+ cacheRead: 0.135,
13163
13650
  cacheWrite: 0,
13164
13651
  },
13165
13652
  contextWindow: 163840,
13166
- maxTokens: 65536,
13653
+ maxTokens: 147456,
13167
13654
  },
13168
13655
  "deepseek/deepseek-v3.2-exp": {
13169
13656
  id: "deepseek/deepseek-v3.2-exp",
@@ -13193,13 +13680,13 @@ export const MODELS = {
13193
13680
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
13194
13681
  input: ["text"],
13195
13682
  cost: {
13196
- input: 0.028,
13197
- output: 0.056,
13198
- cacheRead: 0.0056,
13683
+ input: 0.009699999999999999,
13684
+ output: 1.28,
13685
+ cacheRead: 0.009699999999999999,
13199
13686
  cacheWrite: 0,
13200
13687
  },
13201
13688
  contextWindow: 1048576,
13202
- maxTokens: 384000,
13689
+ maxTokens: 943718,
13203
13690
  },
13204
13691
  "deepseek/deepseek-v4-flash-0731": {
13205
13692
  id: "deepseek/deepseek-v4-flash-0731",
@@ -13212,9 +13699,9 @@ export const MODELS = {
13212
13699
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
13213
13700
  input: ["text"],
13214
13701
  cost: {
13215
- input: 0.0188,
13702
+ input: 0.010400000000000001,
13216
13703
  output: 1.28,
13217
- cacheRead: 0.0188,
13704
+ cacheRead: 0.010400000000000001,
13218
13705
  cacheWrite: 0,
13219
13706
  },
13220
13707
  contextWindow: 1048576,
@@ -13250,9 +13737,9 @@ export const MODELS = {
13250
13737
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
13251
13738
  input: ["text"],
13252
13739
  cost: {
13253
- input: 0.20879999999999999,
13254
- output: 0.41759999999999997,
13255
- cacheRead: 0.0174,
13740
+ input: 0.9552599999999999,
13741
+ output: 1.9105199999999998,
13742
+ cacheRead: 0.07960500000000001,
13256
13743
  cacheWrite: 0,
13257
13744
  },
13258
13745
  contextWindow: 1048576,
@@ -13269,13 +13756,13 @@ export const MODELS = {
13269
13756
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
13270
13757
  input: ["text"],
13271
13758
  cost: {
13272
- input: 0.22,
13273
- output: 4.199999999999999,
13274
- cacheRead: 0.14,
13759
+ input: 1.32,
13760
+ output: 3.9600000000000004,
13761
+ cacheRead: 0.044,
13275
13762
  cacheWrite: 0,
13276
13763
  },
13277
13764
  contextWindow: 1048576,
13278
- maxTokens: 943718,
13765
+ maxTokens: 393216,
13279
13766
  },
13280
13767
  "deepseek/deepseek-v4.1-flash": {
13281
13768
  id: "deepseek/deepseek-v4.1-flash",
@@ -13791,6 +14278,23 @@ export const MODELS = {
13791
14278
  contextWindow: 1048576,
13792
14279
  maxTokens: 65536,
13793
14280
  },
14281
+ "google/gemini-nano-banana-2.1": {
14282
+ id: "google/gemini-nano-banana-2.1",
14283
+ name: "Google: Nano Banana 2.1",
14284
+ api: "openai-completions",
14285
+ provider: "openrouter",
14286
+ baseUrl: "https://openrouter.ai/api/v1",
14287
+ reasoning: true,
14288
+ input: ["text", "image"],
14289
+ cost: {
14290
+ input: 1.5,
14291
+ output: 7.5,
14292
+ cacheRead: 0,
14293
+ cacheWrite: 0,
14294
+ },
14295
+ contextWindow: 65536,
14296
+ maxTokens: 58982,
14297
+ },
13794
14298
  "google/gemma-3-12b-it": {
13795
14299
  id: "google/gemma-3-12b-it",
13796
14300
  name: "Google: Gemma 3 12B",
@@ -13834,9 +14338,9 @@ export const MODELS = {
13834
14338
  reasoning: true,
13835
14339
  input: ["text", "image"],
13836
14340
  cost: {
13837
- input: 0.0675,
13838
- output: 0.22499999999999998,
13839
- cacheRead: 0.0375,
14341
+ input: 0.09,
14342
+ output: 0.3,
14343
+ cacheRead: 0.049999999999999996,
13840
14344
  cacheWrite: 0,
13841
14345
  },
13842
14346
  contextWindow: 262144,
@@ -14029,23 +14533,6 @@ export const MODELS = {
14029
14533
  contextWindow: 262144,
14030
14534
  maxTokens: 32768,
14031
14535
  },
14032
- "kwaipilot/kat-coder-pro-v2.5": {
14033
- id: "kwaipilot/kat-coder-pro-v2.5",
14034
- name: "Kwaipilot: KAT-Coder-Pro V2.5",
14035
- api: "openai-completions",
14036
- provider: "openrouter",
14037
- baseUrl: "https://openrouter.ai/api/v1",
14038
- reasoning: false,
14039
- input: ["text"],
14040
- cost: {
14041
- input: 0.74,
14042
- output: 2.96,
14043
- cacheRead: 0.15,
14044
- cacheWrite: 0,
14045
- },
14046
- contextWindow: 262144,
14047
- maxTokens: 235929,
14048
- },
14049
14536
  "liquid/lfm-2.5-2.6b:free": {
14050
14537
  id: "liquid/lfm-2.5-2.6b:free",
14051
14538
  name: "LiquidAI: LFM2.5-2.6B (free)",
@@ -14142,7 +14629,7 @@ export const MODELS = {
14142
14629
  cost: {
14143
14630
  input: 0.1875,
14144
14631
  output: 0.6525,
14145
- cacheRead: 0,
14632
+ cacheRead: 0.049999999999999996,
14146
14633
  cacheWrite: 0,
14147
14634
  },
14148
14635
  contextWindow: 1048576,
@@ -14174,13 +14661,13 @@ export const MODELS = {
14174
14661
  reasoning: true,
14175
14662
  input: ["text", "image"],
14176
14663
  cost: {
14177
- input: 0.35,
14178
- output: 1.5,
14664
+ input: 0.3,
14665
+ output: 1.2,
14179
14666
  cacheRead: 0.04,
14180
14667
  cacheWrite: 0,
14181
14668
  },
14182
14669
  contextWindow: 131072,
14183
- maxTokens: 117964,
14670
+ maxTokens: 16384,
14184
14671
  },
14185
14672
  "meta/muse-spark-1.1": {
14186
14673
  id: "meta/muse-spark-1.1",
@@ -14556,6 +15043,23 @@ export const MODELS = {
14556
15043
  contextWindow: 262144,
14557
15044
  maxTokens: 209715,
14558
15045
  },
15046
+ "mistralai/mistral-large-4-0": {
15047
+ id: "mistralai/mistral-large-4-0",
15048
+ name: "Mistral: Mistral Large 4",
15049
+ api: "openai-completions",
15050
+ provider: "openrouter",
15051
+ baseUrl: "https://openrouter.ai/api/v1",
15052
+ reasoning: true,
15053
+ input: ["text", "image"],
15054
+ cost: {
15055
+ input: 0.6799999999999999,
15056
+ output: 2.09,
15057
+ cacheRead: 0.07,
15058
+ cacheWrite: 0,
15059
+ },
15060
+ contextWindow: 524288,
15061
+ maxTokens: 262144,
15062
+ },
14559
15063
  "mistralai/mistral-medium-3": {
14560
15064
  id: "mistralai/mistral-medium-3",
14561
15065
  name: "Mistral: Mistral Medium 3",
@@ -14854,9 +15358,9 @@ export const MODELS = {
14854
15358
  reasoning: true,
14855
15359
  input: ["text", "image"],
14856
15360
  cost: {
14857
- input: 0.95,
14858
- output: 4,
14859
- cacheRead: 0.16,
15361
+ input: 0.46499999999999997,
15362
+ output: 2.4499999999999997,
15363
+ cacheRead: 0.0975,
14860
15364
  cacheWrite: 0,
14861
15365
  },
14862
15366
  contextWindow: 262144,
@@ -14888,9 +15392,9 @@ export const MODELS = {
14888
15392
  reasoning: true,
14889
15393
  input: ["text", "image"],
14890
15394
  cost: {
14891
- input: 0.9900000000000001,
14892
- output: 13,
14893
- cacheRead: 0.7999999999999999,
15395
+ input: 0.58,
15396
+ output: 12.3,
15397
+ cacheRead: 0.39999999999999997,
14894
15398
  cacheWrite: 0,
14895
15399
  },
14896
15400
  contextWindow: 1048576,
@@ -14939,9 +15443,9 @@ export const MODELS = {
14939
15443
  reasoning: true,
14940
15444
  input: ["text"],
14941
15445
  cost: {
14942
- input: 0.049999999999999996,
14943
- output: 0.19999999999999998,
14944
- cacheRead: 0.03,
15446
+ input: 0.06,
15447
+ output: 0.24,
15448
+ cacheRead: 0,
14945
15449
  cacheWrite: 0,
14946
15450
  },
14947
15451
  contextWindow: 262144,
@@ -14973,13 +15477,13 @@ export const MODELS = {
14973
15477
  reasoning: true,
14974
15478
  input: ["text"],
14975
15479
  cost: {
14976
- input: 0.08,
14977
- output: 0.44999999999999996,
15480
+ input: 0.08499999999999999,
15481
+ output: 0.39999999999999997,
14978
15482
  cacheRead: 0,
14979
15483
  cacheWrite: 0,
14980
15484
  },
14981
15485
  contextWindow: 262144,
14982
- maxTokens: 235929,
15486
+ maxTokens: 16384,
14983
15487
  },
14984
15488
  "nvidia/nemotron-3-super-120b-a12b:free": {
14985
15489
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -15041,9 +15545,9 @@ export const MODELS = {
15041
15545
  reasoning: true,
15042
15546
  input: ["text"],
15043
15547
  cost: {
15044
- input: 0.0595,
15045
- output: 0.16999999999999998,
15046
- cacheRead: 0.02975,
15548
+ input: 0.049,
15549
+ output: 0.14,
15550
+ cacheRead: 0.0245,
15047
15551
  cacheWrite: 0,
15048
15552
  },
15049
15553
  contextWindow: 262144,
@@ -15978,10 +16482,10 @@ export const MODELS = {
15978
16482
  reasoning: true,
15979
16483
  input: ["text", "image"],
15980
16484
  cost: {
15981
- input: 4,
15982
- output: 20,
15983
- cacheRead: 0.39999999999999997,
15984
- cacheWrite: 5,
16485
+ input: 2,
16486
+ output: 10,
16487
+ cacheRead: 0.19999999999999998,
16488
+ cacheWrite: 2.5,
15985
16489
  },
15986
16490
  contextWindow: 1050000,
15987
16491
  maxTokens: 128000,
@@ -16913,13 +17417,13 @@ export const MODELS = {
16913
17417
  reasoning: false,
16914
17418
  input: ["text"],
16915
17419
  cost: {
16916
- input: 0.0875,
16917
- output: 0.35,
16918
- cacheRead: 0.0175,
17420
+ input: 0.09,
17421
+ output: 0.55,
17422
+ cacheRead: 0,
16919
17423
  cacheWrite: 0,
16920
17424
  },
16921
17425
  contextWindow: 262144,
16922
- maxTokens: 235929,
17426
+ maxTokens: 16384,
16923
17427
  },
16924
17428
  "qwen/qwen3-235b-a22b-thinking-2507": {
16925
17429
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -17151,13 +17655,13 @@ export const MODELS = {
17151
17655
  reasoning: false,
17152
17656
  input: ["text"],
17153
17657
  cost: {
17154
- input: 0.09999999999999999,
17155
- output: 1.1,
17156
- cacheRead: 0.07,
17658
+ input: 0.15,
17659
+ output: 1.5,
17660
+ cacheRead: 0,
17157
17661
  cacheWrite: 0,
17158
17662
  },
17159
17663
  contextWindow: 262144,
17160
- maxTokens: 235929,
17664
+ maxTokens: 32768,
17161
17665
  },
17162
17666
  "qwen/qwen3-next-80b-a3b-thinking": {
17163
17667
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -17321,13 +17825,13 @@ export const MODELS = {
17321
17825
  reasoning: true,
17322
17826
  input: ["text", "image"],
17323
17827
  cost: {
17324
- input: 0.195,
17325
- output: 1.56,
17828
+ input: 0.26,
17829
+ output: 2.6,
17326
17830
  cacheRead: 0,
17327
17831
  cacheWrite: 0,
17328
17832
  },
17329
17833
  contextWindow: 262144,
17330
- maxTokens: 65536,
17834
+ maxTokens: 81920,
17331
17835
  },
17332
17836
  "qwen/qwen3.5-35b-a3b": {
17333
17837
  id: "qwen/qwen3.5-35b-a3b",
@@ -17355,13 +17859,13 @@ export const MODELS = {
17355
17859
  reasoning: true,
17356
17860
  input: ["text", "image"],
17357
17861
  cost: {
17358
- input: 0.55,
17359
- output: 3.5,
17360
- cacheRead: 0.22499999999999998,
17862
+ input: 0.44999999999999996,
17863
+ output: 3,
17864
+ cacheRead: 0.22,
17361
17865
  cacheWrite: 0,
17362
17866
  },
17363
17867
  contextWindow: 262144,
17364
- maxTokens: 235929,
17868
+ maxTokens: 81920,
17365
17869
  },
17366
17870
  "qwen/qwen3.5-9b": {
17367
17871
  id: "qwen/qwen3.5-9b",
@@ -17440,13 +17944,13 @@ export const MODELS = {
17440
17944
  reasoning: true,
17441
17945
  input: ["text", "image"],
17442
17946
  cost: {
17443
- input: 0.32,
17444
- output: 3.1999999999999997,
17445
- cacheRead: 0,
17947
+ input: 0.3,
17948
+ output: 2,
17949
+ cacheRead: 0.03,
17446
17950
  cacheWrite: 0,
17447
17951
  },
17448
17952
  contextWindow: 262144,
17449
- maxTokens: 81920,
17953
+ maxTokens: 65536,
17450
17954
  },
17451
17955
  "qwen/qwen3.6-35b-a3b": {
17452
17956
  id: "qwen/qwen3.6-35b-a3b",
@@ -17593,31 +18097,14 @@ export const MODELS = {
17593
18097
  reasoning: true,
17594
18098
  input: ["text", "image"],
17595
18099
  cost: {
17596
- input: 0.42,
17597
- output: 3,
18100
+ input: 0.425,
18101
+ output: 2.5500000000000003,
17598
18102
  cacheRead: 0.08499999999999999,
17599
- cacheWrite: 0,
18103
+ cacheWrite: 0.53125,
17600
18104
  },
17601
18105
  contextWindow: 1000000,
17602
18106
  maxTokens: 131072,
17603
18107
  },
17604
- "qwen/qwen3.8-27b:free": {
17605
- id: "qwen/qwen3.8-27b:free",
17606
- name: "Qwen: Qwen3.8 27B (free)",
17607
- api: "openai-completions",
17608
- provider: "openrouter",
17609
- baseUrl: "https://openrouter.ai/api/v1",
17610
- reasoning: true,
17611
- input: ["text", "image"],
17612
- cost: {
17613
- input: 0,
17614
- output: 0,
17615
- cacheRead: 0,
17616
- cacheWrite: 0,
17617
- },
17618
- contextWindow: 262144,
17619
- maxTokens: 235929,
17620
- },
17621
18108
  "qwen/qwen3.8-flash": {
17622
18109
  id: "qwen/qwen3.8-flash",
17623
18110
  name: "Qwen: Qwen3.8 Flash",
@@ -17805,23 +18292,6 @@ export const MODELS = {
17805
18292
  contextWindow: 131072,
17806
18293
  maxTokens: 16384,
17807
18294
  },
17808
- "stealth/space-bunny-alpha": {
17809
- id: "stealth/space-bunny-alpha",
17810
- name: "Space Bunny Alpha",
17811
- api: "openai-completions",
17812
- provider: "openrouter",
17813
- baseUrl: "https://openrouter.ai/api/v1",
17814
- reasoning: true,
17815
- input: ["text", "image"],
17816
- cost: {
17817
- input: 0,
17818
- output: 0,
17819
- cacheRead: 0,
17820
- cacheWrite: 0,
17821
- },
17822
- contextWindow: 1000000,
17823
- maxTokens: 524288,
17824
- },
17825
18295
  "stepfun/step-3.5-flash": {
17826
18296
  id: "stepfun/step-3.5-flash",
17827
18297
  name: "StepFun: Step 3.5 Flash",
@@ -17899,13 +18369,13 @@ export const MODELS = {
17899
18369
  reasoning: true,
17900
18370
  input: ["text", "image"],
17901
18371
  cost: {
17902
- input: 0.95,
18372
+ input: 1,
17903
18373
  output: 4.05,
17904
- cacheRead: 0.16,
18374
+ cacheRead: 0.16999999999999998,
17905
18375
  cacheWrite: 0,
17906
18376
  },
17907
18377
  contextWindow: 524288,
17908
- maxTokens: 262144,
18378
+ maxTokens: 471859,
17909
18379
  },
17910
18380
  "thinkingmachines/inkling-small": {
17911
18381
  id: "thinkingmachines/inkling-small",
@@ -18343,7 +18813,7 @@ export const MODELS = {
18343
18813
  cost: {
18344
18814
  input: 0.3,
18345
18815
  output: 0.8999999999999999,
18346
- cacheRead: 0.049999999999999996,
18816
+ cacheRead: 0.055,
18347
18817
  cacheWrite: 0,
18348
18818
  },
18349
18819
  contextWindow: 131072,
@@ -18426,13 +18896,13 @@ export const MODELS = {
18426
18896
  reasoning: true,
18427
18897
  input: ["text"],
18428
18898
  cost: {
18429
- input: 1.4,
18430
- output: 4.4,
18431
- cacheRead: 0.26,
18899
+ input: 0.966,
18900
+ output: 3.036,
18901
+ cacheRead: 0.1794,
18432
18902
  cacheWrite: 0,
18433
18903
  },
18434
18904
  contextWindow: 204800,
18435
- maxTokens: 131072,
18905
+ maxTokens: 128000,
18436
18906
  },
18437
18907
  "z-ai/glm-5.2": {
18438
18908
  id: "z-ai/glm-5.2",
@@ -18443,13 +18913,13 @@ export const MODELS = {
18443
18913
  reasoning: true,
18444
18914
  input: ["text"],
18445
18915
  cost: {
18446
- input: 0.41,
18447
- output: 3.9899999999999998,
18448
- cacheRead: 0.26,
18916
+ input: 0.03,
18917
+ output: 10,
18918
+ cacheRead: 0.03,
18449
18919
  cacheWrite: 0,
18450
18920
  },
18451
18921
  contextWindow: 1048576,
18452
- maxTokens: 943718,
18922
+ maxTokens: 131072,
18453
18923
  },
18454
18924
  "z-ai/glm-5.3": {
18455
18925
  id: "z-ai/glm-5.3",
@@ -18460,13 +18930,13 @@ export const MODELS = {
18460
18930
  reasoning: true,
18461
18931
  input: ["text"],
18462
18932
  cost: {
18463
- input: 1.4,
18464
- output: 4.4,
18465
- cacheRead: 0.14,
18933
+ input: 0.049,
18934
+ output: 3.39,
18935
+ cacheRead: 0.048,
18466
18936
  cacheWrite: 0,
18467
18937
  },
18468
18938
  contextWindow: 1048576,
18469
- maxTokens: 131072,
18939
+ maxTokens: 943718,
18470
18940
  },
18471
18941
  "z-ai/glm-5.3-flash": {
18472
18942
  id: "z-ai/glm-5.3-flash",
@@ -18597,13 +19067,13 @@ export const MODELS = {
18597
19067
  reasoning: true,
18598
19068
  input: ["text", "image"],
18599
19069
  cost: {
18600
- input: 1,
18601
- output: 5,
18602
- cacheRead: 0.09999999999999999,
18603
- cacheWrite: 1.25,
19070
+ input: 0.09999999999999999,
19071
+ output: 0.5,
19072
+ cacheRead: 0.01,
19073
+ cacheWrite: 0.125,
18604
19074
  },
18605
- contextWindow: 200000,
18606
- maxTokens: 64000,
19075
+ contextWindow: 1000000,
19076
+ maxTokens: 128000,
18607
19077
  },
18608
19078
  "~anthropic/claude-opus-latest": {
18609
19079
  id: "~anthropic/claude-opus-latest",
@@ -18633,7 +19103,7 @@ export const MODELS = {
18633
19103
  cost: {
18634
19104
  input: 2,
18635
19105
  output: 10,
18636
- cacheRead: 0.19999999999999998,
19106
+ cacheRead: 0.09999999999999999,
18637
19107
  cacheWrite: 2.5,
18638
19108
  },
18639
19109
  contextWindow: 1000000,
@@ -18648,9 +19118,9 @@ export const MODELS = {
18648
19118
  reasoning: true,
18649
19119
  input: ["text", "image"],
18650
19120
  cost: {
18651
- input: 0.02,
18652
- output: 0.6,
18653
- cacheRead: 0.01,
19121
+ input: 0.003,
19122
+ output: 2.4,
19123
+ cacheRead: 0.003,
18654
19124
  cacheWrite: 0,
18655
19125
  },
18656
19126
  contextWindow: 1048576,
@@ -18665,13 +19135,13 @@ export const MODELS = {
18665
19135
  reasoning: true,
18666
19136
  input: ["text"],
18667
19137
  cost: {
18668
- input: 0.22,
18669
- output: 4.199999999999999,
18670
- cacheRead: 0.14,
19138
+ input: 0.135,
19139
+ output: 8,
19140
+ cacheRead: 0.09999999999999999,
18671
19141
  cacheWrite: 0,
18672
19142
  },
18673
19143
  contextWindow: 1048576,
18674
- maxTokens: 943718,
19144
+ maxTokens: 393216,
18675
19145
  },
18676
19146
  "~deepseek/deepseek-v4-flash-latest": {
18677
19147
  id: "~deepseek/deepseek-v4-flash-latest",
@@ -18684,9 +19154,9 @@ export const MODELS = {
18684
19154
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
18685
19155
  input: ["text"],
18686
19156
  cost: {
18687
- input: 0.0188,
19157
+ input: 0.010400000000000001,
18688
19158
  output: 1.28,
18689
- cacheRead: 0.0188,
19159
+ cacheRead: 0.010400000000000001,
18690
19160
  cacheWrite: 0,
18691
19161
  },
18692
19162
  contextWindow: 1048576,
@@ -18735,7 +19205,7 @@ export const MODELS = {
18735
19205
  reasoning: true,
18736
19206
  input: ["text", "image"],
18737
19207
  cost: {
18738
- input: 0.9895,
19208
+ input: 0.56,
18739
19209
  output: 13,
18740
19210
  cacheRead: 0.44999999999999996,
18741
19211
  cacheWrite: 0,
@@ -18854,13 +19324,13 @@ export const MODELS = {
18854
19324
  reasoning: true,
18855
19325
  input: ["text", "image"],
18856
19326
  cost: {
18857
- input: 0.035,
18858
- output: 0.5,
18859
- cacheRead: 0.035,
19327
+ input: 0.032,
19328
+ output: 1.414768,
19329
+ cacheRead: 0.02,
18860
19330
  cacheWrite: 0,
18861
19331
  },
18862
19332
  contextWindow: 1048576,
18863
- maxTokens: 131072,
19333
+ maxTokens: 943718,
18864
19334
  },
18865
19335
  "~z-ai/glm-latest": {
18866
19336
  id: "~z-ai/glm-latest",
@@ -18871,9 +19341,9 @@ export const MODELS = {
18871
19341
  reasoning: true,
18872
19342
  input: ["text"],
18873
19343
  cost: {
18874
- input: 0.12,
18875
- output: 4,
18876
- cacheRead: 0.08,
19344
+ input: 0.036,
19345
+ output: 12,
19346
+ cacheRead: 0.0335,
18877
19347
  cacheWrite: 0,
18878
19348
  },
18879
19349
  contextWindow: 1048576,
@@ -19885,6 +20355,23 @@ export const MODELS = {
19885
20355
  contextWindow: 200000,
19886
20356
  maxTokens: 64000,
19887
20357
  },
20358
+ "anthropic/claude-haiku-5.5": {
20359
+ id: "anthropic/claude-haiku-5.5",
20360
+ name: "Claude Haiku 5.5",
20361
+ api: "anthropic-messages",
20362
+ provider: "vercel-ai-gateway",
20363
+ baseUrl: "https://ai-gateway.vercel.sh",
20364
+ reasoning: true,
20365
+ input: ["text", "image"],
20366
+ cost: {
20367
+ input: 0.09999999999999999,
20368
+ output: 0.5,
20369
+ cacheRead: 0.01,
20370
+ cacheWrite: 0.125,
20371
+ },
20372
+ contextWindow: 1000000,
20373
+ maxTokens: 128000,
20374
+ },
19888
20375
  "anthropic/claude-opus-4": {
19889
20376
  id: "anthropic/claude-opus-4",
19890
20377
  name: "Claude Opus 4",
@@ -20138,7 +20625,7 @@ export const MODELS = {
20138
20625
  cost: {
20139
20626
  input: 2,
20140
20627
  output: 10,
20141
- cacheRead: 0.19999999999999998,
20628
+ cacheRead: 0.09999999999999999,
20142
20629
  cacheWrite: 2.5,
20143
20630
  },
20144
20631
  contextWindow: 1000000,
@@ -20154,7 +20641,7 @@ export const MODELS = {
20154
20641
  input: ["text"],
20155
20642
  cost: {
20156
20643
  input: 0.25,
20157
- output: 0.8999999999999999,
20644
+ output: 0.7999999999999999,
20158
20645
  cacheRead: 0,
20159
20646
  cacheWrite: 0,
20160
20647
  },
@@ -20263,23 +20750,6 @@ export const MODELS = {
20263
20750
  contextWindow: 163840,
20264
20751
  maxTokens: 128000,
20265
20752
  },
20266
- "deepseek/deepseek-v3.1-terminus": {
20267
- id: "deepseek/deepseek-v3.1-terminus",
20268
- name: "DeepSeek V3.1 Terminus",
20269
- api: "anthropic-messages",
20270
- provider: "vercel-ai-gateway",
20271
- baseUrl: "https://ai-gateway.vercel.sh",
20272
- reasoning: true,
20273
- input: ["text"],
20274
- cost: {
20275
- input: 0.27,
20276
- output: 1,
20277
- cacheRead: 0.135,
20278
- cacheWrite: 0,
20279
- },
20280
- contextWindow: 131072,
20281
- maxTokens: 65536,
20282
- },
20283
20753
  "deepseek/deepseek-v3.2": {
20284
20754
  id: "deepseek/deepseek-v3.2",
20285
20755
  name: "DeepSeek V3.2",
@@ -20416,23 +20886,6 @@ export const MODELS = {
20416
20886
  contextWindow: 1048576,
20417
20887
  maxTokens: 32768,
20418
20888
  },
20419
- "deepseek/deepseek-v4.1-flash-fast": {
20420
- id: "deepseek/deepseek-v4.1-flash-fast",
20421
- name: "DeepSeek V4.1 Flash Fast",
20422
- api: "anthropic-messages",
20423
- provider: "vercel-ai-gateway",
20424
- baseUrl: "https://ai-gateway.vercel.sh",
20425
- reasoning: true,
20426
- input: ["text", "image"],
20427
- cost: {
20428
- input: 0.3,
20429
- output: 1.2,
20430
- cacheRead: 0.006,
20431
- cacheWrite: 0,
20432
- },
20433
- contextWindow: 1048576,
20434
- maxTokens: 1000000,
20435
- },
20436
20889
  "fireworks/ember-1": {
20437
20890
  id: "fireworks/ember-1",
20438
20891
  name: "Ember-1",
@@ -20765,26 +21218,9 @@ export const MODELS = {
20765
21218
  reasoning: true,
20766
21219
  input: ["text"],
20767
21220
  cost: {
20768
- input: 0,
20769
- output: 0,
20770
- cacheRead: 0,
20771
- cacheWrite: 0,
20772
- },
20773
- contextWindow: 256000,
20774
- maxTokens: 32000,
20775
- },
20776
- "inclusionai/ling-3.0-flash-sante-free": {
20777
- id: "inclusionai/ling-3.0-flash-sante-free",
20778
- name: "Ling 3.0 Flash Sante (Free)",
20779
- api: "anthropic-messages",
20780
- provider: "vercel-ai-gateway",
20781
- baseUrl: "https://ai-gateway.vercel.sh",
20782
- reasoning: true,
20783
- input: ["text"],
20784
- cost: {
20785
- input: 0,
20786
- output: 0,
20787
- cacheRead: 0,
21221
+ input: 0.075,
21222
+ output: 0.22,
21223
+ cacheRead: 0.015,
20788
21224
  cacheWrite: 0,
20789
21225
  },
20790
21226
  contextWindow: 256000,
@@ -21283,6 +21719,23 @@ export const MODELS = {
21283
21719
  contextWindow: 262144,
21284
21720
  maxTokens: 256000,
21285
21721
  },
21722
+ "mistral/mistral-large-4": {
21723
+ id: "mistral/mistral-large-4",
21724
+ name: "Mistral Large 4",
21725
+ api: "anthropic-messages",
21726
+ provider: "vercel-ai-gateway",
21727
+ baseUrl: "https://ai-gateway.vercel.sh",
21728
+ reasoning: true,
21729
+ input: ["text", "image"],
21730
+ cost: {
21731
+ input: 0.6799999999999999,
21732
+ output: 2.09,
21733
+ cacheRead: 0.07,
21734
+ cacheWrite: 0,
21735
+ },
21736
+ contextWindow: 524288,
21737
+ maxTokens: 262144,
21738
+ },
21286
21739
  "mistral/mistral-medium-3.5": {
21287
21740
  id: "mistral/mistral-medium-3.5",
21288
21741
  name: "Mistral Medium Latest",
@@ -21368,23 +21821,6 @@ export const MODELS = {
21368
21821
  contextWindow: 131072,
21369
21822
  maxTokens: 131072,
21370
21823
  },
21371
- "moonshotai/kimi-k2-thinking": {
21372
- id: "moonshotai/kimi-k2-thinking",
21373
- name: "Kimi K2 Thinking",
21374
- api: "anthropic-messages",
21375
- provider: "vercel-ai-gateway",
21376
- baseUrl: "https://ai-gateway.vercel.sh",
21377
- reasoning: true,
21378
- input: ["text"],
21379
- cost: {
21380
- input: 0.47,
21381
- output: 2,
21382
- cacheRead: 0.14100000000000001,
21383
- cacheWrite: 0,
21384
- },
21385
- contextWindow: 216144,
21386
- maxTokens: 216144,
21387
- },
21388
21824
  "moonshotai/kimi-k2.5": {
21389
21825
  id: "moonshotai/kimi-k2.5",
21390
21826
  name: "Kimi K2.5",
@@ -22684,9 +23120,9 @@ export const MODELS = {
22684
23120
  reasoning: true,
22685
23121
  input: ["text"],
22686
23122
  cost: {
22687
- input: 0.09999999999999999,
22688
- output: 0.19999999999999998,
22689
- cacheRead: 0.01,
23123
+ input: 0.09,
23124
+ output: 0.18,
23125
+ cacheRead: 0.009,
22690
23126
  cacheWrite: 0,
22691
23127
  },
22692
23128
  contextWindow: 1000000,
@@ -23032,22 +23468,22 @@ export const MODELS = {
23032
23468
  contextWindow: 256000,
23033
23469
  maxTokens: 256000,
23034
23470
  },
23035
- "stepfun/step-3.5-flash": {
23036
- id: "stepfun/step-3.5-flash",
23037
- name: "StepFun 3.5 Flash",
23471
+ "stealth/glyph-cluster": {
23472
+ id: "stealth/glyph-cluster",
23473
+ name: "Glyph Cluster",
23038
23474
  api: "anthropic-messages",
23039
23475
  provider: "vercel-ai-gateway",
23040
23476
  baseUrl: "https://ai-gateway.vercel.sh",
23041
23477
  reasoning: true,
23042
- input: ["text", "image"],
23478
+ input: ["text"],
23043
23479
  cost: {
23044
- input: 0.09,
23045
- output: 0.3,
23046
- cacheRead: 0.02,
23480
+ input: 0,
23481
+ output: 0,
23482
+ cacheRead: 0,
23047
23483
  cacheWrite: 0,
23048
23484
  },
23049
- contextWindow: 262114,
23050
- maxTokens: 262114,
23485
+ contextWindow: 256000,
23486
+ maxTokens: 256000,
23051
23487
  },
23052
23488
  "stepfun/step-3.7-flash": {
23053
23489
  id: "stepfun/step-3.7-flash",