@apholdings/jensen-ai 1.0.7 → 1.0.8

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.
@@ -8,7 +8,7 @@ export const MODELS = {
8
8
  api: "bedrock-converse-stream",
9
9
  provider: "amazon-bedrock",
10
10
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
11
- reasoning: false,
11
+ reasoning: true,
12
12
  input: ["text", "image"],
13
13
  cost: {
14
14
  input: 0.33,
@@ -342,6 +342,23 @@ export const MODELS = {
342
342
  contextWindow: 163840,
343
343
  maxTokens: 81920,
344
344
  },
345
+ "eu.anthropic.claude-fable-5": {
346
+ id: "eu.anthropic.claude-fable-5",
347
+ name: "Claude Fable 5 (EU)",
348
+ api: "bedrock-converse-stream",
349
+ provider: "amazon-bedrock",
350
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
351
+ reasoning: true,
352
+ input: ["text", "image"],
353
+ cost: {
354
+ input: 11,
355
+ output: 55,
356
+ cacheRead: 1.1,
357
+ cacheWrite: 13.75,
358
+ },
359
+ contextWindow: 1000000,
360
+ maxTokens: 128000,
361
+ },
345
362
  "eu.anthropic.claude-haiku-4-5-20251001-v1:0": {
346
363
  id: "eu.anthropic.claude-haiku-4-5-20251001-v1:0",
347
364
  name: "Claude Haiku 4.5 (EU)",
@@ -461,6 +478,23 @@ export const MODELS = {
461
478
  contextWindow: 1000000,
462
479
  maxTokens: 64000,
463
480
  },
481
+ "global.anthropic.claude-fable-5": {
482
+ id: "global.anthropic.claude-fable-5",
483
+ name: "Claude Fable 5 (Global)",
484
+ api: "bedrock-converse-stream",
485
+ provider: "amazon-bedrock",
486
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
487
+ reasoning: true,
488
+ input: ["text", "image"],
489
+ cost: {
490
+ input: 10,
491
+ output: 50,
492
+ cacheRead: 1,
493
+ cacheWrite: 12.5,
494
+ },
495
+ contextWindow: 1000000,
496
+ maxTokens: 128000,
497
+ },
464
498
  "global.anthropic.claude-haiku-4-5-20251001-v1:0": {
465
499
  id: "global.anthropic.claude-haiku-4-5-20251001-v1:0",
466
500
  name: "Claude Haiku 4.5 (Global)",
@@ -1073,13 +1107,64 @@ export const MODELS = {
1073
1107
  contextWindow: 262144,
1074
1108
  maxTokens: 131072,
1075
1109
  },
1110
+ "openai.gpt-5.4": {
1111
+ id: "openai.gpt-5.4",
1112
+ name: "GPT-5.4",
1113
+ api: "bedrock-converse-stream",
1114
+ provider: "amazon-bedrock",
1115
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1116
+ reasoning: true,
1117
+ input: ["text", "image"],
1118
+ cost: {
1119
+ input: 2.75,
1120
+ output: 16.5,
1121
+ cacheRead: 0.275,
1122
+ cacheWrite: 0,
1123
+ },
1124
+ contextWindow: 272000,
1125
+ maxTokens: 128000,
1126
+ },
1127
+ "openai.gpt-5.5": {
1128
+ id: "openai.gpt-5.5",
1129
+ name: "GPT-5.5",
1130
+ api: "bedrock-converse-stream",
1131
+ provider: "amazon-bedrock",
1132
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1133
+ reasoning: true,
1134
+ input: ["text", "image"],
1135
+ cost: {
1136
+ input: 5.5,
1137
+ output: 33,
1138
+ cacheRead: 0.55,
1139
+ cacheWrite: 0,
1140
+ },
1141
+ contextWindow: 272000,
1142
+ maxTokens: 128000,
1143
+ },
1144
+ "openai.gpt-oss-120b": {
1145
+ id: "openai.gpt-oss-120b",
1146
+ name: "gpt-oss-120b",
1147
+ api: "bedrock-converse-stream",
1148
+ provider: "amazon-bedrock",
1149
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1150
+ reasoning: true,
1151
+ input: ["text"],
1152
+ cost: {
1153
+ input: 0.15,
1154
+ output: 0.6,
1155
+ cacheRead: 0,
1156
+ cacheWrite: 0,
1157
+ },
1158
+ contextWindow: 128000,
1159
+ maxTokens: 16384,
1160
+ },
1076
1161
  "openai.gpt-oss-120b-1:0": {
1077
1162
  id: "openai.gpt-oss-120b-1:0",
1078
1163
  name: "gpt-oss-120b",
1079
1164
  api: "bedrock-converse-stream",
1080
1165
  provider: "amazon-bedrock",
1081
1166
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1082
- reasoning: false,
1167
+ reasoning: true,
1083
1168
  input: ["text"],
1084
1169
  cost: {
1085
1170
  input: 0.15,
@@ -1090,13 +1175,30 @@ export const MODELS = {
1090
1175
  contextWindow: 128000,
1091
1176
  maxTokens: 16384,
1092
1177
  },
1178
+ "openai.gpt-oss-20b": {
1179
+ id: "openai.gpt-oss-20b",
1180
+ name: "gpt-oss-20b",
1181
+ api: "bedrock-converse-stream",
1182
+ provider: "amazon-bedrock",
1183
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1184
+ reasoning: true,
1185
+ input: ["text"],
1186
+ cost: {
1187
+ input: 0.07,
1188
+ output: 0.3,
1189
+ cacheRead: 0,
1190
+ cacheWrite: 0,
1191
+ },
1192
+ contextWindow: 128000,
1193
+ maxTokens: 16384,
1194
+ },
1093
1195
  "openai.gpt-oss-20b-1:0": {
1094
1196
  id: "openai.gpt-oss-20b-1:0",
1095
1197
  name: "gpt-oss-20b",
1096
1198
  api: "bedrock-converse-stream",
1097
1199
  provider: "amazon-bedrock",
1098
1200
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1099
- reasoning: false,
1201
+ reasoning: true,
1100
1202
  input: ["text"],
1101
1203
  cost: {
1102
1204
  input: 0.07,
@@ -1260,6 +1362,23 @@ export const MODELS = {
1260
1362
  contextWindow: 262000,
1261
1363
  maxTokens: 262000,
1262
1364
  },
1365
+ "us.anthropic.claude-fable-5": {
1366
+ id: "us.anthropic.claude-fable-5",
1367
+ name: "Claude Fable 5 (US)",
1368
+ api: "bedrock-converse-stream",
1369
+ provider: "amazon-bedrock",
1370
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1371
+ reasoning: true,
1372
+ input: ["text", "image"],
1373
+ cost: {
1374
+ input: 10,
1375
+ output: 50,
1376
+ cacheRead: 1,
1377
+ cacheWrite: 12.5,
1378
+ },
1379
+ contextWindow: 1000000,
1380
+ maxTokens: 128000,
1381
+ },
1263
1382
  "us.anthropic.claude-haiku-4-5-20251001-v1:0": {
1264
1383
  id: "us.anthropic.claude-haiku-4-5-20251001-v1:0",
1265
1384
  name: "Claude Haiku 4.5 (US)",
@@ -1670,6 +1789,23 @@ export const MODELS = {
1670
1789
  contextWindow: 200000,
1671
1790
  maxTokens: 4096,
1672
1791
  },
1792
+ "claude-fable-5": {
1793
+ id: "claude-fable-5",
1794
+ name: "Claude Fable 5",
1795
+ api: "anthropic-messages",
1796
+ provider: "anthropic",
1797
+ baseUrl: "https://api.anthropic.com",
1798
+ reasoning: true,
1799
+ input: ["text", "image"],
1800
+ cost: {
1801
+ input: 10,
1802
+ output: 50,
1803
+ cacheRead: 1,
1804
+ cacheWrite: 12.5,
1805
+ },
1806
+ contextWindow: 1000000,
1807
+ maxTokens: 128000,
1808
+ },
1673
1809
  "claude-haiku-4-5": {
1674
1810
  id: "claude-haiku-4-5",
1675
1811
  name: "Claude Haiku 4.5 (latest)",
@@ -2669,30 +2805,13 @@ export const MODELS = {
2669
2805
  reasoning: true,
2670
2806
  input: ["text"],
2671
2807
  cost: {
2672
- input: 0.25,
2673
- output: 0.69,
2808
+ input: 0.35,
2809
+ output: 0.75,
2674
2810
  cacheRead: 0,
2675
2811
  cacheWrite: 0,
2676
2812
  },
2677
2813
  contextWindow: 131072,
2678
- maxTokens: 32768,
2679
- },
2680
- "llama3.1-8b": {
2681
- id: "llama3.1-8b",
2682
- name: "Llama 3.1 8B",
2683
- api: "openai-completions",
2684
- provider: "cerebras",
2685
- baseUrl: "https://api.cerebras.ai/v1",
2686
- reasoning: false,
2687
- input: ["text"],
2688
- cost: {
2689
- input: 0.1,
2690
- output: 0.1,
2691
- cacheRead: 0,
2692
- cacheWrite: 0,
2693
- },
2694
- contextWindow: 32000,
2695
- maxTokens: 8000,
2814
+ maxTokens: 40960,
2696
2815
  },
2697
2816
  "zai-glm-4.7": {
2698
2817
  id: "zai-glm-4.7",
@@ -2700,7 +2819,7 @@ export const MODELS = {
2700
2819
  api: "openai-completions",
2701
2820
  provider: "cerebras",
2702
2821
  baseUrl: "https://api.cerebras.ai/v1",
2703
- reasoning: false,
2822
+ reasoning: true,
2704
2823
  input: ["text"],
2705
2824
  cost: {
2706
2825
  input: 2.25,
@@ -2709,13 +2828,32 @@ export const MODELS = {
2709
2828
  cacheWrite: 0,
2710
2829
  },
2711
2830
  contextWindow: 131072,
2712
- maxTokens: 40000,
2831
+ maxTokens: 40960,
2713
2832
  },
2714
2833
  },
2715
2834
  "github-copilot": {
2835
+ "claude-fable-5": {
2836
+ id: "claude-fable-5",
2837
+ name: "Claude Fable 5",
2838
+ api: "openai-completions",
2839
+ provider: "github-copilot",
2840
+ baseUrl: "https://api.individual.githubcopilot.com",
2841
+ 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" },
2842
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
2843
+ reasoning: true,
2844
+ input: ["text", "image"],
2845
+ cost: {
2846
+ input: 10,
2847
+ output: 50,
2848
+ cacheRead: 1,
2849
+ cacheWrite: 12.5,
2850
+ },
2851
+ contextWindow: 1000000,
2852
+ maxTokens: 128000,
2853
+ },
2716
2854
  "claude-haiku-4.5": {
2717
2855
  id: "claude-haiku-4.5",
2718
- name: "Claude Haiku 4.5",
2856
+ name: "Claude Haiku 4.5 (latest)",
2719
2857
  api: "anthropic-messages",
2720
2858
  provider: "github-copilot",
2721
2859
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2723,17 +2861,17 @@ export const MODELS = {
2723
2861
  reasoning: true,
2724
2862
  input: ["text", "image"],
2725
2863
  cost: {
2726
- input: 0,
2727
- output: 0,
2728
- cacheRead: 0,
2729
- cacheWrite: 0,
2864
+ input: 1,
2865
+ output: 5,
2866
+ cacheRead: 0.1,
2867
+ cacheWrite: 1.25,
2730
2868
  },
2731
2869
  contextWindow: 200000,
2732
2870
  maxTokens: 64000,
2733
2871
  },
2734
2872
  "claude-opus-4.5": {
2735
2873
  id: "claude-opus-4.5",
2736
- name: "Claude Opus 4.5",
2874
+ name: "Claude Opus 4.5 (latest)",
2737
2875
  api: "anthropic-messages",
2738
2876
  provider: "github-copilot",
2739
2877
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2741,10 +2879,10 @@ export const MODELS = {
2741
2879
  reasoning: true,
2742
2880
  input: ["text", "image"],
2743
2881
  cost: {
2744
- input: 0,
2745
- output: 0,
2746
- cacheRead: 0,
2747
- cacheWrite: 0,
2882
+ input: 5,
2883
+ output: 25,
2884
+ cacheRead: 0.5,
2885
+ cacheWrite: 6.25,
2748
2886
  },
2749
2887
  contextWindow: 200000,
2750
2888
  maxTokens: 32000,
@@ -2759,10 +2897,10 @@ export const MODELS = {
2759
2897
  reasoning: true,
2760
2898
  input: ["text", "image"],
2761
2899
  cost: {
2762
- input: 0,
2763
- output: 0,
2764
- cacheRead: 0,
2765
- cacheWrite: 0,
2900
+ input: 5,
2901
+ output: 25,
2902
+ cacheRead: 0.5,
2903
+ cacheWrite: 6.25,
2766
2904
  },
2767
2905
  contextWindow: 1000000,
2768
2906
  maxTokens: 32000,
@@ -2777,10 +2915,10 @@ export const MODELS = {
2777
2915
  reasoning: true,
2778
2916
  input: ["text", "image"],
2779
2917
  cost: {
2780
- input: 0,
2781
- output: 0,
2782
- cacheRead: 0,
2783
- cacheWrite: 0,
2918
+ input: 5,
2919
+ output: 25,
2920
+ cacheRead: 0.5,
2921
+ cacheWrite: 6.25,
2784
2922
  },
2785
2923
  contextWindow: 200000,
2786
2924
  maxTokens: 32000,
@@ -2795,17 +2933,35 @@ export const MODELS = {
2795
2933
  reasoning: true,
2796
2934
  input: ["text", "image"],
2797
2935
  cost: {
2798
- input: 0,
2799
- output: 0,
2800
- cacheRead: 0,
2801
- cacheWrite: 0,
2936
+ input: 5,
2937
+ output: 25,
2938
+ cacheRead: 0.5,
2939
+ cacheWrite: 6.25,
2802
2940
  },
2803
2941
  contextWindow: 200000,
2804
2942
  maxTokens: 64000,
2805
2943
  },
2944
+ "claude-sonnet-4": {
2945
+ id: "claude-sonnet-4",
2946
+ name: "Claude Sonnet 4 (latest)",
2947
+ api: "anthropic-messages",
2948
+ provider: "github-copilot",
2949
+ baseUrl: "https://api.individual.githubcopilot.com",
2950
+ 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" },
2951
+ reasoning: true,
2952
+ input: ["text", "image"],
2953
+ cost: {
2954
+ input: 3,
2955
+ output: 15,
2956
+ cacheRead: 0.3,
2957
+ cacheWrite: 3.75,
2958
+ },
2959
+ contextWindow: 216000,
2960
+ maxTokens: 16000,
2961
+ },
2806
2962
  "claude-sonnet-4.5": {
2807
2963
  id: "claude-sonnet-4.5",
2808
- name: "Claude Sonnet 4.5",
2964
+ name: "Claude Sonnet 4.5 (latest)",
2809
2965
  api: "anthropic-messages",
2810
2966
  provider: "github-copilot",
2811
2967
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2813,10 +2969,10 @@ export const MODELS = {
2813
2969
  reasoning: true,
2814
2970
  input: ["text", "image"],
2815
2971
  cost: {
2816
- input: 0,
2817
- output: 0,
2818
- cacheRead: 0,
2819
- cacheWrite: 0,
2972
+ input: 3,
2973
+ output: 15,
2974
+ cacheRead: 0.3,
2975
+ cacheWrite: 3.75,
2820
2976
  },
2821
2977
  contextWindow: 200000,
2822
2978
  maxTokens: 32000,
@@ -2831,10 +2987,10 @@ export const MODELS = {
2831
2987
  reasoning: true,
2832
2988
  input: ["text", "image"],
2833
2989
  cost: {
2834
- input: 0,
2835
- output: 0,
2836
- cacheRead: 0,
2837
- cacheWrite: 0,
2990
+ input: 3,
2991
+ output: 15,
2992
+ cacheRead: 0.3,
2993
+ cacheWrite: 3.75,
2838
2994
  },
2839
2995
  contextWindow: 1000000,
2840
2996
  maxTokens: 32000,
@@ -2847,12 +3003,12 @@ export const MODELS = {
2847
3003
  baseUrl: "https://api.individual.githubcopilot.com",
2848
3004
  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" },
2849
3005
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
2850
- reasoning: false,
3006
+ reasoning: true,
2851
3007
  input: ["text", "image"],
2852
3008
  cost: {
2853
- input: 0,
2854
- output: 0,
2855
- cacheRead: 0,
3009
+ input: 1.25,
3010
+ output: 10,
3011
+ cacheRead: 0.125,
2856
3012
  cacheWrite: 0,
2857
3013
  },
2858
3014
  contextWindow: 128000,
@@ -2860,7 +3016,7 @@ export const MODELS = {
2860
3016
  },
2861
3017
  "gemini-3-flash-preview": {
2862
3018
  id: "gemini-3-flash-preview",
2863
- name: "Gemini 3 Flash",
3019
+ name: "Gemini 3 Flash Preview",
2864
3020
  api: "openai-completions",
2865
3021
  provider: "github-copilot",
2866
3022
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2869,9 +3025,9 @@ export const MODELS = {
2869
3025
  reasoning: true,
2870
3026
  input: ["text", "image"],
2871
3027
  cost: {
2872
- input: 0,
2873
- output: 0,
2874
- cacheRead: 0,
3028
+ input: 0.5,
3029
+ output: 3,
3030
+ cacheRead: 0.05,
2875
3031
  cacheWrite: 0,
2876
3032
  },
2877
3033
  contextWindow: 128000,
@@ -2888,9 +3044,9 @@ export const MODELS = {
2888
3044
  reasoning: true,
2889
3045
  input: ["text", "image"],
2890
3046
  cost: {
2891
- input: 0,
2892
- output: 0,
2893
- cacheRead: 0,
3047
+ input: 2,
3048
+ output: 12,
3049
+ cacheRead: 0.2,
2894
3050
  cacheWrite: 0,
2895
3051
  },
2896
3052
  contextWindow: 200000,
@@ -2907,9 +3063,9 @@ export const MODELS = {
2907
3063
  reasoning: true,
2908
3064
  input: ["text", "image"],
2909
3065
  cost: {
2910
- input: 0,
2911
- output: 0,
2912
- cacheRead: 0,
3066
+ input: 1.5,
3067
+ output: 9,
3068
+ cacheRead: 0.15,
2913
3069
  cacheWrite: 0,
2914
3070
  },
2915
3071
  contextWindow: 200000,
@@ -2926,36 +3082,17 @@ export const MODELS = {
2926
3082
  reasoning: false,
2927
3083
  input: ["text", "image"],
2928
3084
  cost: {
2929
- input: 0,
2930
- output: 0,
2931
- cacheRead: 0,
3085
+ input: 2,
3086
+ output: 8,
3087
+ cacheRead: 0.5,
2932
3088
  cacheWrite: 0,
2933
3089
  },
2934
3090
  contextWindow: 128000,
2935
3091
  maxTokens: 16384,
2936
3092
  },
2937
- "gpt-4o": {
2938
- id: "gpt-4o",
2939
- name: "GPT-4o",
2940
- api: "openai-completions",
2941
- provider: "github-copilot",
2942
- baseUrl: "https://api.individual.githubcopilot.com",
2943
- 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" },
2944
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
2945
- reasoning: false,
2946
- input: ["text", "image"],
2947
- cost: {
2948
- input: 0,
2949
- output: 0,
2950
- cacheRead: 0,
2951
- cacheWrite: 0,
2952
- },
2953
- contextWindow: 128000,
2954
- maxTokens: 4096,
2955
- },
2956
3093
  "gpt-5-mini": {
2957
3094
  id: "gpt-5-mini",
2958
- name: "GPT-5-mini",
3095
+ name: "GPT-5 Mini",
2959
3096
  api: "openai-responses",
2960
3097
  provider: "github-copilot",
2961
3098
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2963,9 +3100,9 @@ export const MODELS = {
2963
3100
  reasoning: true,
2964
3101
  input: ["text", "image"],
2965
3102
  cost: {
2966
- input: 0,
2967
- output: 0,
2968
- cacheRead: 0,
3103
+ input: 0.25,
3104
+ output: 2,
3105
+ cacheRead: 0.025,
2969
3106
  cacheWrite: 0,
2970
3107
  },
2971
3108
  contextWindow: 264000,
@@ -2981,9 +3118,9 @@ export const MODELS = {
2981
3118
  reasoning: true,
2982
3119
  input: ["text", "image"],
2983
3120
  cost: {
2984
- input: 0,
2985
- output: 0,
2986
- cacheRead: 0,
3121
+ input: 1.75,
3122
+ output: 14,
3123
+ cacheRead: 0.175,
2987
3124
  cacheWrite: 0,
2988
3125
  },
2989
3126
  contextWindow: 400000,
@@ -2991,7 +3128,7 @@ export const MODELS = {
2991
3128
  },
2992
3129
  "gpt-5.2-codex": {
2993
3130
  id: "gpt-5.2-codex",
2994
- name: "GPT-5.2-Codex",
3131
+ name: "GPT-5.2 Codex",
2995
3132
  api: "openai-responses",
2996
3133
  provider: "github-copilot",
2997
3134
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2999,9 +3136,9 @@ export const MODELS = {
2999
3136
  reasoning: true,
3000
3137
  input: ["text", "image"],
3001
3138
  cost: {
3002
- input: 0,
3003
- output: 0,
3004
- cacheRead: 0,
3139
+ input: 1.75,
3140
+ output: 14,
3141
+ cacheRead: 0.175,
3005
3142
  cacheWrite: 0,
3006
3143
  },
3007
3144
  contextWindow: 400000,
@@ -3009,7 +3146,7 @@ export const MODELS = {
3009
3146
  },
3010
3147
  "gpt-5.3-codex": {
3011
3148
  id: "gpt-5.3-codex",
3012
- name: "GPT-5.3-Codex",
3149
+ name: "GPT-5.3 Codex",
3013
3150
  api: "openai-responses",
3014
3151
  provider: "github-copilot",
3015
3152
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3017,9 +3154,9 @@ export const MODELS = {
3017
3154
  reasoning: true,
3018
3155
  input: ["text", "image"],
3019
3156
  cost: {
3020
- input: 0,
3021
- output: 0,
3022
- cacheRead: 0,
3157
+ input: 1.75,
3158
+ output: 14,
3159
+ cacheRead: 0.175,
3023
3160
  cacheWrite: 0,
3024
3161
  },
3025
3162
  contextWindow: 400000,
@@ -3035,9 +3172,9 @@ export const MODELS = {
3035
3172
  reasoning: true,
3036
3173
  input: ["text", "image"],
3037
3174
  cost: {
3038
- input: 0,
3039
- output: 0,
3040
- cacheRead: 0,
3175
+ input: 2.5,
3176
+ output: 15,
3177
+ cacheRead: 0.25,
3041
3178
  cacheWrite: 0,
3042
3179
  },
3043
3180
  contextWindow: 400000,
@@ -3045,7 +3182,7 @@ export const MODELS = {
3045
3182
  },
3046
3183
  "gpt-5.4-mini": {
3047
3184
  id: "gpt-5.4-mini",
3048
- name: "GPT-5.4 Mini",
3185
+ name: "GPT-5.4 mini",
3049
3186
  api: "openai-responses",
3050
3187
  provider: "github-copilot",
3051
3188
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3053,17 +3190,17 @@ export const MODELS = {
3053
3190
  reasoning: true,
3054
3191
  input: ["text", "image"],
3055
3192
  cost: {
3056
- input: 0,
3057
- output: 0,
3058
- cacheRead: 0,
3193
+ input: 0.75,
3194
+ output: 4.5,
3195
+ cacheRead: 0.075,
3059
3196
  cacheWrite: 0,
3060
3197
  },
3061
3198
  contextWindow: 400000,
3062
3199
  maxTokens: 128000,
3063
3200
  },
3064
- "gpt-5.5": {
3065
- id: "gpt-5.5",
3066
- name: "GPT-5.5",
3201
+ "gpt-5.4-nano": {
3202
+ id: "gpt-5.4-nano",
3203
+ name: "GPT-5.4 nano",
3067
3204
  api: "openai-responses",
3068
3205
  provider: "github-copilot",
3069
3206
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3071,32 +3208,31 @@ export const MODELS = {
3071
3208
  reasoning: true,
3072
3209
  input: ["text", "image"],
3073
3210
  cost: {
3074
- input: 0,
3075
- output: 0,
3076
- cacheRead: 0,
3211
+ input: 0.2,
3212
+ output: 1.25,
3213
+ cacheRead: 0.02,
3077
3214
  cacheWrite: 0,
3078
3215
  },
3079
3216
  contextWindow: 400000,
3080
3217
  maxTokens: 128000,
3081
3218
  },
3082
- "grok-code-fast-1": {
3083
- id: "grok-code-fast-1",
3084
- name: "Grok Code Fast 1",
3085
- api: "openai-completions",
3219
+ "gpt-5.5": {
3220
+ id: "gpt-5.5",
3221
+ name: "GPT-5.5",
3222
+ api: "openai-responses",
3086
3223
  provider: "github-copilot",
3087
3224
  baseUrl: "https://api.individual.githubcopilot.com",
3088
3225
  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" },
3089
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
3090
3226
  reasoning: true,
3091
- input: ["text"],
3227
+ input: ["text", "image"],
3092
3228
  cost: {
3093
- input: 0,
3094
- output: 0,
3095
- cacheRead: 0,
3229
+ input: 5,
3230
+ output: 30,
3231
+ cacheRead: 0.5,
3096
3232
  cacheWrite: 0,
3097
3233
  },
3098
- contextWindow: 128000,
3099
- maxTokens: 64000,
3234
+ contextWindow: 400000,
3235
+ maxTokens: 128000,
3100
3236
  },
3101
3237
  },
3102
3238
  "google": {
@@ -3839,77 +3975,9 @@ export const MODELS = {
3839
3975
  },
3840
3976
  },
3841
3977
  "groq": {
3842
- "deepseek-r1-distill-llama-70b": {
3843
- id: "deepseek-r1-distill-llama-70b",
3844
- name: "DeepSeek R1 Distill Llama 70B",
3845
- api: "openai-completions",
3846
- provider: "groq",
3847
- baseUrl: "https://api.groq.com/openai/v1",
3848
- reasoning: true,
3849
- input: ["text"],
3850
- cost: {
3851
- input: 0.75,
3852
- output: 0.99,
3853
- cacheRead: 0,
3854
- cacheWrite: 0,
3855
- },
3856
- contextWindow: 131072,
3857
- maxTokens: 8192,
3858
- },
3859
- "gemma2-9b-it": {
3860
- id: "gemma2-9b-it",
3861
- name: "Gemma 2 9B",
3862
- api: "openai-completions",
3863
- provider: "groq",
3864
- baseUrl: "https://api.groq.com/openai/v1",
3865
- reasoning: false,
3866
- input: ["text"],
3867
- cost: {
3868
- input: 0.2,
3869
- output: 0.2,
3870
- cacheRead: 0,
3871
- cacheWrite: 0,
3872
- },
3873
- contextWindow: 8192,
3874
- maxTokens: 8192,
3875
- },
3876
- "groq/compound": {
3877
- id: "groq/compound",
3878
- name: "Compound",
3879
- api: "openai-completions",
3880
- provider: "groq",
3881
- baseUrl: "https://api.groq.com/openai/v1",
3882
- reasoning: true,
3883
- input: ["text"],
3884
- cost: {
3885
- input: 0,
3886
- output: 0,
3887
- cacheRead: 0,
3888
- cacheWrite: 0,
3889
- },
3890
- contextWindow: 131072,
3891
- maxTokens: 8192,
3892
- },
3893
- "groq/compound-mini": {
3894
- id: "groq/compound-mini",
3895
- name: "Compound Mini",
3896
- api: "openai-completions",
3897
- provider: "groq",
3898
- baseUrl: "https://api.groq.com/openai/v1",
3899
- reasoning: true,
3900
- input: ["text"],
3901
- cost: {
3902
- input: 0,
3903
- output: 0,
3904
- cacheRead: 0,
3905
- cacheWrite: 0,
3906
- },
3907
- contextWindow: 131072,
3908
- maxTokens: 8192,
3909
- },
3910
3978
  "llama-3.1-8b-instant": {
3911
3979
  id: "llama-3.1-8b-instant",
3912
- name: "Llama 3.1 8B Instant",
3980
+ name: "Llama 3.1 8B",
3913
3981
  api: "openai-completions",
3914
3982
  provider: "groq",
3915
3983
  baseUrl: "https://api.groq.com/openai/v1",
@@ -3926,7 +3994,7 @@ export const MODELS = {
3926
3994
  },
3927
3995
  "llama-3.3-70b-versatile": {
3928
3996
  id: "llama-3.3-70b-versatile",
3929
- name: "Llama 3.3 70B Versatile",
3997
+ name: "Llama 3.3 70B",
3930
3998
  api: "openai-completions",
3931
3999
  provider: "groq",
3932
4000
  baseUrl: "https://api.groq.com/openai/v1",
@@ -3941,60 +4009,9 @@ export const MODELS = {
3941
4009
  contextWindow: 131072,
3942
4010
  maxTokens: 32768,
3943
4011
  },
3944
- "llama3-70b-8192": {
3945
- id: "llama3-70b-8192",
3946
- name: "Llama 3 70B",
3947
- api: "openai-completions",
3948
- provider: "groq",
3949
- baseUrl: "https://api.groq.com/openai/v1",
3950
- reasoning: false,
3951
- input: ["text"],
3952
- cost: {
3953
- input: 0.59,
3954
- output: 0.79,
3955
- cacheRead: 0,
3956
- cacheWrite: 0,
3957
- },
3958
- contextWindow: 8192,
3959
- maxTokens: 8192,
3960
- },
3961
- "llama3-8b-8192": {
3962
- id: "llama3-8b-8192",
3963
- name: "Llama 3 8B",
3964
- api: "openai-completions",
3965
- provider: "groq",
3966
- baseUrl: "https://api.groq.com/openai/v1",
3967
- reasoning: false,
3968
- input: ["text"],
3969
- cost: {
3970
- input: 0.05,
3971
- output: 0.08,
3972
- cacheRead: 0,
3973
- cacheWrite: 0,
3974
- },
3975
- contextWindow: 8192,
3976
- maxTokens: 8192,
3977
- },
3978
- "meta-llama/llama-4-maverick-17b-128e-instruct": {
3979
- id: "meta-llama/llama-4-maverick-17b-128e-instruct",
3980
- name: "Llama 4 Maverick 17B",
3981
- api: "openai-completions",
3982
- provider: "groq",
3983
- baseUrl: "https://api.groq.com/openai/v1",
3984
- reasoning: false,
3985
- input: ["text", "image"],
3986
- cost: {
3987
- input: 0.2,
3988
- output: 0.6,
3989
- cacheRead: 0,
3990
- cacheWrite: 0,
3991
- },
3992
- contextWindow: 131072,
3993
- maxTokens: 8192,
3994
- },
3995
4012
  "meta-llama/llama-4-scout-17b-16e-instruct": {
3996
4013
  id: "meta-llama/llama-4-scout-17b-16e-instruct",
3997
- name: "Llama 4 Scout 17B",
4014
+ name: "Llama 4 Scout 17B 16E",
3998
4015
  api: "openai-completions",
3999
4016
  provider: "groq",
4000
4017
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4009,57 +4026,6 @@ export const MODELS = {
4009
4026
  contextWindow: 131072,
4010
4027
  maxTokens: 8192,
4011
4028
  },
4012
- "mistral-saba-24b": {
4013
- id: "mistral-saba-24b",
4014
- name: "Mistral Saba 24B",
4015
- api: "openai-completions",
4016
- provider: "groq",
4017
- baseUrl: "https://api.groq.com/openai/v1",
4018
- reasoning: false,
4019
- input: ["text"],
4020
- cost: {
4021
- input: 0.79,
4022
- output: 0.79,
4023
- cacheRead: 0,
4024
- cacheWrite: 0,
4025
- },
4026
- contextWindow: 32768,
4027
- maxTokens: 32768,
4028
- },
4029
- "moonshotai/kimi-k2-instruct": {
4030
- id: "moonshotai/kimi-k2-instruct",
4031
- name: "Kimi K2 Instruct",
4032
- api: "openai-completions",
4033
- provider: "groq",
4034
- baseUrl: "https://api.groq.com/openai/v1",
4035
- reasoning: false,
4036
- input: ["text"],
4037
- cost: {
4038
- input: 1,
4039
- output: 3,
4040
- cacheRead: 0,
4041
- cacheWrite: 0,
4042
- },
4043
- contextWindow: 131072,
4044
- maxTokens: 16384,
4045
- },
4046
- "moonshotai/kimi-k2-instruct-0905": {
4047
- id: "moonshotai/kimi-k2-instruct-0905",
4048
- name: "Kimi K2 Instruct 0905",
4049
- api: "openai-completions",
4050
- provider: "groq",
4051
- baseUrl: "https://api.groq.com/openai/v1",
4052
- reasoning: false,
4053
- input: ["text"],
4054
- cost: {
4055
- input: 1,
4056
- output: 3,
4057
- cacheRead: 0.5,
4058
- cacheWrite: 0,
4059
- },
4060
- contextWindow: 262144,
4061
- maxTokens: 16384,
4062
- },
4063
4029
  "openai/gpt-oss-120b": {
4064
4030
  id: "openai/gpt-oss-120b",
4065
4031
  name: "GPT OSS 120B",
@@ -4111,9 +4077,9 @@ export const MODELS = {
4111
4077
  contextWindow: 131072,
4112
4078
  maxTokens: 65536,
4113
4079
  },
4114
- "qwen-qwq-32b": {
4115
- id: "qwen-qwq-32b",
4116
- name: "Qwen QwQ 32B",
4080
+ "qwen/qwen3-32b": {
4081
+ id: "qwen/qwen3-32b",
4082
+ name: "Qwen3-32B",
4117
4083
  api: "openai-completions",
4118
4084
  provider: "groq",
4119
4085
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4121,32 +4087,33 @@ export const MODELS = {
4121
4087
  input: ["text"],
4122
4088
  cost: {
4123
4089
  input: 0.29,
4124
- output: 0.39,
4090
+ output: 0.59,
4125
4091
  cacheRead: 0,
4126
4092
  cacheWrite: 0,
4127
4093
  },
4128
4094
  contextWindow: 131072,
4129
- maxTokens: 16384,
4095
+ maxTokens: 40960,
4130
4096
  },
4131
- "qwen/qwen3-32b": {
4132
- id: "qwen/qwen3-32b",
4133
- name: "Qwen3 32B",
4097
+ },
4098
+ "huggingface": {
4099
+ "MiniMaxAI/MiniMax-M2": {
4100
+ id: "MiniMaxAI/MiniMax-M2",
4101
+ name: "MiniMax-M2",
4134
4102
  api: "openai-completions",
4135
- provider: "groq",
4136
- baseUrl: "https://api.groq.com/openai/v1",
4103
+ provider: "huggingface",
4104
+ baseUrl: "https://router.huggingface.co/v1",
4105
+ compat: { "supportsDeveloperRole": false },
4137
4106
  reasoning: true,
4138
4107
  input: ["text"],
4139
4108
  cost: {
4140
- input: 0.29,
4141
- output: 0.59,
4109
+ input: 0.3,
4110
+ output: 1.2,
4142
4111
  cacheRead: 0,
4143
4112
  cacheWrite: 0,
4144
4113
  },
4145
- contextWindow: 131072,
4146
- maxTokens: 40960,
4114
+ contextWindow: 204800,
4115
+ maxTokens: 128000,
4147
4116
  },
4148
- },
4149
- "huggingface": {
4150
4117
  "MiniMaxAI/MiniMax-M2.1": {
4151
4118
  id: "MiniMaxAI/MiniMax-M2.1",
4152
4119
  name: "MiniMax-M2.1",
@@ -4201,6 +4168,42 @@ export const MODELS = {
4201
4168
  contextWindow: 204800,
4202
4169
  maxTokens: 131072,
4203
4170
  },
4171
+ "MiniMaxAI/MiniMax-M3": {
4172
+ id: "MiniMaxAI/MiniMax-M3",
4173
+ name: "MiniMax-M3",
4174
+ api: "openai-completions",
4175
+ provider: "huggingface",
4176
+ baseUrl: "https://router.huggingface.co/v1",
4177
+ compat: { "supportsDeveloperRole": false },
4178
+ reasoning: true,
4179
+ input: ["text", "image"],
4180
+ cost: {
4181
+ input: 0.3,
4182
+ output: 1.2,
4183
+ cacheRead: 0,
4184
+ cacheWrite: 0,
4185
+ },
4186
+ contextWindow: 524288,
4187
+ maxTokens: 128000,
4188
+ },
4189
+ "Qwen/Qwen3-235B-A22B": {
4190
+ id: "Qwen/Qwen3-235B-A22B",
4191
+ name: "Qwen3 235B-A22B",
4192
+ api: "openai-completions",
4193
+ provider: "huggingface",
4194
+ baseUrl: "https://router.huggingface.co/v1",
4195
+ compat: { "supportsDeveloperRole": false },
4196
+ reasoning: true,
4197
+ input: ["text"],
4198
+ cost: {
4199
+ input: 0.2,
4200
+ output: 0.8,
4201
+ cacheRead: 0,
4202
+ cacheWrite: 0,
4203
+ },
4204
+ contextWindow: 40960,
4205
+ maxTokens: 16384,
4206
+ },
4204
4207
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
4205
4208
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
4206
4209
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -4219,6 +4222,42 @@ export const MODELS = {
4219
4222
  contextWindow: 262144,
4220
4223
  maxTokens: 131072,
4221
4224
  },
4225
+ "Qwen/Qwen3-32B": {
4226
+ id: "Qwen/Qwen3-32B",
4227
+ name: "Qwen3 32B",
4228
+ api: "openai-completions",
4229
+ provider: "huggingface",
4230
+ baseUrl: "https://router.huggingface.co/v1",
4231
+ compat: { "supportsDeveloperRole": false },
4232
+ reasoning: true,
4233
+ input: ["text"],
4234
+ cost: {
4235
+ input: 0.29,
4236
+ output: 0.59,
4237
+ cacheRead: 0,
4238
+ cacheWrite: 0,
4239
+ },
4240
+ contextWindow: 131072,
4241
+ maxTokens: 16384,
4242
+ },
4243
+ "Qwen/Qwen3-Coder-30B-A3B-Instruct": {
4244
+ id: "Qwen/Qwen3-Coder-30B-A3B-Instruct",
4245
+ name: "Qwen3-Coder 30B-A3B Instruct",
4246
+ api: "openai-completions",
4247
+ provider: "huggingface",
4248
+ baseUrl: "https://router.huggingface.co/v1",
4249
+ compat: { "supportsDeveloperRole": false },
4250
+ reasoning: false,
4251
+ input: ["text"],
4252
+ cost: {
4253
+ input: 0.07,
4254
+ output: 0.26,
4255
+ cacheRead: 0,
4256
+ cacheWrite: 0,
4257
+ },
4258
+ contextWindow: 262144,
4259
+ maxTokens: 65536,
4260
+ },
4222
4261
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
4223
4262
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
4224
4263
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -4291,6 +4330,60 @@ export const MODELS = {
4291
4330
  contextWindow: 262144,
4292
4331
  maxTokens: 131072,
4293
4332
  },
4333
+ "Qwen/Qwen3.5-122B-A10B": {
4334
+ id: "Qwen/Qwen3.5-122B-A10B",
4335
+ name: "Qwen3.5 122B-A10B",
4336
+ api: "openai-completions",
4337
+ provider: "huggingface",
4338
+ baseUrl: "https://router.huggingface.co/v1",
4339
+ compat: { "supportsDeveloperRole": false },
4340
+ reasoning: true,
4341
+ input: ["text", "image"],
4342
+ cost: {
4343
+ input: 0.4,
4344
+ output: 3.2,
4345
+ cacheRead: 0,
4346
+ cacheWrite: 0,
4347
+ },
4348
+ contextWindow: 262144,
4349
+ maxTokens: 65536,
4350
+ },
4351
+ "Qwen/Qwen3.5-27B": {
4352
+ id: "Qwen/Qwen3.5-27B",
4353
+ name: "Qwen3.5 27B",
4354
+ api: "openai-completions",
4355
+ provider: "huggingface",
4356
+ baseUrl: "https://router.huggingface.co/v1",
4357
+ compat: { "supportsDeveloperRole": false },
4358
+ reasoning: true,
4359
+ input: ["text", "image"],
4360
+ cost: {
4361
+ input: 0.3,
4362
+ output: 2.4,
4363
+ cacheRead: 0,
4364
+ cacheWrite: 0,
4365
+ },
4366
+ contextWindow: 262144,
4367
+ maxTokens: 65536,
4368
+ },
4369
+ "Qwen/Qwen3.5-35B-A3B": {
4370
+ id: "Qwen/Qwen3.5-35B-A3B",
4371
+ name: "Qwen3.5 35B-A3B",
4372
+ api: "openai-completions",
4373
+ provider: "huggingface",
4374
+ baseUrl: "https://router.huggingface.co/v1",
4375
+ compat: { "supportsDeveloperRole": false },
4376
+ reasoning: true,
4377
+ input: ["text", "image"],
4378
+ cost: {
4379
+ input: 0.25,
4380
+ output: 2,
4381
+ cacheRead: 0,
4382
+ cacheWrite: 0,
4383
+ },
4384
+ contextWindow: 262144,
4385
+ maxTokens: 65536,
4386
+ },
4294
4387
  "Qwen/Qwen3.5-397B-A17B": {
4295
4388
  id: "Qwen/Qwen3.5-397B-A17B",
4296
4389
  name: "Qwen3.5-397B-A17B",
@@ -4309,6 +4402,60 @@ export const MODELS = {
4309
4402
  contextWindow: 262144,
4310
4403
  maxTokens: 32768,
4311
4404
  },
4405
+ "Qwen/Qwen3.5-9B": {
4406
+ id: "Qwen/Qwen3.5-9B",
4407
+ name: "Qwen3.5 9B",
4408
+ api: "openai-completions",
4409
+ provider: "huggingface",
4410
+ baseUrl: "https://router.huggingface.co/v1",
4411
+ compat: { "supportsDeveloperRole": false },
4412
+ reasoning: true,
4413
+ input: ["text", "image"],
4414
+ cost: {
4415
+ input: 0.17,
4416
+ output: 0.25,
4417
+ cacheRead: 0,
4418
+ cacheWrite: 0,
4419
+ },
4420
+ contextWindow: 262144,
4421
+ maxTokens: 65536,
4422
+ },
4423
+ "Qwen/Qwen3.6-27B": {
4424
+ id: "Qwen/Qwen3.6-27B",
4425
+ name: "Qwen3.6 27B",
4426
+ api: "openai-completions",
4427
+ provider: "huggingface",
4428
+ baseUrl: "https://router.huggingface.co/v1",
4429
+ compat: { "supportsDeveloperRole": false },
4430
+ reasoning: true,
4431
+ input: ["text", "image"],
4432
+ cost: {
4433
+ input: 0.47,
4434
+ output: 3.19,
4435
+ cacheRead: 0,
4436
+ cacheWrite: 0,
4437
+ },
4438
+ contextWindow: 262144,
4439
+ maxTokens: 65536,
4440
+ },
4441
+ "Qwen/Qwen3.6-35B-A3B": {
4442
+ id: "Qwen/Qwen3.6-35B-A3B",
4443
+ name: "Qwen3.6 35B-A3B",
4444
+ api: "openai-completions",
4445
+ provider: "huggingface",
4446
+ baseUrl: "https://router.huggingface.co/v1",
4447
+ compat: { "supportsDeveloperRole": false },
4448
+ reasoning: true,
4449
+ input: ["text", "image"],
4450
+ cost: {
4451
+ input: 0.15,
4452
+ output: 0.95,
4453
+ cacheRead: 0,
4454
+ cacheWrite: 0,
4455
+ },
4456
+ contextWindow: 262144,
4457
+ maxTokens: 65536,
4458
+ },
4312
4459
  "XiaomiMiMo/MiMo-V2-Flash": {
4313
4460
  id: "XiaomiMiMo/MiMo-V2-Flash",
4314
4461
  name: "MiMo-V2-Flash",
@@ -4327,6 +4474,42 @@ export const MODELS = {
4327
4474
  contextWindow: 262144,
4328
4475
  maxTokens: 4096,
4329
4476
  },
4477
+ "XiaomiMiMo/MiMo-V2.5-Pro": {
4478
+ id: "XiaomiMiMo/MiMo-V2.5-Pro",
4479
+ name: "MiMo-V2.5-Pro",
4480
+ api: "openai-completions",
4481
+ provider: "huggingface",
4482
+ baseUrl: "https://router.huggingface.co/v1",
4483
+ compat: { "supportsDeveloperRole": false },
4484
+ reasoning: true,
4485
+ input: ["text"],
4486
+ cost: {
4487
+ input: 1,
4488
+ output: 3,
4489
+ cacheRead: 0,
4490
+ cacheWrite: 0,
4491
+ },
4492
+ contextWindow: 1048576,
4493
+ maxTokens: 131072,
4494
+ },
4495
+ "deepseek-ai/DeepSeek-R1": {
4496
+ id: "deepseek-ai/DeepSeek-R1",
4497
+ name: "DeepSeek-R1",
4498
+ api: "openai-completions",
4499
+ provider: "huggingface",
4500
+ baseUrl: "https://router.huggingface.co/v1",
4501
+ compat: { "supportsDeveloperRole": false },
4502
+ reasoning: true,
4503
+ input: ["text"],
4504
+ cost: {
4505
+ input: 0.7,
4506
+ output: 2.5,
4507
+ cacheRead: 0,
4508
+ cacheWrite: 0,
4509
+ },
4510
+ contextWindow: 64000,
4511
+ maxTokens: 32768,
4512
+ },
4330
4513
  "deepseek-ai/DeepSeek-R1-0528": {
4331
4514
  id: "deepseek-ai/DeepSeek-R1-0528",
4332
4515
  name: "DeepSeek-R1-0528",
@@ -4363,6 +4546,24 @@ export const MODELS = {
4363
4546
  contextWindow: 163840,
4364
4547
  maxTokens: 65536,
4365
4548
  },
4549
+ "deepseek-ai/DeepSeek-V4-Flash": {
4550
+ id: "deepseek-ai/DeepSeek-V4-Flash",
4551
+ name: "DeepSeek V4 Flash",
4552
+ api: "openai-completions",
4553
+ provider: "huggingface",
4554
+ baseUrl: "https://router.huggingface.co/v1",
4555
+ compat: { "supportsDeveloperRole": false },
4556
+ reasoning: true,
4557
+ input: ["text"],
4558
+ cost: {
4559
+ input: 0.14,
4560
+ output: 0.28,
4561
+ cacheRead: 0,
4562
+ cacheWrite: 0,
4563
+ },
4564
+ contextWindow: 1048576,
4565
+ maxTokens: 384000,
4566
+ },
4366
4567
  "deepseek-ai/DeepSeek-V4-Pro": {
4367
4568
  id: "deepseek-ai/DeepSeek-V4-Pro",
4368
4569
  name: "DeepSeek V4 Pro",
@@ -4381,6 +4582,60 @@ export const MODELS = {
4381
4582
  contextWindow: 1048576,
4382
4583
  maxTokens: 393216,
4383
4584
  },
4585
+ "google/gemma-4-26B-A4B-it": {
4586
+ id: "google/gemma-4-26B-A4B-it",
4587
+ name: "Gemma 4 26B A4B IT",
4588
+ api: "openai-completions",
4589
+ provider: "huggingface",
4590
+ baseUrl: "https://router.huggingface.co/v1",
4591
+ compat: { "supportsDeveloperRole": false },
4592
+ reasoning: true,
4593
+ input: ["text", "image"],
4594
+ cost: {
4595
+ input: 0.13,
4596
+ output: 0.4,
4597
+ cacheRead: 0,
4598
+ cacheWrite: 0,
4599
+ },
4600
+ contextWindow: 262144,
4601
+ maxTokens: 32768,
4602
+ },
4603
+ "google/gemma-4-31B-it": {
4604
+ id: "google/gemma-4-31B-it",
4605
+ name: "Gemma 4 31B IT",
4606
+ api: "openai-completions",
4607
+ provider: "huggingface",
4608
+ baseUrl: "https://router.huggingface.co/v1",
4609
+ compat: { "supportsDeveloperRole": false },
4610
+ reasoning: true,
4611
+ input: ["text", "image"],
4612
+ cost: {
4613
+ input: 0.14,
4614
+ output: 0.4,
4615
+ cacheRead: 0,
4616
+ cacheWrite: 0,
4617
+ },
4618
+ contextWindow: 262144,
4619
+ maxTokens: 32768,
4620
+ },
4621
+ "meta-llama/Llama-3.3-70B-Instruct": {
4622
+ id: "meta-llama/Llama-3.3-70B-Instruct",
4623
+ name: "Llama-3.3-70B-Instruct",
4624
+ api: "openai-completions",
4625
+ provider: "huggingface",
4626
+ baseUrl: "https://router.huggingface.co/v1",
4627
+ compat: { "supportsDeveloperRole": false },
4628
+ reasoning: false,
4629
+ input: ["text"],
4630
+ cost: {
4631
+ input: 0.59,
4632
+ output: 0.79,
4633
+ cacheRead: 0,
4634
+ cacheWrite: 0,
4635
+ },
4636
+ contextWindow: 131072,
4637
+ maxTokens: 4096,
4638
+ },
4384
4639
  "moonshotai/Kimi-K2-Instruct": {
4385
4640
  id: "moonshotai/Kimi-K2-Instruct",
4386
4641
  name: "Kimi-K2-Instruct",
@@ -4468,8 +4723,134 @@ export const MODELS = {
4468
4723
  cacheRead: 0.16,
4469
4724
  cacheWrite: 0,
4470
4725
  },
4471
- contextWindow: 262144,
4472
- maxTokens: 262144,
4726
+ contextWindow: 262144,
4727
+ maxTokens: 262144,
4728
+ },
4729
+ "moonshotai/Kimi-K2.7-Code": {
4730
+ id: "moonshotai/Kimi-K2.7-Code",
4731
+ name: "Kimi K2.7 Code",
4732
+ api: "openai-completions",
4733
+ provider: "huggingface",
4734
+ baseUrl: "https://router.huggingface.co/v1",
4735
+ compat: { "supportsDeveloperRole": false },
4736
+ reasoning: true,
4737
+ input: ["text", "image"],
4738
+ cost: {
4739
+ input: 0.95,
4740
+ output: 4,
4741
+ cacheRead: 0,
4742
+ cacheWrite: 0,
4743
+ },
4744
+ contextWindow: 262144,
4745
+ maxTokens: 262144,
4746
+ },
4747
+ "stepfun-ai/Step-3.5-Flash": {
4748
+ id: "stepfun-ai/Step-3.5-Flash",
4749
+ name: "Step 3.5 Flash",
4750
+ api: "openai-completions",
4751
+ provider: "huggingface",
4752
+ baseUrl: "https://router.huggingface.co/v1",
4753
+ compat: { "supportsDeveloperRole": false },
4754
+ reasoning: true,
4755
+ input: ["text"],
4756
+ cost: {
4757
+ input: 0.1,
4758
+ output: 0.3,
4759
+ cacheRead: 0,
4760
+ cacheWrite: 0,
4761
+ },
4762
+ contextWindow: 262144,
4763
+ maxTokens: 256000,
4764
+ },
4765
+ "stepfun-ai/Step-3.7-Flash": {
4766
+ id: "stepfun-ai/Step-3.7-Flash",
4767
+ name: "Step 3.7 Flash",
4768
+ api: "openai-completions",
4769
+ provider: "huggingface",
4770
+ baseUrl: "https://router.huggingface.co/v1",
4771
+ compat: { "supportsDeveloperRole": false },
4772
+ reasoning: true,
4773
+ input: ["text", "image"],
4774
+ cost: {
4775
+ input: 0.2,
4776
+ output: 1.15,
4777
+ cacheRead: 0,
4778
+ cacheWrite: 0,
4779
+ },
4780
+ contextWindow: 262144,
4781
+ maxTokens: 256000,
4782
+ },
4783
+ "zai-org/GLM-4.5": {
4784
+ id: "zai-org/GLM-4.5",
4785
+ name: "GLM-4.5",
4786
+ api: "openai-completions",
4787
+ provider: "huggingface",
4788
+ baseUrl: "https://router.huggingface.co/v1",
4789
+ compat: { "supportsDeveloperRole": false },
4790
+ reasoning: true,
4791
+ input: ["text"],
4792
+ cost: {
4793
+ input: 0.6,
4794
+ output: 2.2,
4795
+ cacheRead: 0,
4796
+ cacheWrite: 0,
4797
+ },
4798
+ contextWindow: 131072,
4799
+ maxTokens: 98304,
4800
+ },
4801
+ "zai-org/GLM-4.5-Air": {
4802
+ id: "zai-org/GLM-4.5-Air",
4803
+ name: "GLM-4.5-Air",
4804
+ api: "openai-completions",
4805
+ provider: "huggingface",
4806
+ baseUrl: "https://router.huggingface.co/v1",
4807
+ compat: { "supportsDeveloperRole": false },
4808
+ reasoning: true,
4809
+ input: ["text"],
4810
+ cost: {
4811
+ input: 0.13,
4812
+ output: 0.85,
4813
+ cacheRead: 0,
4814
+ cacheWrite: 0,
4815
+ },
4816
+ contextWindow: 131072,
4817
+ maxTokens: 98304,
4818
+ },
4819
+ "zai-org/GLM-4.5V": {
4820
+ id: "zai-org/GLM-4.5V",
4821
+ name: "GLM-4.5V",
4822
+ api: "openai-completions",
4823
+ provider: "huggingface",
4824
+ baseUrl: "https://router.huggingface.co/v1",
4825
+ compat: { "supportsDeveloperRole": false },
4826
+ reasoning: true,
4827
+ input: ["text", "image"],
4828
+ cost: {
4829
+ input: 0.6,
4830
+ output: 1.8,
4831
+ cacheRead: 0,
4832
+ cacheWrite: 0,
4833
+ },
4834
+ contextWindow: 65536,
4835
+ maxTokens: 16384,
4836
+ },
4837
+ "zai-org/GLM-4.6": {
4838
+ id: "zai-org/GLM-4.6",
4839
+ name: "GLM-4.6",
4840
+ api: "openai-completions",
4841
+ provider: "huggingface",
4842
+ baseUrl: "https://router.huggingface.co/v1",
4843
+ compat: { "supportsDeveloperRole": false },
4844
+ reasoning: true,
4845
+ input: ["text"],
4846
+ cost: {
4847
+ input: 0.55,
4848
+ output: 2.2,
4849
+ cacheRead: 0,
4850
+ cacheWrite: 0,
4851
+ },
4852
+ contextWindow: 204800,
4853
+ maxTokens: 131072,
4473
4854
  },
4474
4855
  "zai-org/GLM-4.7": {
4475
4856
  id: "zai-org/GLM-4.7",
@@ -4543,6 +4924,24 @@ export const MODELS = {
4543
4924
  contextWindow: 202752,
4544
4925
  maxTokens: 131072,
4545
4926
  },
4927
+ "zai-org/GLM-5.2": {
4928
+ id: "zai-org/GLM-5.2",
4929
+ name: "GLM-5.2",
4930
+ api: "openai-completions",
4931
+ provider: "huggingface",
4932
+ baseUrl: "https://router.huggingface.co/v1",
4933
+ compat: { "supportsDeveloperRole": false },
4934
+ reasoning: true,
4935
+ input: ["text"],
4936
+ cost: {
4937
+ input: 1.4,
4938
+ output: 4.4,
4939
+ cacheRead: 0,
4940
+ cacheWrite: 0,
4941
+ },
4942
+ contextWindow: 262144,
4943
+ maxTokens: 131072,
4944
+ },
4546
4945
  },
4547
4946
  "kimi-coding": {
4548
4947
  "k2p5": {
@@ -4579,6 +4978,23 @@ export const MODELS = {
4579
4978
  contextWindow: 262144,
4580
4979
  maxTokens: 32768,
4581
4980
  },
4981
+ "k2p7": {
4982
+ id: "k2p7",
4983
+ name: "Kimi K2.7 Code",
4984
+ api: "anthropic-messages",
4985
+ provider: "kimi-coding",
4986
+ baseUrl: "https://api.kimi.com/coding",
4987
+ reasoning: true,
4988
+ input: ["text", "image"],
4989
+ cost: {
4990
+ input: 0,
4991
+ output: 0,
4992
+ cacheRead: 0,
4993
+ cacheWrite: 0,
4994
+ },
4995
+ contextWindow: 262144,
4996
+ maxTokens: 32768,
4997
+ },
4582
4998
  "kimi-k2-thinking": {
4583
4999
  id: "kimi-k2-thinking",
4584
5000
  name: "Kimi K2 Thinking",
@@ -4700,6 +5116,23 @@ export const MODELS = {
4700
5116
  contextWindow: 204800,
4701
5117
  maxTokens: 131072,
4702
5118
  },
5119
+ "MiniMax-M3": {
5120
+ id: "MiniMax-M3",
5121
+ name: "MiniMax-M3",
5122
+ api: "anthropic-messages",
5123
+ provider: "minimax",
5124
+ baseUrl: "https://api.minimax.io/anthropic",
5125
+ reasoning: true,
5126
+ input: ["text", "image"],
5127
+ cost: {
5128
+ input: 0.3,
5129
+ output: 1.2,
5130
+ cacheRead: 0.06,
5131
+ cacheWrite: 0,
5132
+ },
5133
+ contextWindow: 1000000,
5134
+ maxTokens: 128000,
5135
+ },
4703
5136
  },
4704
5137
  "minimax-cn": {
4705
5138
  "MiniMax-M2": {
@@ -4804,6 +5237,23 @@ export const MODELS = {
4804
5237
  contextWindow: 204800,
4805
5238
  maxTokens: 131072,
4806
5239
  },
5240
+ "MiniMax-M3": {
5241
+ id: "MiniMax-M3",
5242
+ name: "MiniMax-M3",
5243
+ api: "anthropic-messages",
5244
+ provider: "minimax-cn",
5245
+ baseUrl: "https://api.minimaxi.com/anthropic",
5246
+ reasoning: true,
5247
+ input: ["text", "image"],
5248
+ cost: {
5249
+ input: 0.3,
5250
+ output: 1.2,
5251
+ cacheRead: 0.06,
5252
+ cacheWrite: 0,
5253
+ },
5254
+ contextWindow: 1000000,
5255
+ maxTokens: 128000,
5256
+ },
4807
5257
  },
4808
5258
  "mistral": {
4809
5259
  "codestral-latest": {
@@ -4840,6 +5290,23 @@ export const MODELS = {
4840
5290
  contextWindow: 262144,
4841
5291
  maxTokens: 262144,
4842
5292
  },
5293
+ "devstral-latest": {
5294
+ id: "devstral-latest",
5295
+ name: "Devstral 2",
5296
+ api: "mistral-conversations",
5297
+ provider: "mistral",
5298
+ baseUrl: "https://api.mistral.ai",
5299
+ reasoning: false,
5300
+ input: ["text"],
5301
+ cost: {
5302
+ input: 0.4,
5303
+ output: 2,
5304
+ cacheRead: 0,
5305
+ cacheWrite: 0,
5306
+ },
5307
+ contextWindow: 262144,
5308
+ maxTokens: 262144,
5309
+ },
4843
5310
  "devstral-medium-2507": {
4844
5311
  id: "devstral-medium-2507",
4845
5312
  name: "Devstral Medium",
@@ -5101,11 +5568,11 @@ export const MODELS = {
5101
5568
  api: "mistral-conversations",
5102
5569
  provider: "mistral",
5103
5570
  baseUrl: "https://api.mistral.ai",
5104
- reasoning: true,
5571
+ reasoning: false,
5105
5572
  input: ["text", "image"],
5106
5573
  cost: {
5107
- input: 1.5,
5108
- output: 7.5,
5574
+ input: 0.4,
5575
+ output: 2,
5109
5576
  cacheRead: 0,
5110
5577
  cacheWrite: 0,
5111
5578
  },
@@ -5197,6 +5664,23 @@ export const MODELS = {
5197
5664
  contextWindow: 8000,
5198
5665
  maxTokens: 8000,
5199
5666
  },
5667
+ "open-mistral-nemo": {
5668
+ id: "open-mistral-nemo",
5669
+ name: "Open Mistral Nemo",
5670
+ api: "mistral-conversations",
5671
+ provider: "mistral",
5672
+ baseUrl: "https://api.mistral.ai",
5673
+ reasoning: false,
5674
+ input: ["text"],
5675
+ cost: {
5676
+ input: 0.15,
5677
+ output: 0.15,
5678
+ cacheRead: 0,
5679
+ cacheWrite: 0,
5680
+ },
5681
+ contextWindow: 128000,
5682
+ maxTokens: 128000,
5683
+ },
5200
5684
  "open-mixtral-8x22b": {
5201
5685
  id: "open-mixtral-8x22b",
5202
5686
  name: "Mixtral 8x22B",
@@ -6319,7 +6803,7 @@ export const MODELS = {
6319
6803
  cost: {
6320
6804
  input: 0.14,
6321
6805
  output: 0.28,
6322
- cacheRead: 0.03,
6806
+ cacheRead: 0.028,
6323
6807
  cacheWrite: 0,
6324
6808
  },
6325
6809
  contextWindow: 1000000,
@@ -6342,6 +6826,23 @@ export const MODELS = {
6342
6826
  contextWindow: 200000,
6343
6827
  maxTokens: 128000,
6344
6828
  },
6829
+ "deepseek-v4-pro": {
6830
+ id: "deepseek-v4-pro",
6831
+ name: "DeepSeek V4 Pro",
6832
+ api: "openai-completions",
6833
+ provider: "opencode",
6834
+ baseUrl: "https://opencode.ai/zen/v1",
6835
+ reasoning: true,
6836
+ input: ["text"],
6837
+ cost: {
6838
+ input: 1.74,
6839
+ output: 3.84,
6840
+ cacheRead: 0.145,
6841
+ cacheWrite: 0,
6842
+ },
6843
+ contextWindow: 1000000,
6844
+ maxTokens: 384000,
6845
+ },
6345
6846
  "gemini-3-flash": {
6346
6847
  id: "gemini-3-flash",
6347
6848
  name: "Gemini 3 Flash",
@@ -6427,6 +6928,23 @@ export const MODELS = {
6427
6928
  contextWindow: 204800,
6428
6929
  maxTokens: 131072,
6429
6930
  },
6931
+ "glm-5.2": {
6932
+ id: "glm-5.2",
6933
+ name: "GLM-5.2",
6934
+ api: "openai-completions",
6935
+ provider: "opencode",
6936
+ baseUrl: "https://opencode.ai/zen/v1",
6937
+ reasoning: true,
6938
+ input: ["text"],
6939
+ cost: {
6940
+ input: 1.4,
6941
+ output: 4.4,
6942
+ cacheRead: 0.26,
6943
+ cacheWrite: 0,
6944
+ },
6945
+ contextWindow: 1000000,
6946
+ maxTokens: 131072,
6947
+ },
6430
6948
  "gpt-5": {
6431
6949
  id: "gpt-5",
6432
6950
  name: "GPT-5",
@@ -6801,26 +7319,26 @@ export const MODELS = {
6801
7319
  contextWindow: 204800,
6802
7320
  maxTokens: 131072,
6803
7321
  },
6804
- "minimax-m3-free": {
6805
- id: "minimax-m3-free",
6806
- name: "MiniMax M3 Free",
6807
- api: "anthropic-messages",
7322
+ "nemotron-3-ultra-free": {
7323
+ id: "nemotron-3-ultra-free",
7324
+ name: "Nemotron 3 Ultra Free",
7325
+ api: "openai-completions",
6808
7326
  provider: "opencode",
6809
- baseUrl: "https://opencode.ai/zen",
7327
+ baseUrl: "https://opencode.ai/zen/v1",
6810
7328
  reasoning: true,
6811
- input: ["text", "image"],
7329
+ input: ["text"],
6812
7330
  cost: {
6813
7331
  input: 0,
6814
7332
  output: 0,
6815
7333
  cacheRead: 0,
6816
7334
  cacheWrite: 0,
6817
7335
  },
6818
- contextWindow: 200000,
6819
- maxTokens: 32000,
7336
+ contextWindow: 1000000,
7337
+ maxTokens: 128000,
6820
7338
  },
6821
- "nemotron-3-super-free": {
6822
- id: "nemotron-3-super-free",
6823
- name: "Nemotron 3 Super Free",
7339
+ "north-mini-code-free": {
7340
+ id: "north-mini-code-free",
7341
+ name: "North Mini Code Free",
6824
7342
  api: "openai-completions",
6825
7343
  provider: "opencode",
6826
7344
  baseUrl: "https://opencode.ai/zen/v1",
@@ -6832,8 +7350,8 @@ export const MODELS = {
6832
7350
  cacheRead: 0,
6833
7351
  cacheWrite: 0,
6834
7352
  },
6835
- contextWindow: 204800,
6836
- maxTokens: 128000,
7353
+ contextWindow: 256000,
7354
+ maxTokens: 64000,
6837
7355
  },
6838
7356
  "qwen3.5-plus": {
6839
7357
  id: "qwen3.5-plus",
@@ -6905,26 +7423,26 @@ export const MODELS = {
6905
7423
  contextWindow: 1000000,
6906
7424
  maxTokens: 384000,
6907
7425
  },
6908
- "glm-5": {
6909
- id: "glm-5",
6910
- name: "GLM-5",
7426
+ "glm-5.1": {
7427
+ id: "glm-5.1",
7428
+ name: "GLM-5.1",
6911
7429
  api: "openai-completions",
6912
7430
  provider: "opencode-go",
6913
7431
  baseUrl: "https://opencode.ai/zen/go/v1",
6914
7432
  reasoning: true,
6915
7433
  input: ["text"],
6916
7434
  cost: {
6917
- input: 1,
6918
- output: 3.2,
6919
- cacheRead: 0.2,
7435
+ input: 1.4,
7436
+ output: 4.4,
7437
+ cacheRead: 0.26,
6920
7438
  cacheWrite: 0,
6921
7439
  },
6922
7440
  contextWindow: 202752,
6923
7441
  maxTokens: 32768,
6924
7442
  },
6925
- "glm-5.1": {
6926
- id: "glm-5.1",
6927
- name: "GLM-5.1",
7443
+ "glm-5.2": {
7444
+ id: "glm-5.2",
7445
+ name: "GLM-5.2",
6928
7446
  api: "openai-completions",
6929
7447
  provider: "opencode-go",
6930
7448
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -6936,29 +7454,29 @@ export const MODELS = {
6936
7454
  cacheRead: 0.26,
6937
7455
  cacheWrite: 0,
6938
7456
  },
6939
- contextWindow: 202752,
6940
- maxTokens: 32768,
7457
+ contextWindow: 1000000,
7458
+ maxTokens: 131072,
6941
7459
  },
6942
- "kimi-k2.5": {
6943
- id: "kimi-k2.5",
6944
- name: "Kimi K2.5",
7460
+ "kimi-k2.6": {
7461
+ id: "kimi-k2.6",
7462
+ name: "Kimi K2.6",
6945
7463
  api: "openai-completions",
6946
7464
  provider: "opencode-go",
6947
7465
  baseUrl: "https://opencode.ai/zen/go/v1",
6948
7466
  reasoning: true,
6949
7467
  input: ["text", "image"],
6950
7468
  cost: {
6951
- input: 0.6,
6952
- output: 3,
6953
- cacheRead: 0.1,
7469
+ input: 0.95,
7470
+ output: 4,
7471
+ cacheRead: 0.16,
6954
7472
  cacheWrite: 0,
6955
7473
  },
6956
7474
  contextWindow: 262144,
6957
7475
  maxTokens: 65536,
6958
7476
  },
6959
- "kimi-k2.6": {
6960
- id: "kimi-k2.6",
6961
- name: "Kimi K2.6",
7477
+ "kimi-k2.7-code": {
7478
+ id: "kimi-k2.7-code",
7479
+ name: "Kimi K2.7 Code",
6962
7480
  api: "openai-completions",
6963
7481
  provider: "opencode-go",
6964
7482
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -6967,11 +7485,11 @@ export const MODELS = {
6967
7485
  cost: {
6968
7486
  input: 0.95,
6969
7487
  output: 4,
6970
- cacheRead: 0.16,
7488
+ cacheRead: 0.19,
6971
7489
  cacheWrite: 0,
6972
7490
  },
6973
7491
  contextWindow: 262144,
6974
- maxTokens: 65536,
7492
+ maxTokens: 262144,
6975
7493
  },
6976
7494
  "mimo-v2.5": {
6977
7495
  id: "mimo-v2.5",
@@ -7007,23 +7525,6 @@ export const MODELS = {
7007
7525
  contextWindow: 1048576,
7008
7526
  maxTokens: 128000,
7009
7527
  },
7010
- "minimax-m2.5": {
7011
- id: "minimax-m2.5",
7012
- name: "MiniMax M2.5",
7013
- api: "anthropic-messages",
7014
- provider: "opencode-go",
7015
- baseUrl: "https://opencode.ai/zen/go",
7016
- reasoning: true,
7017
- input: ["text"],
7018
- cost: {
7019
- input: 0.3,
7020
- output: 1.2,
7021
- cacheRead: 0.03,
7022
- cacheWrite: 0,
7023
- },
7024
- contextWindow: 204800,
7025
- maxTokens: 65536,
7026
- },
7027
7528
  "minimax-m2.7": {
7028
7529
  id: "minimax-m2.7",
7029
7530
  name: "MiniMax M2.7",
@@ -7043,19 +7544,19 @@ export const MODELS = {
7043
7544
  },
7044
7545
  "minimax-m3": {
7045
7546
  id: "minimax-m3",
7046
- name: "MiniMax M3",
7547
+ name: "MiniMax M3 (3x usage)",
7047
7548
  api: "anthropic-messages",
7048
7549
  provider: "opencode-go",
7049
7550
  baseUrl: "https://opencode.ai/zen/go",
7050
7551
  reasoning: true,
7051
7552
  input: ["text", "image"],
7052
7553
  cost: {
7053
- input: 0.6,
7054
- output: 2.4,
7055
- cacheRead: 0.12,
7554
+ input: 0.1,
7555
+ output: 0.4,
7556
+ cacheRead: 0.02,
7056
7557
  cacheWrite: 0,
7057
7558
  },
7058
- contextWindow: 512000,
7559
+ contextWindow: 1000000,
7059
7560
  maxTokens: 131072,
7060
7561
  },
7061
7562
  "qwen3.6-plus": {
@@ -7072,7 +7573,7 @@ export const MODELS = {
7072
7573
  cacheRead: 0.05,
7073
7574
  cacheWrite: 0.625,
7074
7575
  },
7075
- contextWindow: 262144,
7576
+ contextWindow: 1000000,
7076
7577
  maxTokens: 65536,
7077
7578
  },
7078
7579
  "qwen3.7-max": {
@@ -7092,6 +7593,23 @@ export const MODELS = {
7092
7593
  contextWindow: 1000000,
7093
7594
  maxTokens: 65536,
7094
7595
  },
7596
+ "qwen3.7-plus": {
7597
+ id: "qwen3.7-plus",
7598
+ name: "Qwen3.7 Plus",
7599
+ api: "anthropic-messages",
7600
+ provider: "opencode-go",
7601
+ baseUrl: "https://opencode.ai/zen/go",
7602
+ reasoning: true,
7603
+ input: ["text", "image"],
7604
+ cost: {
7605
+ input: 0.4,
7606
+ output: 1.6,
7607
+ cacheRead: 0.04,
7608
+ cacheWrite: 0.5,
7609
+ },
7610
+ contextWindow: 1000000,
7611
+ maxTokens: 65536,
7612
+ },
7095
7613
  },
7096
7614
  "openrouter": {
7097
7615
  "ai21/jamba-large-1.7": {
@@ -7213,22 +7731,22 @@ export const MODELS = {
7213
7731
  contextWindow: 200000,
7214
7732
  maxTokens: 4096,
7215
7733
  },
7216
- "anthropic/claude-3.5-haiku": {
7217
- id: "anthropic/claude-3.5-haiku",
7218
- name: "Anthropic: Claude 3.5 Haiku",
7734
+ "anthropic/claude-fable-5": {
7735
+ id: "anthropic/claude-fable-5",
7736
+ name: "Anthropic: Claude Fable 5",
7219
7737
  api: "openai-completions",
7220
7738
  provider: "openrouter",
7221
7739
  baseUrl: "https://openrouter.ai/api/v1",
7222
- reasoning: false,
7740
+ reasoning: true,
7223
7741
  input: ["text", "image"],
7224
7742
  cost: {
7225
- input: 0.7999999999999999,
7226
- output: 4,
7227
- cacheRead: 0.08,
7228
- cacheWrite: 1,
7743
+ input: 10,
7744
+ output: 50,
7745
+ cacheRead: 1,
7746
+ cacheWrite: 12.5,
7229
7747
  },
7230
- contextWindow: 200000,
7231
- maxTokens: 8192,
7748
+ contextWindow: 1000000,
7749
+ maxTokens: 128000,
7232
7750
  },
7233
7751
  "anthropic/claude-haiku-4.5": {
7234
7752
  id: "anthropic/claude-haiku-4.5",
@@ -7460,13 +7978,13 @@ export const MODELS = {
7460
7978
  reasoning: true,
7461
7979
  input: ["text"],
7462
7980
  cost: {
7463
- input: 0.22,
7464
- output: 0.85,
7981
+ input: 0.25,
7982
+ output: 0.7999999999999999,
7465
7983
  cacheRead: 0.06,
7466
7984
  cacheWrite: 0,
7467
7985
  },
7468
7986
  contextWindow: 262144,
7469
- maxTokens: 262144,
7987
+ maxTokens: 80000,
7470
7988
  },
7471
7989
  "arcee-ai/trinity-mini": {
7472
7990
  id: "arcee-ai/trinity-mini",
@@ -7516,25 +8034,8 @@ export const MODELS = {
7516
8034
  cacheRead: 0,
7517
8035
  cacheWrite: 0,
7518
8036
  },
7519
- contextWindow: 2000000,
7520
- maxTokens: 30000,
7521
- },
7522
- "baidu/ernie-4.5-vl-28b-a3b": {
7523
- id: "baidu/ernie-4.5-vl-28b-a3b",
7524
- name: "Baidu: ERNIE 4.5 VL 28B A3B",
7525
- api: "openai-completions",
7526
- provider: "openrouter",
7527
- baseUrl: "https://openrouter.ai/api/v1",
7528
- reasoning: true,
7529
- input: ["text", "image"],
7530
- cost: {
7531
- input: 0.14,
7532
- output: 0.56,
7533
- cacheRead: 0,
7534
- cacheWrite: 0,
7535
- },
7536
- contextWindow: 131072,
7537
- maxTokens: 8000,
8037
+ contextWindow: 2000000,
8038
+ maxTokens: 30000,
7538
8039
  },
7539
8040
  "bytedance-seed/seed-1.6": {
7540
8041
  id: "bytedance-seed/seed-1.6",
@@ -7638,6 +8139,23 @@ export const MODELS = {
7638
8139
  contextWindow: 128000,
7639
8140
  maxTokens: 4000,
7640
8141
  },
8142
+ "cohere/north-mini-code:free": {
8143
+ id: "cohere/north-mini-code:free",
8144
+ name: "Cohere: North Mini Code (free)",
8145
+ api: "openai-completions",
8146
+ provider: "openrouter",
8147
+ baseUrl: "https://openrouter.ai/api/v1",
8148
+ reasoning: true,
8149
+ input: ["text"],
8150
+ cost: {
8151
+ input: 0,
8152
+ output: 0,
8153
+ cacheRead: 0,
8154
+ cacheWrite: 0,
8155
+ },
8156
+ contextWindow: 256000,
8157
+ maxTokens: 64000,
8158
+ },
7641
8159
  "deepseek/deepseek-chat": {
7642
8160
  id: "deepseek/deepseek-chat",
7643
8161
  name: "DeepSeek: DeepSeek V3",
@@ -7647,8 +8165,8 @@ export const MODELS = {
7647
8165
  reasoning: false,
7648
8166
  input: ["text"],
7649
8167
  cost: {
7650
- input: 0.2288,
7651
- output: 0.9144,
8168
+ input: 0.20020000000000002,
8169
+ output: 0.8000999999999999,
7652
8170
  cacheRead: 0,
7653
8171
  cacheWrite: 0,
7654
8172
  },
@@ -7749,13 +8267,13 @@ export const MODELS = {
7749
8267
  reasoning: true,
7750
8268
  input: ["text"],
7751
8269
  cost: {
7752
- input: 0.252,
7753
- output: 0.378,
7754
- cacheRead: 0.0252,
8270
+ input: 0.2288,
8271
+ output: 0.3432,
8272
+ cacheRead: 0.02288,
7755
8273
  cacheWrite: 0,
7756
8274
  },
7757
8275
  contextWindow: 131072,
7758
- maxTokens: 65536,
8276
+ maxTokens: 64000,
7759
8277
  },
7760
8278
  "deepseek/deepseek-v3.2-exp": {
7761
8279
  id: "deepseek/deepseek-v3.2-exp",
@@ -7783,13 +8301,13 @@ export const MODELS = {
7783
8301
  reasoning: true,
7784
8302
  input: ["text"],
7785
8303
  cost: {
7786
- input: 0.0983,
7787
- output: 0.1966,
7788
- cacheRead: 0.019700000000000002,
8304
+ input: 0.09,
8305
+ output: 0.18,
8306
+ cacheRead: 0.02,
7789
8307
  cacheWrite: 0,
7790
8308
  },
7791
8309
  contextWindow: 1048576,
7792
- maxTokens: 131072,
8310
+ maxTokens: 65536,
7793
8311
  },
7794
8312
  "deepseek/deepseek-v4-pro": {
7795
8313
  id: "deepseek/deepseek-v4-pro",
@@ -7808,57 +8326,6 @@ export const MODELS = {
7808
8326
  contextWindow: 1048576,
7809
8327
  maxTokens: 384000,
7810
8328
  },
7811
- "essentialai/rnj-1-instruct": {
7812
- id: "essentialai/rnj-1-instruct",
7813
- name: "EssentialAI: Rnj 1 Instruct",
7814
- api: "openai-completions",
7815
- provider: "openrouter",
7816
- baseUrl: "https://openrouter.ai/api/v1",
7817
- reasoning: false,
7818
- input: ["text"],
7819
- cost: {
7820
- input: 0.15,
7821
- output: 0.15,
7822
- cacheRead: 0,
7823
- cacheWrite: 0,
7824
- },
7825
- contextWindow: 32768,
7826
- maxTokens: 4096,
7827
- },
7828
- "google/gemini-2.0-flash-001": {
7829
- id: "google/gemini-2.0-flash-001",
7830
- name: "Google: Gemini 2.0 Flash",
7831
- api: "openai-completions",
7832
- provider: "openrouter",
7833
- baseUrl: "https://openrouter.ai/api/v1",
7834
- reasoning: false,
7835
- input: ["text", "image"],
7836
- cost: {
7837
- input: 0.09999999999999999,
7838
- output: 0.39999999999999997,
7839
- cacheRead: 0.024999999999999998,
7840
- cacheWrite: 0.08333333333333334,
7841
- },
7842
- contextWindow: 1048576,
7843
- maxTokens: 8192,
7844
- },
7845
- "google/gemini-2.0-flash-lite-001": {
7846
- id: "google/gemini-2.0-flash-lite-001",
7847
- name: "Google: Gemini 2.0 Flash Lite",
7848
- api: "openai-completions",
7849
- provider: "openrouter",
7850
- baseUrl: "https://openrouter.ai/api/v1",
7851
- reasoning: false,
7852
- input: ["text", "image"],
7853
- cost: {
7854
- input: 0.075,
7855
- output: 0.3,
7856
- cacheRead: 0,
7857
- cacheWrite: 0,
7858
- },
7859
- contextWindow: 1048576,
7860
- maxTokens: 8192,
7861
- },
7862
8329
  "google/gemini-2.5-flash": {
7863
8330
  id: "google/gemini-2.5-flash",
7864
8331
  name: "Google: Gemini 2.5 Flash",
@@ -7976,7 +8443,24 @@ export const MODELS = {
7976
8443
  cacheWrite: 0.08333333333333334,
7977
8444
  },
7978
8445
  contextWindow: 1048576,
7979
- maxTokens: 65536,
8446
+ maxTokens: 65535,
8447
+ },
8448
+ "google/gemini-3-pro-image": {
8449
+ id: "google/gemini-3-pro-image",
8450
+ name: "Google: Nano Banana Pro (Gemini 3 Pro Image)",
8451
+ api: "openai-completions",
8452
+ provider: "openrouter",
8453
+ baseUrl: "https://openrouter.ai/api/v1",
8454
+ reasoning: true,
8455
+ input: ["text", "image"],
8456
+ cost: {
8457
+ input: 2,
8458
+ output: 12,
8459
+ cacheRead: 0.19999999999999998,
8460
+ cacheWrite: 0.375,
8461
+ },
8462
+ contextWindow: 65536,
8463
+ maxTokens: 32768,
7980
8464
  },
7981
8465
  "google/gemini-3.1-flash-lite": {
7982
8466
  id: "google/gemini-3.1-flash-lite",
@@ -8072,8 +8556,8 @@ export const MODELS = {
8072
8556
  reasoning: false,
8073
8557
  input: ["text", "image"],
8074
8558
  cost: {
8075
- input: 0.04,
8076
- output: 0.13,
8559
+ input: 0.049999999999999996,
8560
+ output: 0.15,
8077
8561
  cacheRead: 0,
8078
8562
  cacheWrite: 0,
8079
8563
  },
@@ -8141,12 +8625,12 @@ export const MODELS = {
8141
8625
  input: ["text", "image"],
8142
8626
  cost: {
8143
8627
  input: 0.12,
8144
- output: 0.37,
8145
- cacheRead: 0,
8628
+ output: 0.35,
8629
+ cacheRead: 0.09,
8146
8630
  cacheWrite: 0,
8147
8631
  },
8148
8632
  contextWindow: 262144,
8149
- maxTokens: 16384,
8633
+ maxTokens: 262144,
8150
8634
  },
8151
8635
  "google/gemma-4-31b-it:free": {
8152
8636
  id: "google/gemma-4-31b-it:free",
@@ -8163,7 +8647,7 @@ export const MODELS = {
8163
8647
  cacheWrite: 0,
8164
8648
  },
8165
8649
  contextWindow: 262144,
8166
- maxTokens: 32768,
8650
+ maxTokens: 8192,
8167
8651
  },
8168
8652
  "ibm-granite/granite-4.1-8b": {
8169
8653
  id: "ibm-granite/granite-4.1-8b",
@@ -8242,9 +8726,9 @@ export const MODELS = {
8242
8726
  reasoning: true,
8243
8727
  input: ["text"],
8244
8728
  cost: {
8245
- input: 0.3,
8246
- output: 2.5,
8247
- cacheRead: 0.06,
8729
+ input: 0.075,
8730
+ output: 0.625,
8731
+ cacheRead: 0.015,
8248
8732
  cacheWrite: 0,
8249
8733
  },
8250
8734
  contextWindow: 262144,
@@ -8267,6 +8751,23 @@ export const MODELS = {
8267
8751
  contextWindow: 256000,
8268
8752
  maxTokens: 80000,
8269
8753
  },
8754
+ "liquid/lfm-2.5-1.2b-thinking:free": {
8755
+ id: "liquid/lfm-2.5-1.2b-thinking:free",
8756
+ name: "LiquidAI: LFM2.5-1.2B-Thinking (free)",
8757
+ api: "openai-completions",
8758
+ provider: "openrouter",
8759
+ baseUrl: "https://openrouter.ai/api/v1",
8760
+ reasoning: true,
8761
+ input: ["text"],
8762
+ cost: {
8763
+ input: 0,
8764
+ output: 0,
8765
+ cacheRead: 0,
8766
+ cacheWrite: 0,
8767
+ },
8768
+ contextWindow: 32768,
8769
+ maxTokens: 4096,
8770
+ },
8270
8771
  "meta-llama/llama-3.1-70b-instruct": {
8271
8772
  id: "meta-llama/llama-3.1-70b-instruct",
8272
8773
  name: "Meta: Llama 3.1 70B Instruct",
@@ -8294,7 +8795,7 @@ export const MODELS = {
8294
8795
  input: ["text"],
8295
8796
  cost: {
8296
8797
  input: 0.02,
8297
- output: 0.049999999999999996,
8798
+ output: 0.03,
8298
8799
  cacheRead: 0,
8299
8800
  cacheWrite: 0,
8300
8801
  },
@@ -8361,7 +8862,7 @@ export const MODELS = {
8361
8862
  reasoning: false,
8362
8863
  input: ["text", "image"],
8363
8864
  cost: {
8364
- input: 0.08,
8865
+ input: 0.09999999999999999,
8365
8866
  output: 0.3,
8366
8867
  cacheRead: 0,
8367
8868
  cacheWrite: 0,
@@ -8429,8 +8930,8 @@ export const MODELS = {
8429
8930
  reasoning: true,
8430
8931
  input: ["text"],
8431
8932
  cost: {
8432
- input: 0.15,
8433
- output: 1.15,
8933
+ input: 0.12,
8934
+ output: 0.48,
8434
8935
  cacheRead: 0,
8435
8936
  cacheWrite: 0,
8436
8937
  },
@@ -8446,13 +8947,13 @@ export const MODELS = {
8446
8947
  reasoning: true,
8447
8948
  input: ["text"],
8448
8949
  cost: {
8449
- input: 0.26,
8450
- output: 1.2,
8950
+ input: 0.18,
8951
+ output: 0.72,
8451
8952
  cacheRead: 0,
8452
8953
  cacheWrite: 0,
8453
8954
  },
8454
8955
  contextWindow: 204800,
8455
- maxTokens: 4096,
8956
+ maxTokens: 196608,
8456
8957
  },
8457
8958
  "minimax/minimax-m3": {
8458
8959
  id: "minimax/minimax-m3",
@@ -8805,7 +9306,7 @@ export const MODELS = {
8805
9306
  cost: {
8806
9307
  input: 0.6,
8807
9308
  output: 2.5,
8808
- cacheRead: 0,
9309
+ cacheRead: 0.6,
8809
9310
  cacheWrite: 0,
8810
9311
  },
8811
9312
  contextWindow: 262144,
@@ -8837,47 +9338,30 @@ export const MODELS = {
8837
9338
  reasoning: true,
8838
9339
  input: ["text", "image"],
8839
9340
  cost: {
8840
- input: 0.684,
8841
- output: 3.42,
9341
+ input: 0.66,
9342
+ output: 3.41,
8842
9343
  cacheRead: 0.144,
8843
9344
  cacheWrite: 0,
8844
9345
  },
8845
9346
  contextWindow: 262144,
8846
9347
  maxTokens: 262144,
8847
9348
  },
8848
- "moonshotai/kimi-k2.6:free": {
8849
- id: "moonshotai/kimi-k2.6:free",
8850
- name: "MoonshotAI: Kimi K2.6 (free)",
9349
+ "moonshotai/kimi-k2.7-code": {
9350
+ id: "moonshotai/kimi-k2.7-code",
9351
+ name: "MoonshotAI: Kimi K2.7 Code",
8851
9352
  api: "openai-completions",
8852
9353
  provider: "openrouter",
8853
9354
  baseUrl: "https://openrouter.ai/api/v1",
8854
9355
  reasoning: true,
8855
9356
  input: ["text", "image"],
8856
9357
  cost: {
8857
- input: 0,
8858
- output: 0,
8859
- cacheRead: 0,
9358
+ input: 0.74,
9359
+ output: 3.5,
9360
+ cacheRead: 0.15,
8860
9361
  cacheWrite: 0,
8861
9362
  },
8862
9363
  contextWindow: 262144,
8863
- maxTokens: 4096,
8864
- },
8865
- "nex-agi/deepseek-v3.1-nex-n1": {
8866
- id: "nex-agi/deepseek-v3.1-nex-n1",
8867
- name: "Nex AGI: DeepSeek V3.1 Nex N1",
8868
- api: "openai-completions",
8869
- provider: "openrouter",
8870
- baseUrl: "https://openrouter.ai/api/v1",
8871
- reasoning: false,
8872
- input: ["text"],
8873
- cost: {
8874
- input: 0.135,
8875
- output: 0.5,
8876
- cacheRead: 0,
8877
- cacheWrite: 0,
8878
- },
8879
- contextWindow: 131072,
8880
- maxTokens: 163840,
9364
+ maxTokens: 16384,
8881
9365
  },
8882
9366
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
8883
9367
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -8888,7 +9372,7 @@ export const MODELS = {
8888
9372
  reasoning: true,
8889
9373
  input: ["text"],
8890
9374
  cost: {
8891
- input: 0.09999999999999999,
9375
+ input: 0.39999999999999997,
8892
9376
  output: 0.39999999999999997,
8893
9377
  cacheRead: 0,
8894
9378
  cacheWrite: 0,
@@ -8956,13 +9440,13 @@ export const MODELS = {
8956
9440
  reasoning: true,
8957
9441
  input: ["text"],
8958
9442
  cost: {
8959
- input: 0.09,
8960
- output: 0.44999999999999996,
9443
+ input: 0.08499999999999999,
9444
+ output: 0.39999999999999997,
8961
9445
  cacheRead: 0,
8962
9446
  cacheWrite: 0,
8963
9447
  },
8964
9448
  contextWindow: 1000000,
8965
- maxTokens: 4096,
9449
+ maxTokens: 16384,
8966
9450
  },
8967
9451
  "nvidia/nemotron-3-super-120b-a12b:free": {
8968
9452
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -8981,39 +9465,56 @@ export const MODELS = {
8981
9465
  contextWindow: 1000000,
8982
9466
  maxTokens: 262144,
8983
9467
  },
8984
- "nvidia/nemotron-nano-12b-v2-vl:free": {
8985
- id: "nvidia/nemotron-nano-12b-v2-vl:free",
8986
- name: "NVIDIA: Nemotron Nano 12B 2 VL (free)",
9468
+ "nvidia/nemotron-3-ultra-550b-a55b": {
9469
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
9470
+ name: "NVIDIA: Nemotron 3 Ultra",
8987
9471
  api: "openai-completions",
8988
9472
  provider: "openrouter",
8989
9473
  baseUrl: "https://openrouter.ai/api/v1",
8990
9474
  reasoning: true,
8991
- input: ["text", "image"],
9475
+ input: ["text"],
9476
+ cost: {
9477
+ input: 0.5,
9478
+ output: 2.2,
9479
+ cacheRead: 0.09999999999999999,
9480
+ cacheWrite: 0,
9481
+ },
9482
+ contextWindow: 1000000,
9483
+ maxTokens: 16384,
9484
+ },
9485
+ "nvidia/nemotron-3-ultra-550b-a55b:free": {
9486
+ id: "nvidia/nemotron-3-ultra-550b-a55b:free",
9487
+ name: "NVIDIA: Nemotron 3 Ultra (free)",
9488
+ api: "openai-completions",
9489
+ provider: "openrouter",
9490
+ baseUrl: "https://openrouter.ai/api/v1",
9491
+ reasoning: true,
9492
+ input: ["text"],
8992
9493
  cost: {
8993
9494
  input: 0,
8994
9495
  output: 0,
8995
9496
  cacheRead: 0,
8996
9497
  cacheWrite: 0,
8997
9498
  },
8998
- contextWindow: 128000,
8999
- maxTokens: 128000,
9499
+ contextWindow: 1000000,
9500
+ maxTokens: 65536,
9000
9501
  },
9001
- "nvidia/nemotron-nano-9b-v2": {
9002
- id: "nvidia/nemotron-nano-9b-v2",
9003
- name: "NVIDIA: Nemotron Nano 9B V2",
9502
+ "nvidia/nemotron-nano-12b-v2-vl:free": {
9503
+ id: "nvidia/nemotron-nano-12b-v2-vl:free",
9504
+ name: "NVIDIA: Nemotron Nano 12B 2 VL (free)",
9004
9505
  api: "openai-completions",
9005
9506
  provider: "openrouter",
9006
9507
  baseUrl: "https://openrouter.ai/api/v1",
9007
9508
  reasoning: true,
9008
- input: ["text"],
9509
+ input: ["text", "image"],
9009
9510
  cost: {
9010
- input: 0.04,
9011
- output: 0.16,
9511
+ input: 0,
9512
+ output: 0,
9012
9513
  cacheRead: 0,
9013
9514
  cacheWrite: 0,
9014
9515
  },
9015
- contextWindow: 131072,
9016
- maxTokens: 16384,
9516
+ contextWindow: 128000,
9517
+ maxTokens: 128000,
9017
9518
  },
9018
9519
  "nvidia/nemotron-nano-9b-v2:free": {
9019
9520
  id: "nvidia/nemotron-nano-9b-v2:free",
@@ -9100,40 +9601,6 @@ export const MODELS = {
9100
9601
  contextWindow: 8191,
9101
9602
  maxTokens: 4096,
9102
9603
  },
9103
- "openai/gpt-4-0314": {
9104
- id: "openai/gpt-4-0314",
9105
- name: "OpenAI: GPT-4 (older v0314)",
9106
- api: "openai-completions",
9107
- provider: "openrouter",
9108
- baseUrl: "https://openrouter.ai/api/v1",
9109
- reasoning: false,
9110
- input: ["text"],
9111
- cost: {
9112
- input: 30,
9113
- output: 60,
9114
- cacheRead: 0,
9115
- cacheWrite: 0,
9116
- },
9117
- contextWindow: 8191,
9118
- maxTokens: 4096,
9119
- },
9120
- "openai/gpt-4-1106-preview": {
9121
- id: "openai/gpt-4-1106-preview",
9122
- name: "OpenAI: GPT-4 Turbo (older v1106)",
9123
- api: "openai-completions",
9124
- provider: "openrouter",
9125
- baseUrl: "https://openrouter.ai/api/v1",
9126
- reasoning: false,
9127
- input: ["text"],
9128
- cost: {
9129
- input: 10,
9130
- output: 30,
9131
- cacheRead: 0,
9132
- cacheWrite: 0,
9133
- },
9134
- contextWindow: 128000,
9135
- maxTokens: 4096,
9136
- },
9137
9604
  "openai/gpt-4-turbo": {
9138
9605
  id: "openai/gpt-4-turbo",
9139
9606
  name: "OpenAI: GPT-4 Turbo",
@@ -9755,13 +10222,13 @@ export const MODELS = {
9755
10222
  reasoning: true,
9756
10223
  input: ["text"],
9757
10224
  cost: {
9758
- input: 0.039,
9759
- output: 0.18,
10225
+ input: 0.03,
10226
+ output: 0.15,
9760
10227
  cacheRead: 0,
9761
10228
  cacheWrite: 0,
9762
10229
  },
9763
10230
  contextWindow: 131072,
9764
- maxTokens: 4096,
10231
+ maxTokens: 131072,
9765
10232
  },
9766
10233
  "openai/gpt-oss-120b:free": {
9767
10234
  id: "openai/gpt-oss-120b:free",
@@ -9812,7 +10279,7 @@ export const MODELS = {
9812
10279
  cacheWrite: 0,
9813
10280
  },
9814
10281
  contextWindow: 131072,
9815
- maxTokens: 8192,
10282
+ maxTokens: 32768,
9816
10283
  },
9817
10284
  "openai/gpt-oss-safeguard-20b": {
9818
10285
  id: "openai/gpt-oss-safeguard-20b",
@@ -9825,7 +10292,7 @@ export const MODELS = {
9825
10292
  cost: {
9826
10293
  input: 0.075,
9827
10294
  output: 0.3,
9828
- cacheRead: 0.037,
10295
+ cacheRead: 0.0375,
9829
10296
  cacheWrite: 0,
9830
10297
  },
9831
10298
  contextWindow: 131072,
@@ -10015,16 +10482,50 @@ export const MODELS = {
10015
10482
  cacheRead: 0,
10016
10483
  cacheWrite: 0,
10017
10484
  },
10018
- contextWindow: 200000,
10019
- maxTokens: 4096,
10485
+ contextWindow: 200000,
10486
+ maxTokens: 4096,
10487
+ },
10488
+ "openrouter/owl-alpha": {
10489
+ id: "openrouter/owl-alpha",
10490
+ name: "Owl Alpha",
10491
+ api: "openai-completions",
10492
+ provider: "openrouter",
10493
+ baseUrl: "https://openrouter.ai/api/v1",
10494
+ reasoning: false,
10495
+ input: ["text"],
10496
+ cost: {
10497
+ input: 0,
10498
+ output: 0,
10499
+ cacheRead: 0,
10500
+ cacheWrite: 0,
10501
+ },
10502
+ contextWindow: 1048756,
10503
+ maxTokens: 262144,
10504
+ },
10505
+ "poolside/laguna-m.1": {
10506
+ id: "poolside/laguna-m.1",
10507
+ name: "Poolside: Laguna M.1",
10508
+ api: "openai-completions",
10509
+ provider: "openrouter",
10510
+ baseUrl: "https://openrouter.ai/api/v1",
10511
+ reasoning: true,
10512
+ input: ["text"],
10513
+ cost: {
10514
+ input: 0.19999999999999998,
10515
+ output: 0.39999999999999997,
10516
+ cacheRead: 0.09999999999999999,
10517
+ cacheWrite: 0,
10518
+ },
10519
+ contextWindow: 262144,
10520
+ maxTokens: 32768,
10020
10521
  },
10021
- "openrouter/owl-alpha": {
10022
- id: "openrouter/owl-alpha",
10023
- name: "Owl Alpha",
10522
+ "poolside/laguna-m.1:free": {
10523
+ id: "poolside/laguna-m.1:free",
10524
+ name: "Poolside: Laguna M.1 (free)",
10024
10525
  api: "openai-completions",
10025
10526
  provider: "openrouter",
10026
10527
  baseUrl: "https://openrouter.ai/api/v1",
10027
- reasoning: false,
10528
+ reasoning: true,
10028
10529
  input: ["text"],
10029
10530
  cost: {
10030
10531
  input: 0,
@@ -10032,21 +10533,21 @@ export const MODELS = {
10032
10533
  cacheRead: 0,
10033
10534
  cacheWrite: 0,
10034
10535
  },
10035
- contextWindow: 1048756,
10036
- maxTokens: 262144,
10536
+ contextWindow: 262144,
10537
+ maxTokens: 32768,
10037
10538
  },
10038
- "poolside/laguna-m.1:free": {
10039
- id: "poolside/laguna-m.1:free",
10040
- name: "Poolside: Laguna M.1 (free)",
10539
+ "poolside/laguna-xs.2": {
10540
+ id: "poolside/laguna-xs.2",
10541
+ name: "Poolside: Laguna XS.2",
10041
10542
  api: "openai-completions",
10042
10543
  provider: "openrouter",
10043
10544
  baseUrl: "https://openrouter.ai/api/v1",
10044
10545
  reasoning: true,
10045
10546
  input: ["text"],
10046
10547
  cost: {
10047
- input: 0,
10048
- output: 0,
10049
- cacheRead: 0,
10548
+ input: 0.09999999999999999,
10549
+ output: 0.19999999999999998,
10550
+ cacheRead: 0.049999999999999996,
10050
10551
  cacheWrite: 0,
10051
10552
  },
10052
10553
  contextWindow: 262144,
@@ -10069,39 +10570,39 @@ export const MODELS = {
10069
10570
  contextWindow: 262144,
10070
10571
  maxTokens: 32768,
10071
10572
  },
10072
- "prime-intellect/intellect-3": {
10073
- id: "prime-intellect/intellect-3",
10074
- name: "Prime Intellect: INTELLECT-3",
10573
+ "qwen/qwen-2.5-72b-instruct": {
10574
+ id: "qwen/qwen-2.5-72b-instruct",
10575
+ name: "Qwen2.5 72B Instruct",
10075
10576
  api: "openai-completions",
10076
10577
  provider: "openrouter",
10077
10578
  baseUrl: "https://openrouter.ai/api/v1",
10078
- reasoning: true,
10579
+ reasoning: false,
10079
10580
  input: ["text"],
10080
10581
  cost: {
10081
- input: 0.19999999999999998,
10082
- output: 1.1,
10582
+ input: 0.36,
10583
+ output: 0.39999999999999997,
10083
10584
  cacheRead: 0,
10084
10585
  cacheWrite: 0,
10085
10586
  },
10086
10587
  contextWindow: 131072,
10087
- maxTokens: 131072,
10588
+ maxTokens: 16384,
10088
10589
  },
10089
- "qwen/qwen-2.5-72b-instruct": {
10090
- id: "qwen/qwen-2.5-72b-instruct",
10091
- name: "Qwen2.5 72B Instruct",
10590
+ "qwen/qwen-2.5-7b-instruct": {
10591
+ id: "qwen/qwen-2.5-7b-instruct",
10592
+ name: "Qwen: Qwen2.5 7B Instruct",
10092
10593
  api: "openai-completions",
10093
10594
  provider: "openrouter",
10094
10595
  baseUrl: "https://openrouter.ai/api/v1",
10095
10596
  reasoning: false,
10096
10597
  input: ["text"],
10097
10598
  cost: {
10098
- input: 0.36,
10099
- output: 0.39999999999999997,
10599
+ input: 0.04,
10600
+ output: 0.09999999999999999,
10100
10601
  cacheRead: 0,
10101
10602
  cacheWrite: 0,
10102
10603
  },
10103
10604
  contextWindow: 131072,
10104
- maxTokens: 16384,
10605
+ maxTokens: 32768,
10105
10606
  },
10106
10607
  "qwen/qwen-plus": {
10107
10608
  id: "qwen/qwen-plus",
@@ -10197,7 +10698,7 @@ export const MODELS = {
10197
10698
  reasoning: false,
10198
10699
  input: ["text"],
10199
10700
  cost: {
10200
- input: 0.071,
10701
+ input: 0.09,
10201
10702
  output: 0.09999999999999999,
10202
10703
  cacheRead: 0,
10203
10704
  cacheWrite: 0,
@@ -10231,13 +10732,13 @@ export const MODELS = {
10231
10732
  reasoning: true,
10232
10733
  input: ["text"],
10233
10734
  cost: {
10234
- input: 0.09,
10235
- output: 0.44999999999999996,
10735
+ input: 0.12,
10736
+ output: 0.5,
10236
10737
  cacheRead: 0,
10237
10738
  cacheWrite: 0,
10238
10739
  },
10239
10740
  contextWindow: 131072,
10240
- maxTokens: 20000,
10741
+ maxTokens: 16384,
10241
10742
  },
10242
10743
  "qwen/qwen3-30b-a3b-instruct-2507": {
10243
10744
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -10248,13 +10749,13 @@ export const MODELS = {
10248
10749
  reasoning: false,
10249
10750
  input: ["text"],
10250
10751
  cost: {
10251
- input: 0.09,
10252
- output: 0.3,
10752
+ input: 0.04815,
10753
+ output: 0.19305,
10253
10754
  cacheRead: 0,
10254
10755
  cacheWrite: 0,
10255
10756
  },
10256
- contextWindow: 262144,
10257
- maxTokens: 262144,
10757
+ contextWindow: 131072,
10758
+ maxTokens: 32000,
10258
10759
  },
10259
10760
  "qwen/qwen3-30b-a3b-thinking-2507": {
10260
10761
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -10662,7 +11163,7 @@ export const MODELS = {
10662
11163
  cacheWrite: 0,
10663
11164
  },
10664
11165
  contextWindow: 262144,
10665
- maxTokens: 262144,
11166
+ maxTokens: 81920,
10666
11167
  },
10667
11168
  "qwen/qwen3.5-397b-a17b": {
10668
11169
  id: "qwen/qwen3.5-397b-a17b",
@@ -10673,13 +11174,13 @@ export const MODELS = {
10673
11174
  reasoning: true,
10674
11175
  input: ["text", "image"],
10675
11176
  cost: {
10676
- input: 0.39,
10677
- output: 2.34,
11177
+ input: 0.385,
11178
+ output: 2.4499999999999997,
10678
11179
  cacheRead: 0,
10679
11180
  cacheWrite: 0,
10680
11181
  },
10681
- contextWindow: 262144,
10682
- maxTokens: 65536,
11182
+ contextWindow: 256000,
11183
+ maxTokens: 4096,
10683
11184
  },
10684
11185
  "qwen/qwen3.5-9b": {
10685
11186
  id: "qwen/qwen3.5-9b",
@@ -10690,13 +11191,13 @@ export const MODELS = {
10690
11191
  reasoning: true,
10691
11192
  input: ["text", "image"],
10692
11193
  cost: {
10693
- input: 0.04,
11194
+ input: 0.09999999999999999,
10694
11195
  output: 0.15,
10695
11196
  cacheRead: 0,
10696
11197
  cacheWrite: 0,
10697
11198
  },
10698
11199
  contextWindow: 262144,
10699
- maxTokens: 81920,
11200
+ maxTokens: 262144,
10700
11201
  },
10701
11202
  "qwen/qwen3.5-flash-02-23": {
10702
11203
  id: "qwen/qwen3.5-flash-02-23",
@@ -10758,8 +11259,8 @@ export const MODELS = {
10758
11259
  reasoning: true,
10759
11260
  input: ["text", "image"],
10760
11261
  cost: {
10761
- input: 0.29,
10762
- output: 3.1999999999999997,
11262
+ input: 0.28850000000000003,
11263
+ output: 2.65,
10763
11264
  cacheRead: 0,
10764
11265
  cacheWrite: 0,
10765
11266
  },
@@ -10781,7 +11282,7 @@ export const MODELS = {
10781
11282
  cacheWrite: 0,
10782
11283
  },
10783
11284
  contextWindow: 262144,
10784
- maxTokens: 262140,
11285
+ maxTokens: 262144,
10785
11286
  },
10786
11287
  "qwen/qwen3.6-flash": {
10787
11288
  id: "qwen/qwen3.6-flash",
@@ -10851,6 +11352,23 @@ export const MODELS = {
10851
11352
  contextWindow: 1000000,
10852
11353
  maxTokens: 65536,
10853
11354
  },
11355
+ "qwen/qwen3.7-plus": {
11356
+ id: "qwen/qwen3.7-plus",
11357
+ name: "Qwen: Qwen3.7 Plus",
11358
+ api: "openai-completions",
11359
+ provider: "openrouter",
11360
+ baseUrl: "https://openrouter.ai/api/v1",
11361
+ reasoning: true,
11362
+ input: ["text", "image"],
11363
+ cost: {
11364
+ input: 0.32,
11365
+ output: 1.28,
11366
+ cacheRead: 0.064,
11367
+ cacheWrite: 0.39999999999999997,
11368
+ },
11369
+ contextWindow: 1000000,
11370
+ maxTokens: 65536,
11371
+ },
10854
11372
  "rekaai/reka-edge": {
10855
11373
  id: "rekaai/reka-edge",
10856
11374
  name: "Reka Edge",
@@ -10885,22 +11403,22 @@ export const MODELS = {
10885
11403
  contextWindow: 256000,
10886
11404
  maxTokens: 128000,
10887
11405
  },
10888
- "sao10k/l3-euryale-70b": {
10889
- id: "sao10k/l3-euryale-70b",
10890
- name: "Sao10k: Llama 3 Euryale 70B v2.1",
11406
+ "sakana/fugu-ultra": {
11407
+ id: "sakana/fugu-ultra",
11408
+ name: "Sakana: Fugu Ultra",
10891
11409
  api: "openai-completions",
10892
11410
  provider: "openrouter",
10893
11411
  baseUrl: "https://openrouter.ai/api/v1",
10894
- reasoning: false,
10895
- input: ["text"],
11412
+ reasoning: true,
11413
+ input: ["text", "image"],
10896
11414
  cost: {
10897
- input: 1.48,
10898
- output: 1.48,
10899
- cacheRead: 0,
11415
+ input: 5,
11416
+ output: 30,
11417
+ cacheRead: 0.5,
10900
11418
  cacheWrite: 0,
10901
11419
  },
10902
- contextWindow: 8192,
10903
- maxTokens: 8192,
11420
+ contextWindow: 1000000,
11421
+ maxTokens: 128000,
10904
11422
  },
10905
11423
  "sao10k/l3.1-euryale-70b": {
10906
11424
  id: "sao10k/l3.1-euryale-70b",
@@ -10970,23 +11488,6 @@ export const MODELS = {
10970
11488
  contextWindow: 262144,
10971
11489
  maxTokens: 4096,
10972
11490
  },
10973
- "thedrummer/rocinante-12b": {
10974
- id: "thedrummer/rocinante-12b",
10975
- name: "TheDrummer: Rocinante 12B",
10976
- api: "openai-completions",
10977
- provider: "openrouter",
10978
- baseUrl: "https://openrouter.ai/api/v1",
10979
- reasoning: false,
10980
- input: ["text"],
10981
- cost: {
10982
- input: 0.16999999999999998,
10983
- output: 0.43,
10984
- cacheRead: 0,
10985
- cacheWrite: 0,
10986
- },
10987
- contextWindow: 32768,
10988
- maxTokens: 32768,
10989
- },
10990
11491
  "thedrummer/unslopnemo-12b": {
10991
11492
  id: "thedrummer/unslopnemo-12b",
10992
11493
  name: "TheDrummer: UnslopNemo 12B",
@@ -11072,23 +11573,6 @@ export const MODELS = {
11072
11573
  contextWindow: 256000,
11073
11574
  maxTokens: 4096,
11074
11575
  },
11075
- "xiaomi/mimo-v2-flash": {
11076
- id: "xiaomi/mimo-v2-flash",
11077
- name: "Xiaomi: MiMo-V2-Flash",
11078
- api: "openai-completions",
11079
- provider: "openrouter",
11080
- baseUrl: "https://openrouter.ai/api/v1",
11081
- reasoning: true,
11082
- input: ["text"],
11083
- cost: {
11084
- input: 0.09999999999999999,
11085
- output: 0.3,
11086
- cacheRead: 0.01,
11087
- cacheWrite: 0,
11088
- },
11089
- contextWindow: 262144,
11090
- maxTokens: 65536,
11091
- },
11092
11576
  "xiaomi/mimo-v2.5": {
11093
11577
  id: "xiaomi/mimo-v2.5",
11094
11578
  name: "Xiaomi: MiMo-V2.5",
@@ -11098,13 +11582,13 @@ export const MODELS = {
11098
11582
  reasoning: true,
11099
11583
  input: ["text", "image"],
11100
11584
  cost: {
11101
- input: 0.14,
11585
+ input: 0.105,
11102
11586
  output: 0.28,
11103
- cacheRead: 0.0028,
11587
+ cacheRead: 0,
11104
11588
  cacheWrite: 0,
11105
11589
  },
11106
11590
  contextWindow: 1048576,
11107
- maxTokens: 131072,
11591
+ maxTokens: 4096,
11108
11592
  },
11109
11593
  "xiaomi/mimo-v2.5-pro": {
11110
11594
  id: "xiaomi/mimo-v2.5-pro",
@@ -11123,23 +11607,6 @@ export const MODELS = {
11123
11607
  contextWindow: 1048576,
11124
11608
  maxTokens: 131072,
11125
11609
  },
11126
- "z-ai/glm-4-32b": {
11127
- id: "z-ai/glm-4-32b",
11128
- name: "Z.ai: GLM 4 32B ",
11129
- api: "openai-completions",
11130
- provider: "openrouter",
11131
- baseUrl: "https://openrouter.ai/api/v1",
11132
- reasoning: false,
11133
- input: ["text"],
11134
- cost: {
11135
- input: 0.09999999999999999,
11136
- output: 0.09999999999999999,
11137
- cacheRead: 0,
11138
- cacheWrite: 0,
11139
- },
11140
- contextWindow: 128000,
11141
- maxTokens: 4096,
11142
- },
11143
11610
  "z-ai/glm-4.5": {
11144
11611
  id: "z-ai/glm-4.5",
11145
11612
  name: "Z.ai: GLM 4.5",
@@ -11166,30 +11633,13 @@ export const MODELS = {
11166
11633
  reasoning: true,
11167
11634
  input: ["text"],
11168
11635
  cost: {
11169
- input: 0.125,
11636
+ input: 0.13,
11170
11637
  output: 0.85,
11171
- cacheRead: 0.06,
11172
- cacheWrite: 0,
11173
- },
11174
- contextWindow: 131072,
11175
- maxTokens: 131070,
11176
- },
11177
- "z-ai/glm-4.5-air:free": {
11178
- id: "z-ai/glm-4.5-air:free",
11179
- name: "Z.ai: GLM 4.5 Air (free)",
11180
- api: "openai-completions",
11181
- provider: "openrouter",
11182
- baseUrl: "https://openrouter.ai/api/v1",
11183
- reasoning: true,
11184
- input: ["text"],
11185
- cost: {
11186
- input: 0,
11187
- output: 0,
11188
- cacheRead: 0,
11638
+ cacheRead: 0.024999999999999998,
11189
11639
  cacheWrite: 0,
11190
11640
  },
11191
11641
  contextWindow: 131072,
11192
- maxTokens: 96000,
11642
+ maxTokens: 98304,
11193
11643
  },
11194
11644
  "z-ai/glm-4.5v": {
11195
11645
  id: "z-ai/glm-4.5v",
@@ -11236,11 +11686,11 @@ export const MODELS = {
11236
11686
  cost: {
11237
11687
  input: 0.3,
11238
11688
  output: 0.8999999999999999,
11239
- cacheRead: 0.049999999999999996,
11689
+ cacheRead: 0.055,
11240
11690
  cacheWrite: 0,
11241
11691
  },
11242
11692
  contextWindow: 131072,
11243
- maxTokens: 24000,
11693
+ maxTokens: 32768,
11244
11694
  },
11245
11695
  "z-ai/glm-4.7": {
11246
11696
  id: "z-ai/glm-4.7",
@@ -11307,7 +11757,7 @@ export const MODELS = {
11307
11757
  cacheRead: 0.24,
11308
11758
  cacheWrite: 0,
11309
11759
  },
11310
- contextWindow: 202752,
11760
+ contextWindow: 262144,
11311
11761
  maxTokens: 131072,
11312
11762
  },
11313
11763
  "z-ai/glm-5.1": {
@@ -11327,6 +11777,23 @@ export const MODELS = {
11327
11777
  contextWindow: 202752,
11328
11778
  maxTokens: 4096,
11329
11779
  },
11780
+ "z-ai/glm-5.2": {
11781
+ id: "z-ai/glm-5.2",
11782
+ name: "Z.ai: GLM 5.2",
11783
+ api: "openai-completions",
11784
+ provider: "openrouter",
11785
+ baseUrl: "https://openrouter.ai/api/v1",
11786
+ reasoning: true,
11787
+ input: ["text"],
11788
+ cost: {
11789
+ input: 0.95,
11790
+ output: 3,
11791
+ cacheRead: 0.18,
11792
+ cacheWrite: 0,
11793
+ },
11794
+ contextWindow: 1048576,
11795
+ maxTokens: 32768,
11796
+ },
11330
11797
  "z-ai/glm-5v-turbo": {
11331
11798
  id: "z-ai/glm-5v-turbo",
11332
11799
  name: "Z.ai: GLM 5V Turbo",
@@ -11344,6 +11811,23 @@ export const MODELS = {
11344
11811
  contextWindow: 202752,
11345
11812
  maxTokens: 131072,
11346
11813
  },
11814
+ "~anthropic/claude-fable-latest": {
11815
+ id: "~anthropic/claude-fable-latest",
11816
+ name: "Anthropic: Claude Fable Latest",
11817
+ api: "openai-completions",
11818
+ provider: "openrouter",
11819
+ baseUrl: "https://openrouter.ai/api/v1",
11820
+ reasoning: true,
11821
+ input: ["text", "image"],
11822
+ cost: {
11823
+ input: 10,
11824
+ output: 50,
11825
+ cacheRead: 1,
11826
+ cacheWrite: 12.5,
11827
+ },
11828
+ contextWindow: 1000000,
11829
+ maxTokens: 128000,
11830
+ },
11347
11831
  "~anthropic/claude-haiku-latest": {
11348
11832
  id: "~anthropic/claude-haiku-latest",
11349
11833
  name: "Anthropic Claude Haiku Latest",
@@ -11438,8 +11922,8 @@ export const MODELS = {
11438
11922
  reasoning: true,
11439
11923
  input: ["text", "image"],
11440
11924
  cost: {
11441
- input: 0.684,
11442
- output: 3.42,
11925
+ input: 0.66,
11926
+ output: 3.41,
11443
11927
  cacheRead: 0.144,
11444
11928
  cacheWrite: 0,
11445
11929
  },
@@ -11505,7 +11989,7 @@ export const MODELS = {
11505
11989
  api: "anthropic-messages",
11506
11990
  provider: "vercel-ai-gateway",
11507
11991
  baseUrl: "https://ai-gateway.vercel.sh",
11508
- reasoning: false,
11992
+ reasoning: true,
11509
11993
  input: ["text"],
11510
11994
  cost: {
11511
11995
  input: 0.22,
@@ -11525,8 +12009,8 @@ export const MODELS = {
11525
12009
  reasoning: true,
11526
12010
  input: ["text"],
11527
12011
  cost: {
11528
- input: 0.08,
11529
- output: 0.29,
12012
+ input: 0.12,
12013
+ output: 0.5,
11530
12014
  cacheRead: 0,
11531
12015
  cacheWrite: 0,
11532
12016
  },
@@ -11557,7 +12041,7 @@ export const MODELS = {
11557
12041
  provider: "vercel-ai-gateway",
11558
12042
  baseUrl: "https://ai-gateway.vercel.sh",
11559
12043
  reasoning: true,
11560
- input: ["text", "image"],
12044
+ input: ["text"],
11561
12045
  cost: {
11562
12046
  input: 1.3,
11563
12047
  output: 7.8,
@@ -11607,7 +12091,7 @@ export const MODELS = {
11607
12091
  api: "anthropic-messages",
11608
12092
  provider: "vercel-ai-gateway",
11609
12093
  baseUrl: "https://ai-gateway.vercel.sh",
11610
- reasoning: true,
12094
+ reasoning: false,
11611
12095
  input: ["text"],
11612
12096
  cost: {
11613
12097
  input: 0.15,
@@ -11686,22 +12170,90 @@ export const MODELS = {
11686
12170
  contextWindow: 262144,
11687
12171
  maxTokens: 32768,
11688
12172
  },
11689
- "alibaba/qwen3-max-thinking": {
11690
- id: "alibaba/qwen3-max-thinking",
11691
- name: "Qwen 3 Max Thinking",
12173
+ "alibaba/qwen3-max-thinking": {
12174
+ id: "alibaba/qwen3-max-thinking",
12175
+ name: "Qwen 3 Max Thinking",
12176
+ api: "anthropic-messages",
12177
+ provider: "vercel-ai-gateway",
12178
+ baseUrl: "https://ai-gateway.vercel.sh",
12179
+ reasoning: true,
12180
+ input: ["text"],
12181
+ cost: {
12182
+ input: 1.2,
12183
+ output: 6,
12184
+ cacheRead: 0.24,
12185
+ cacheWrite: 0,
12186
+ },
12187
+ contextWindow: 256000,
12188
+ maxTokens: 65536,
12189
+ },
12190
+ "alibaba/qwen3-next-80b-a3b-instruct": {
12191
+ id: "alibaba/qwen3-next-80b-a3b-instruct",
12192
+ name: "Qwen3 Next 80B A3B Instruct",
12193
+ api: "anthropic-messages",
12194
+ provider: "vercel-ai-gateway",
12195
+ baseUrl: "https://ai-gateway.vercel.sh",
12196
+ reasoning: false,
12197
+ input: ["text"],
12198
+ cost: {
12199
+ input: 0.15,
12200
+ output: 1.2,
12201
+ cacheRead: 0,
12202
+ cacheWrite: 0,
12203
+ },
12204
+ contextWindow: 131072,
12205
+ maxTokens: 32768,
12206
+ },
12207
+ "alibaba/qwen3-next-80b-a3b-thinking": {
12208
+ id: "alibaba/qwen3-next-80b-a3b-thinking",
12209
+ name: "Qwen3 Next 80B A3B Thinking",
12210
+ api: "anthropic-messages",
12211
+ provider: "vercel-ai-gateway",
12212
+ baseUrl: "https://ai-gateway.vercel.sh",
12213
+ reasoning: true,
12214
+ input: ["text"],
12215
+ cost: {
12216
+ input: 0.15,
12217
+ output: 1.2,
12218
+ cacheRead: 0,
12219
+ cacheWrite: 0,
12220
+ },
12221
+ contextWindow: 131072,
12222
+ maxTokens: 32768,
12223
+ },
12224
+ "alibaba/qwen3-vl-235b-a22b-instruct": {
12225
+ id: "alibaba/qwen3-vl-235b-a22b-instruct",
12226
+ name: "Qwen3 VL 235B A22B Instruct",
12227
+ api: "anthropic-messages",
12228
+ provider: "vercel-ai-gateway",
12229
+ baseUrl: "https://ai-gateway.vercel.sh",
12230
+ reasoning: false,
12231
+ input: ["text", "image"],
12232
+ cost: {
12233
+ input: 0.39999999999999997,
12234
+ output: 1.5999999999999999,
12235
+ cacheRead: 0,
12236
+ cacheWrite: 0,
12237
+ },
12238
+ contextWindow: 131072,
12239
+ maxTokens: 129024,
12240
+ },
12241
+ "alibaba/qwen3-vl-instruct": {
12242
+ id: "alibaba/qwen3-vl-instruct",
12243
+ name: "Qwen3 VL 235B A22B Instruct",
11692
12244
  api: "anthropic-messages",
11693
12245
  provider: "vercel-ai-gateway",
11694
12246
  baseUrl: "https://ai-gateway.vercel.sh",
11695
- reasoning: true,
11696
- input: ["text"],
12247
+ reasoning: false,
12248
+ input: ["text", "image"],
11697
12249
  cost: {
11698
- input: 1.2,
11699
- output: 6,
11700
- cacheRead: 0.24,
12250
+ input: 0.39999999999999997,
12251
+ output: 1.5999999999999999,
12252
+ cacheRead: 0,
11701
12253
  cacheWrite: 0,
11702
12254
  },
11703
- contextWindow: 256000,
11704
- maxTokens: 65536,
12255
+ contextWindow: 131072,
12256
+ maxTokens: 129024,
11705
12257
  },
11706
12258
  "alibaba/qwen3-vl-thinking": {
11707
12259
  id: "alibaba/qwen3-vl-thinking",
@@ -11795,7 +12347,7 @@ export const MODELS = {
11795
12347
  provider: "vercel-ai-gateway",
11796
12348
  baseUrl: "https://ai-gateway.vercel.sh",
11797
12349
  reasoning: true,
11798
- input: ["text", "image"],
12350
+ input: ["text"],
11799
12351
  cost: {
11800
12352
  input: 1.25,
11801
12353
  output: 3.75,
@@ -11805,6 +12357,91 @@ export const MODELS = {
11805
12357
  contextWindow: 991000,
11806
12358
  maxTokens: 64000,
11807
12359
  },
12360
+ "alibaba/qwen3.7-plus": {
12361
+ id: "alibaba/qwen3.7-plus",
12362
+ name: "Qwen 3.7 Plus",
12363
+ api: "anthropic-messages",
12364
+ provider: "vercel-ai-gateway",
12365
+ baseUrl: "https://ai-gateway.vercel.sh",
12366
+ reasoning: true,
12367
+ input: ["text", "image"],
12368
+ cost: {
12369
+ input: 0.39999999999999997,
12370
+ output: 1.5999999999999999,
12371
+ cacheRead: 0.08,
12372
+ cacheWrite: 0.5,
12373
+ },
12374
+ contextWindow: 1000000,
12375
+ maxTokens: 64000,
12376
+ },
12377
+ "amazon/nova-2-lite": {
12378
+ id: "amazon/nova-2-lite",
12379
+ name: "Nova 2 Lite",
12380
+ api: "anthropic-messages",
12381
+ provider: "vercel-ai-gateway",
12382
+ baseUrl: "https://ai-gateway.vercel.sh",
12383
+ reasoning: true,
12384
+ input: ["text", "image"],
12385
+ cost: {
12386
+ input: 0.3,
12387
+ output: 2.5,
12388
+ cacheRead: 0.075,
12389
+ cacheWrite: 0,
12390
+ },
12391
+ contextWindow: 1000000,
12392
+ maxTokens: 1000000,
12393
+ },
12394
+ "amazon/nova-lite": {
12395
+ id: "amazon/nova-lite",
12396
+ name: "Nova Lite",
12397
+ api: "anthropic-messages",
12398
+ provider: "vercel-ai-gateway",
12399
+ baseUrl: "https://ai-gateway.vercel.sh",
12400
+ reasoning: false,
12401
+ input: ["text", "image"],
12402
+ cost: {
12403
+ input: 0.06,
12404
+ output: 0.24,
12405
+ cacheRead: 0,
12406
+ cacheWrite: 0,
12407
+ },
12408
+ contextWindow: 300000,
12409
+ maxTokens: 8192,
12410
+ },
12411
+ "amazon/nova-micro": {
12412
+ id: "amazon/nova-micro",
12413
+ name: "Nova Micro",
12414
+ api: "anthropic-messages",
12415
+ provider: "vercel-ai-gateway",
12416
+ baseUrl: "https://ai-gateway.vercel.sh",
12417
+ reasoning: false,
12418
+ input: ["text"],
12419
+ cost: {
12420
+ input: 0.035,
12421
+ output: 0.14,
12422
+ cacheRead: 0,
12423
+ cacheWrite: 0,
12424
+ },
12425
+ contextWindow: 128000,
12426
+ maxTokens: 8192,
12427
+ },
12428
+ "amazon/nova-pro": {
12429
+ id: "amazon/nova-pro",
12430
+ name: "Nova Pro",
12431
+ api: "anthropic-messages",
12432
+ provider: "vercel-ai-gateway",
12433
+ baseUrl: "https://ai-gateway.vercel.sh",
12434
+ reasoning: false,
12435
+ input: ["text", "image"],
12436
+ cost: {
12437
+ input: 0.7999999999999999,
12438
+ output: 3.1999999999999997,
12439
+ cacheRead: 0,
12440
+ cacheWrite: 0,
12441
+ },
12442
+ contextWindow: 300000,
12443
+ maxTokens: 8192,
12444
+ },
11808
12445
  "anthropic/claude-3-haiku": {
11809
12446
  id: "anthropic/claude-3-haiku",
11810
12447
  name: "Claude 3 Haiku",
@@ -12043,6 +12680,23 @@ export const MODELS = {
12043
12680
  contextWindow: 262100,
12044
12681
  maxTokens: 80000,
12045
12682
  },
12683
+ "arcee-ai/trinity-mini": {
12684
+ id: "arcee-ai/trinity-mini",
12685
+ name: "Trinity Mini",
12686
+ api: "anthropic-messages",
12687
+ provider: "vercel-ai-gateway",
12688
+ baseUrl: "https://ai-gateway.vercel.sh",
12689
+ reasoning: false,
12690
+ input: ["text"],
12691
+ cost: {
12692
+ input: 0.045,
12693
+ output: 0.15,
12694
+ cacheRead: 0,
12695
+ cacheWrite: 0,
12696
+ },
12697
+ contextWindow: 131072,
12698
+ maxTokens: 131072,
12699
+ },
12046
12700
  "bytedance/seed-1.6": {
12047
12701
  id: "bytedance/seed-1.6",
12048
12702
  name: "Seed 1.6",
@@ -12050,7 +12704,7 @@ export const MODELS = {
12050
12704
  provider: "vercel-ai-gateway",
12051
12705
  baseUrl: "https://ai-gateway.vercel.sh",
12052
12706
  reasoning: true,
12053
- input: ["text"],
12707
+ input: ["text", "image"],
12054
12708
  cost: {
12055
12709
  input: 0.25,
12056
12710
  output: 2,
@@ -12060,6 +12714,23 @@ export const MODELS = {
12060
12714
  contextWindow: 256000,
12061
12715
  maxTokens: 32000,
12062
12716
  },
12717
+ "bytedance/seed-1.8": {
12718
+ id: "bytedance/seed-1.8",
12719
+ name: "Bytedance Seed 1.8",
12720
+ api: "anthropic-messages",
12721
+ provider: "vercel-ai-gateway",
12722
+ baseUrl: "https://ai-gateway.vercel.sh",
12723
+ reasoning: true,
12724
+ input: ["text", "image"],
12725
+ cost: {
12726
+ input: 0.25,
12727
+ output: 2,
12728
+ cacheRead: 0.049999999999999996,
12729
+ cacheWrite: 0,
12730
+ },
12731
+ contextWindow: 256000,
12732
+ maxTokens: 64000,
12733
+ },
12063
12734
  "cohere/command-a": {
12064
12735
  id: "cohere/command-a",
12065
12736
  name: "Command A",
@@ -12168,7 +12839,7 @@ export const MODELS = {
12168
12839
  api: "anthropic-messages",
12169
12840
  provider: "vercel-ai-gateway",
12170
12841
  baseUrl: "https://ai-gateway.vercel.sh",
12171
- reasoning: false,
12842
+ reasoning: true,
12172
12843
  input: ["text"],
12173
12844
  cost: {
12174
12845
  input: 0.62,
@@ -12213,40 +12884,6 @@ export const MODELS = {
12213
12884
  contextWindow: 1000000,
12214
12885
  maxTokens: 384000,
12215
12886
  },
12216
- "google/gemini-2.0-flash": {
12217
- id: "google/gemini-2.0-flash",
12218
- name: "Gemini 2.0 Flash",
12219
- api: "anthropic-messages",
12220
- provider: "vercel-ai-gateway",
12221
- baseUrl: "https://ai-gateway.vercel.sh",
12222
- reasoning: false,
12223
- input: ["text", "image"],
12224
- cost: {
12225
- input: 0.15,
12226
- output: 0.6,
12227
- cacheRead: 0.024999999999999998,
12228
- cacheWrite: 0,
12229
- },
12230
- contextWindow: 1048576,
12231
- maxTokens: 8192,
12232
- },
12233
- "google/gemini-2.0-flash-lite": {
12234
- id: "google/gemini-2.0-flash-lite",
12235
- name: "Gemini 2.0 Flash Lite",
12236
- api: "anthropic-messages",
12237
- provider: "vercel-ai-gateway",
12238
- baseUrl: "https://ai-gateway.vercel.sh",
12239
- reasoning: false,
12240
- input: ["text", "image"],
12241
- cost: {
12242
- input: 0.075,
12243
- output: 0.3,
12244
- cacheRead: 0.02,
12245
- cacheWrite: 0,
12246
- },
12247
- contextWindow: 1048576,
12248
- maxTokens: 8192,
12249
- },
12250
12887
  "google/gemini-2.5-flash": {
12251
12888
  id: "google/gemini-2.5-flash",
12252
12889
  name: "Gemini 2.5 Flash",
@@ -12406,12 +13043,12 @@ export const MODELS = {
12406
13043
  api: "anthropic-messages",
12407
13044
  provider: "vercel-ai-gateway",
12408
13045
  baseUrl: "https://ai-gateway.vercel.sh",
12409
- reasoning: false,
13046
+ reasoning: true,
12410
13047
  input: ["text", "image"],
12411
13048
  cost: {
12412
- input: 0.13,
12413
- output: 0.39999999999999997,
12414
- cacheRead: 0,
13049
+ input: 0.15,
13050
+ output: 0.6,
13051
+ cacheRead: 0.015,
12415
13052
  cacheWrite: 0,
12416
13053
  },
12417
13054
  contextWindow: 262144,
@@ -12423,7 +13060,7 @@ export const MODELS = {
12423
13060
  api: "anthropic-messages",
12424
13061
  provider: "vercel-ai-gateway",
12425
13062
  baseUrl: "https://ai-gateway.vercel.sh",
12426
- reasoning: false,
13063
+ reasoning: true,
12427
13064
  input: ["text", "image"],
12428
13065
  cost: {
12429
13066
  input: 0.14,
@@ -12468,6 +13105,40 @@ export const MODELS = {
12468
13105
  contextWindow: 32000,
12469
13106
  maxTokens: 16384,
12470
13107
  },
13108
+ "interfaze/interfaze-beta": {
13109
+ id: "interfaze/interfaze-beta",
13110
+ name: "Interfaze Beta",
13111
+ api: "anthropic-messages",
13112
+ provider: "vercel-ai-gateway",
13113
+ baseUrl: "https://ai-gateway.vercel.sh",
13114
+ reasoning: true,
13115
+ input: ["text", "image"],
13116
+ cost: {
13117
+ input: 1.5,
13118
+ output: 3.5,
13119
+ cacheRead: 0,
13120
+ cacheWrite: 0,
13121
+ },
13122
+ contextWindow: 1000000,
13123
+ maxTokens: 32000,
13124
+ },
13125
+ "kwaipilot/kat-coder-pro-v1": {
13126
+ id: "kwaipilot/kat-coder-pro-v1",
13127
+ name: "KAT-Coder-Pro V1",
13128
+ api: "anthropic-messages",
13129
+ provider: "vercel-ai-gateway",
13130
+ baseUrl: "https://ai-gateway.vercel.sh",
13131
+ reasoning: false,
13132
+ input: ["text"],
13133
+ cost: {
13134
+ input: 0.3,
13135
+ output: 1.2,
13136
+ cacheRead: 0.06,
13137
+ cacheWrite: 0,
13138
+ },
13139
+ contextWindow: 256000,
13140
+ maxTokens: 32000,
13141
+ },
12471
13142
  "kwaipilot/kat-coder-pro-v2": {
12472
13143
  id: "kwaipilot/kat-coder-pro-v2",
12473
13144
  name: "Kat Coder Pro V2",
@@ -12502,6 +13173,23 @@ export const MODELS = {
12502
13173
  contextWindow: 128000,
12503
13174
  maxTokens: 100000,
12504
13175
  },
13176
+ "meituan/longcat-flash-thinking-2601": {
13177
+ id: "meituan/longcat-flash-thinking-2601",
13178
+ name: "LongCat Flash Thinking 2601",
13179
+ api: "anthropic-messages",
13180
+ provider: "vercel-ai-gateway",
13181
+ baseUrl: "https://ai-gateway.vercel.sh",
13182
+ reasoning: true,
13183
+ input: ["text"],
13184
+ cost: {
13185
+ input: 0,
13186
+ output: 0,
13187
+ cacheRead: 0,
13188
+ cacheWrite: 0,
13189
+ },
13190
+ contextWindow: 32768,
13191
+ maxTokens: 32768,
13192
+ },
12505
13193
  "meta/llama-3.1-70b": {
12506
13194
  id: "meta/llama-3.1-70b",
12507
13195
  name: "Llama 3.1 70B Instruct",
@@ -12713,7 +13401,7 @@ export const MODELS = {
12713
13401
  provider: "vercel-ai-gateway",
12714
13402
  baseUrl: "https://ai-gateway.vercel.sh",
12715
13403
  reasoning: true,
12716
- input: ["text", "image"],
13404
+ input: ["text"],
12717
13405
  cost: {
12718
13406
  input: 0.3,
12719
13407
  output: 1.2,
@@ -12730,7 +13418,7 @@ export const MODELS = {
12730
13418
  provider: "vercel-ai-gateway",
12731
13419
  baseUrl: "https://ai-gateway.vercel.sh",
12732
13420
  reasoning: true,
12733
- input: ["text", "image"],
13421
+ input: ["text"],
12734
13422
  cost: {
12735
13423
  input: 0.6,
12736
13424
  output: 2.4,
@@ -12808,17 +13496,68 @@ export const MODELS = {
12808
13496
  contextWindow: 128000,
12809
13497
  maxTokens: 64000,
12810
13498
  },
12811
- "mistral/devstral-small-2": {
12812
- id: "mistral/devstral-small-2",
12813
- name: "Devstral Small 2",
13499
+ "mistral/devstral-small-2": {
13500
+ id: "mistral/devstral-small-2",
13501
+ name: "Devstral Small 2",
13502
+ api: "anthropic-messages",
13503
+ provider: "vercel-ai-gateway",
13504
+ baseUrl: "https://ai-gateway.vercel.sh",
13505
+ reasoning: false,
13506
+ input: ["text", "image"],
13507
+ cost: {
13508
+ input: 0.09999999999999999,
13509
+ output: 0.3,
13510
+ cacheRead: 0,
13511
+ cacheWrite: 0,
13512
+ },
13513
+ contextWindow: 256000,
13514
+ maxTokens: 256000,
13515
+ },
13516
+ "mistral/magistral-medium": {
13517
+ id: "mistral/magistral-medium",
13518
+ name: "Magistral Medium 2509",
13519
+ api: "anthropic-messages",
13520
+ provider: "vercel-ai-gateway",
13521
+ baseUrl: "https://ai-gateway.vercel.sh",
13522
+ reasoning: true,
13523
+ input: ["text", "image"],
13524
+ cost: {
13525
+ input: 2,
13526
+ output: 5,
13527
+ cacheRead: 0,
13528
+ cacheWrite: 0,
13529
+ },
13530
+ contextWindow: 128000,
13531
+ maxTokens: 64000,
13532
+ },
13533
+ "mistral/magistral-small": {
13534
+ id: "mistral/magistral-small",
13535
+ name: "Magistral Small 2509",
13536
+ api: "anthropic-messages",
13537
+ provider: "vercel-ai-gateway",
13538
+ baseUrl: "https://ai-gateway.vercel.sh",
13539
+ reasoning: true,
13540
+ input: ["text", "image"],
13541
+ cost: {
13542
+ input: 0.5,
13543
+ output: 1.5,
13544
+ cacheRead: 0,
13545
+ cacheWrite: 0,
13546
+ },
13547
+ contextWindow: 128000,
13548
+ maxTokens: 64000,
13549
+ },
13550
+ "mistral/ministral-14b": {
13551
+ id: "mistral/ministral-14b",
13552
+ name: "Ministral 14B",
12814
13553
  api: "anthropic-messages",
12815
13554
  provider: "vercel-ai-gateway",
12816
13555
  baseUrl: "https://ai-gateway.vercel.sh",
12817
13556
  reasoning: false,
12818
- input: ["text"],
13557
+ input: ["text", "image"],
12819
13558
  cost: {
12820
- input: 0.09999999999999999,
12821
- output: 0.3,
13559
+ input: 0.19999999999999998,
13560
+ output: 0.19999999999999998,
12822
13561
  cacheRead: 0,
12823
13562
  cacheWrite: 0,
12824
13563
  },
@@ -12859,6 +13598,23 @@ export const MODELS = {
12859
13598
  contextWindow: 128000,
12860
13599
  maxTokens: 4000,
12861
13600
  },
13601
+ "mistral/mistral-large-3": {
13602
+ id: "mistral/mistral-large-3",
13603
+ name: "Mistral Large 3",
13604
+ api: "anthropic-messages",
13605
+ provider: "vercel-ai-gateway",
13606
+ baseUrl: "https://ai-gateway.vercel.sh",
13607
+ reasoning: false,
13608
+ input: ["text", "image"],
13609
+ cost: {
13610
+ input: 0.5,
13611
+ output: 1.5,
13612
+ cacheRead: 0,
13613
+ cacheWrite: 0,
13614
+ },
13615
+ contextWindow: 256000,
13616
+ maxTokens: 256000,
13617
+ },
12862
13618
  "mistral/mistral-medium": {
12863
13619
  id: "mistral/mistral-medium",
12864
13620
  name: "Mistral Medium 3.1",
@@ -12883,7 +13639,7 @@ export const MODELS = {
12883
13639
  provider: "vercel-ai-gateway",
12884
13640
  baseUrl: "https://ai-gateway.vercel.sh",
12885
13641
  reasoning: true,
12886
- input: ["text"],
13642
+ input: ["text", "image"],
12887
13643
  cost: {
12888
13644
  input: 1.5,
12889
13645
  output: 7.5,
@@ -12893,6 +13649,23 @@ export const MODELS = {
12893
13649
  contextWindow: 256000,
12894
13650
  maxTokens: 256000,
12895
13651
  },
13652
+ "mistral/mistral-nemo": {
13653
+ id: "mistral/mistral-nemo",
13654
+ name: "Mistral Nemo 12B",
13655
+ api: "anthropic-messages",
13656
+ provider: "vercel-ai-gateway",
13657
+ baseUrl: "https://ai-gateway.vercel.sh",
13658
+ reasoning: false,
13659
+ input: ["text"],
13660
+ cost: {
13661
+ input: 0.15,
13662
+ output: 0.15,
13663
+ cacheRead: 0,
13664
+ cacheWrite: 0,
13665
+ },
13666
+ contextWindow: 128000,
13667
+ maxTokens: 128000,
13668
+ },
12896
13669
  "mistral/mistral-small": {
12897
13670
  id: "mistral/mistral-small",
12898
13671
  name: "Mistral Small",
@@ -12978,73 +13751,124 @@ export const MODELS = {
12978
13751
  contextWindow: 262114,
12979
13752
  maxTokens: 262114,
12980
13753
  },
12981
- "moonshotai/kimi-k2-thinking-turbo": {
12982
- id: "moonshotai/kimi-k2-thinking-turbo",
12983
- name: "Kimi K2 Thinking Turbo",
13754
+ "moonshotai/kimi-k2.5": {
13755
+ id: "moonshotai/kimi-k2.5",
13756
+ name: "Kimi K2.5",
12984
13757
  api: "anthropic-messages",
12985
13758
  provider: "vercel-ai-gateway",
12986
13759
  baseUrl: "https://ai-gateway.vercel.sh",
12987
13760
  reasoning: true,
12988
- input: ["text"],
13761
+ input: ["text", "image"],
12989
13762
  cost: {
12990
- input: 1.15,
12991
- output: 8,
12992
- cacheRead: 0.15,
13763
+ input: 0.6,
13764
+ output: 3,
13765
+ cacheRead: 0.09999999999999999,
12993
13766
  cacheWrite: 0,
12994
13767
  },
12995
13768
  contextWindow: 262114,
12996
13769
  maxTokens: 262114,
12997
13770
  },
12998
- "moonshotai/kimi-k2-turbo": {
12999
- id: "moonshotai/kimi-k2-turbo",
13000
- name: "Kimi K2 Turbo",
13771
+ "moonshotai/kimi-k2.6": {
13772
+ id: "moonshotai/kimi-k2.6",
13773
+ name: "Kimi K2.6",
13001
13774
  api: "anthropic-messages",
13002
13775
  provider: "vercel-ai-gateway",
13003
13776
  baseUrl: "https://ai-gateway.vercel.sh",
13004
- reasoning: false,
13005
- input: ["text"],
13777
+ reasoning: true,
13778
+ input: ["text", "image"],
13006
13779
  cost: {
13007
- input: 1.15,
13008
- output: 8,
13009
- cacheRead: 0.15,
13780
+ input: 0.95,
13781
+ output: 4,
13782
+ cacheRead: 0.16,
13010
13783
  cacheWrite: 0,
13011
13784
  },
13012
- contextWindow: 256000,
13013
- maxTokens: 16384,
13785
+ contextWindow: 262000,
13786
+ maxTokens: 262000,
13014
13787
  },
13015
- "moonshotai/kimi-k2.5": {
13016
- id: "moonshotai/kimi-k2.5",
13017
- name: "Kimi K2.5",
13788
+ "moonshotai/kimi-k2.7-code": {
13789
+ id: "moonshotai/kimi-k2.7-code",
13790
+ name: "Kimi K2.7 Code",
13018
13791
  api: "anthropic-messages",
13019
13792
  provider: "vercel-ai-gateway",
13020
13793
  baseUrl: "https://ai-gateway.vercel.sh",
13021
13794
  reasoning: true,
13022
13795
  input: ["text", "image"],
13023
13796
  cost: {
13024
- input: 0.6,
13025
- output: 3,
13026
- cacheRead: 0.09999999999999999,
13797
+ input: 0.95,
13798
+ output: 4,
13799
+ cacheRead: 0.19,
13027
13800
  cacheWrite: 0,
13028
13801
  },
13029
- contextWindow: 262114,
13030
- maxTokens: 262114,
13802
+ contextWindow: 256000,
13803
+ maxTokens: 32768,
13031
13804
  },
13032
- "moonshotai/kimi-k2.6": {
13033
- id: "moonshotai/kimi-k2.6",
13034
- name: "Kimi K2.6",
13805
+ "moonshotai/kimi-k2.7-code-highspeed": {
13806
+ id: "moonshotai/kimi-k2.7-code-highspeed",
13807
+ name: "Kimi K2.7 Code High Speed",
13035
13808
  api: "anthropic-messages",
13036
13809
  provider: "vercel-ai-gateway",
13037
13810
  baseUrl: "https://ai-gateway.vercel.sh",
13038
13811
  reasoning: true,
13039
13812
  input: ["text", "image"],
13040
13813
  cost: {
13041
- input: 0.95,
13042
- output: 4,
13043
- cacheRead: 0.16,
13814
+ input: 1.9,
13815
+ output: 8,
13816
+ cacheRead: 0.38,
13044
13817
  cacheWrite: 0,
13045
13818
  },
13046
- contextWindow: 262000,
13047
- maxTokens: 262000,
13819
+ contextWindow: 262144,
13820
+ maxTokens: 32768,
13821
+ },
13822
+ "nvidia/nemotron-3-nano-30b-a3b": {
13823
+ id: "nvidia/nemotron-3-nano-30b-a3b",
13824
+ name: "Nemotron 3 Nano 30B A3B",
13825
+ api: "anthropic-messages",
13826
+ provider: "vercel-ai-gateway",
13827
+ baseUrl: "https://ai-gateway.vercel.sh",
13828
+ reasoning: true,
13829
+ input: ["text"],
13830
+ cost: {
13831
+ input: 0.049999999999999996,
13832
+ output: 0.24,
13833
+ cacheRead: 0,
13834
+ cacheWrite: 0,
13835
+ },
13836
+ contextWindow: 262144,
13837
+ maxTokens: 262144,
13838
+ },
13839
+ "nvidia/nemotron-3-super-120b-a12b": {
13840
+ id: "nvidia/nemotron-3-super-120b-a12b",
13841
+ name: "NVIDIA Nemotron 3 Super 120B A12B",
13842
+ api: "anthropic-messages",
13843
+ provider: "vercel-ai-gateway",
13844
+ baseUrl: "https://ai-gateway.vercel.sh",
13845
+ reasoning: true,
13846
+ input: ["text"],
13847
+ cost: {
13848
+ input: 0.15,
13849
+ output: 0.65,
13850
+ cacheRead: 0,
13851
+ cacheWrite: 0,
13852
+ },
13853
+ contextWindow: 256000,
13854
+ maxTokens: 32000,
13855
+ },
13856
+ "nvidia/nemotron-3-ultra-550b-a55b": {
13857
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
13858
+ name: "Nemotron 3 Ultra",
13859
+ api: "anthropic-messages",
13860
+ provider: "vercel-ai-gateway",
13861
+ baseUrl: "https://ai-gateway.vercel.sh",
13862
+ reasoning: true,
13863
+ input: ["text"],
13864
+ cost: {
13865
+ input: 0.6,
13866
+ output: 2.4,
13867
+ cacheRead: 0.12,
13868
+ cacheWrite: 0,
13869
+ },
13870
+ contextWindow: 1000000,
13871
+ maxTokens: 65000,
13048
13872
  },
13049
13873
  "nvidia/nemotron-nano-12b-v2-vl": {
13050
13874
  id: "nvidia/nemotron-nano-12b-v2-vl",
@@ -13080,6 +13904,23 @@ export const MODELS = {
13080
13904
  contextWindow: 131072,
13081
13905
  maxTokens: 131072,
13082
13906
  },
13907
+ "openai/gpt-3.5-turbo": {
13908
+ id: "openai/gpt-3.5-turbo",
13909
+ name: "GPT-3.5 Turbo",
13910
+ api: "anthropic-messages",
13911
+ provider: "vercel-ai-gateway",
13912
+ baseUrl: "https://ai-gateway.vercel.sh",
13913
+ reasoning: false,
13914
+ input: ["text"],
13915
+ cost: {
13916
+ input: 0.5,
13917
+ output: 1.5,
13918
+ cacheRead: 0,
13919
+ cacheWrite: 0,
13920
+ },
13921
+ contextWindow: 16385,
13922
+ maxTokens: 4096,
13923
+ },
13083
13924
  "openai/gpt-4-turbo": {
13084
13925
  id: "openai/gpt-4-turbo",
13085
13926
  name: "GPT-4 Turbo",
@@ -13205,7 +14046,7 @@ export const MODELS = {
13205
14046
  api: "anthropic-messages",
13206
14047
  provider: "vercel-ai-gateway",
13207
14048
  baseUrl: "https://ai-gateway.vercel.sh",
13208
- reasoning: true,
14049
+ reasoning: false,
13209
14050
  input: ["text", "image"],
13210
14051
  cost: {
13211
14052
  input: 1.25,
@@ -13223,7 +14064,7 @@ export const MODELS = {
13223
14064
  provider: "vercel-ai-gateway",
13224
14065
  baseUrl: "https://ai-gateway.vercel.sh",
13225
14066
  reasoning: true,
13226
- input: ["text"],
14067
+ input: ["text", "image"],
13227
14068
  cost: {
13228
14069
  input: 1.25,
13229
14070
  output: 10,
@@ -13341,7 +14182,7 @@ export const MODELS = {
13341
14182
  api: "anthropic-messages",
13342
14183
  provider: "vercel-ai-gateway",
13343
14184
  baseUrl: "https://ai-gateway.vercel.sh",
13344
- reasoning: true,
14185
+ reasoning: false,
13345
14186
  input: ["text", "image"],
13346
14187
  cost: {
13347
14188
  input: 1.25,
@@ -13392,7 +14233,7 @@ export const MODELS = {
13392
14233
  api: "anthropic-messages",
13393
14234
  provider: "vercel-ai-gateway",
13394
14235
  baseUrl: "https://ai-gateway.vercel.sh",
13395
- reasoning: true,
14236
+ reasoning: false,
13396
14237
  input: ["text", "image"],
13397
14238
  cost: {
13398
14239
  input: 1.75,
@@ -13443,7 +14284,7 @@ export const MODELS = {
13443
14284
  api: "anthropic-messages",
13444
14285
  provider: "vercel-ai-gateway",
13445
14286
  baseUrl: "https://ai-gateway.vercel.sh",
13446
- reasoning: true,
14287
+ reasoning: false,
13447
14288
  input: ["text", "image"],
13448
14289
  cost: {
13449
14290
  input: 1.75,
@@ -13573,6 +14414,23 @@ export const MODELS = {
13573
14414
  contextWindow: 1000000,
13574
14415
  maxTokens: 128000,
13575
14416
  },
14417
+ "openai/gpt-oss-120b": {
14418
+ id: "openai/gpt-oss-120b",
14419
+ name: "GPT OSS 120B",
14420
+ api: "anthropic-messages",
14421
+ provider: "vercel-ai-gateway",
14422
+ baseUrl: "https://ai-gateway.vercel.sh",
14423
+ reasoning: true,
14424
+ input: ["text"],
14425
+ cost: {
14426
+ input: 0.35,
14427
+ output: 0.75,
14428
+ cacheRead: 0.25,
14429
+ cacheWrite: 0,
14430
+ },
14431
+ contextWindow: 131072,
14432
+ maxTokens: 131000,
14433
+ },
13576
14434
  "openai/gpt-oss-20b": {
13577
14435
  id: "openai/gpt-oss-20b",
13578
14436
  name: "GPT OSS 20B",
@@ -13709,39 +14567,22 @@ export const MODELS = {
13709
14567
  contextWindow: 200000,
13710
14568
  maxTokens: 100000,
13711
14569
  },
13712
- "perplexity/sonar": {
13713
- id: "perplexity/sonar",
13714
- name: "Sonar",
13715
- api: "anthropic-messages",
13716
- provider: "vercel-ai-gateway",
13717
- baseUrl: "https://ai-gateway.vercel.sh",
13718
- reasoning: false,
13719
- input: ["text", "image"],
13720
- cost: {
13721
- input: 0,
13722
- output: 0,
13723
- cacheRead: 0,
13724
- cacheWrite: 0,
13725
- },
13726
- contextWindow: 127000,
13727
- maxTokens: 8000,
13728
- },
13729
- "perplexity/sonar-pro": {
13730
- id: "perplexity/sonar-pro",
13731
- name: "Sonar Pro",
14570
+ "sakana/fugu-ultra": {
14571
+ id: "sakana/fugu-ultra",
14572
+ name: "Fugu Ultra",
13732
14573
  api: "anthropic-messages",
13733
14574
  provider: "vercel-ai-gateway",
13734
14575
  baseUrl: "https://ai-gateway.vercel.sh",
13735
- reasoning: false,
14576
+ reasoning: true,
13736
14577
  input: ["text", "image"],
13737
14578
  cost: {
13738
- input: 0,
13739
- output: 0,
13740
- cacheRead: 0,
14579
+ input: 5,
14580
+ output: 30,
14581
+ cacheRead: 0.5,
13741
14582
  cacheWrite: 0,
13742
14583
  },
13743
- contextWindow: 200000,
13744
- maxTokens: 8000,
14584
+ contextWindow: 1000000,
14585
+ maxTokens: 1000000,
13745
14586
  },
13746
14587
  "stepfun/step-3.5-flash": {
13747
14588
  id: "stepfun/step-3.5-flash",
@@ -13754,8 +14595,8 @@ export const MODELS = {
13754
14595
  cost: {
13755
14596
  input: 0.09,
13756
14597
  output: 0.3,
13757
- cacheRead: 0,
13758
- cacheWrite: 0.02,
14598
+ cacheRead: 0.02,
14599
+ cacheWrite: 0,
13759
14600
  },
13760
14601
  contextWindow: 262114,
13761
14602
  maxTokens: 262114,
@@ -14005,7 +14846,7 @@ export const MODELS = {
14005
14846
  provider: "vercel-ai-gateway",
14006
14847
  baseUrl: "https://ai-gateway.vercel.sh",
14007
14848
  reasoning: true,
14008
- input: ["text", "image"],
14849
+ input: ["text"],
14009
14850
  cost: {
14010
14851
  input: 0.435,
14011
14852
  output: 0.87,
@@ -14055,7 +14896,7 @@ export const MODELS = {
14055
14896
  api: "anthropic-messages",
14056
14897
  provider: "vercel-ai-gateway",
14057
14898
  baseUrl: "https://ai-gateway.vercel.sh",
14058
- reasoning: false,
14899
+ reasoning: true,
14059
14900
  input: ["text", "image"],
14060
14901
  cost: {
14061
14902
  input: 0.6,
@@ -14219,6 +15060,40 @@ export const MODELS = {
14219
15060
  contextWindow: 202800,
14220
15061
  maxTokens: 64000,
14221
15062
  },
15063
+ "zai/glm-5.2": {
15064
+ id: "zai/glm-5.2",
15065
+ name: "GLM 5.2",
15066
+ api: "anthropic-messages",
15067
+ provider: "vercel-ai-gateway",
15068
+ baseUrl: "https://ai-gateway.vercel.sh",
15069
+ reasoning: true,
15070
+ input: ["text"],
15071
+ cost: {
15072
+ input: 1.5,
15073
+ output: 4.5,
15074
+ cacheRead: 0.3,
15075
+ cacheWrite: 0,
15076
+ },
15077
+ contextWindow: 1000000,
15078
+ maxTokens: 128000,
15079
+ },
15080
+ "zai/glm-5.2-fast": {
15081
+ id: "zai/glm-5.2-fast",
15082
+ name: "GLM 5.2 Fast",
15083
+ api: "anthropic-messages",
15084
+ provider: "vercel-ai-gateway",
15085
+ baseUrl: "https://ai-gateway.vercel.sh",
15086
+ reasoning: true,
15087
+ input: ["text"],
15088
+ cost: {
15089
+ input: 3,
15090
+ output: 10.25,
15091
+ cacheRead: 0.5,
15092
+ cacheWrite: 0,
15093
+ },
15094
+ contextWindow: 1000000,
15095
+ maxTokens: 128000,
15096
+ },
14222
15097
  "zai/glm-5v-turbo": {
14223
15098
  id: "zai/glm-5v-turbo",
14224
15099
  name: "GLM 5V Turbo",
@@ -14252,7 +15127,7 @@ export const MODELS = {
14252
15127
  cacheRead: 0.2,
14253
15128
  cacheWrite: 0,
14254
15129
  },
14255
- contextWindow: 2000000,
15130
+ contextWindow: 1000000,
14256
15131
  maxTokens: 30000,
14257
15132
  },
14258
15133
  "grok-4.20-0309-reasoning": {
@@ -14269,7 +15144,7 @@ export const MODELS = {
14269
15144
  cacheRead: 0.2,
14270
15145
  cacheWrite: 0,
14271
15146
  },
14272
- contextWindow: 2000000,
15147
+ contextWindow: 1000000,
14273
15148
  maxTokens: 30000,
14274
15149
  },
14275
15150
  "grok-4.3": {
@@ -14541,6 +15416,24 @@ export const MODELS = {
14541
15416
  contextWindow: 200000,
14542
15417
  maxTokens: 131072,
14543
15418
  },
15419
+ "glm-5.2": {
15420
+ id: "glm-5.2",
15421
+ name: "GLM-5.2",
15422
+ api: "openai-completions",
15423
+ provider: "zai",
15424
+ baseUrl: "https://api.z.ai/api/paas/v4",
15425
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
15426
+ reasoning: true,
15427
+ input: ["text"],
15428
+ cost: {
15429
+ input: 1.4,
15430
+ output: 4.4,
15431
+ cacheRead: 0.26,
15432
+ cacheWrite: 0,
15433
+ },
15434
+ contextWindow: 1000000,
15435
+ maxTokens: 131072,
15436
+ },
14544
15437
  "glm-5v-turbo": {
14545
15438
  id: "glm-5v-turbo",
14546
15439
  name: "GLM-5V-Turbo",