@dreb/ai 2.21.1 → 2.22.0

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.
Files changed (32) hide show
  1. package/dist/models.generated.d.ts +285 -227
  2. package/dist/models.generated.d.ts.map +1 -1
  3. package/dist/models.generated.js +483 -431
  4. package/dist/models.generated.js.map +1 -1
  5. package/dist/providers/amazon-bedrock.d.ts.map +1 -1
  6. package/dist/providers/amazon-bedrock.js +5 -0
  7. package/dist/providers/amazon-bedrock.js.map +1 -1
  8. package/dist/providers/anthropic.d.ts.map +1 -1
  9. package/dist/providers/anthropic.js +5 -0
  10. package/dist/providers/anthropic.js.map +1 -1
  11. package/dist/providers/google-gemini-cli.d.ts.map +1 -1
  12. package/dist/providers/google-gemini-cli.js +5 -0
  13. package/dist/providers/google-gemini-cli.js.map +1 -1
  14. package/dist/providers/google-vertex.d.ts.map +1 -1
  15. package/dist/providers/google-vertex.js +5 -0
  16. package/dist/providers/google-vertex.js.map +1 -1
  17. package/dist/providers/google.d.ts.map +1 -1
  18. package/dist/providers/google.js +5 -0
  19. package/dist/providers/google.js.map +1 -1
  20. package/dist/providers/mistral.d.ts.map +1 -1
  21. package/dist/providers/mistral.js +5 -0
  22. package/dist/providers/mistral.js.map +1 -1
  23. package/dist/providers/openai-codex-responses.d.ts.map +1 -1
  24. package/dist/providers/openai-codex-responses.js +2 -1
  25. package/dist/providers/openai-codex-responses.js.map +1 -1
  26. package/dist/providers/openai-completions.d.ts.map +1 -1
  27. package/dist/providers/openai-completions.js +5 -0
  28. package/dist/providers/openai-completions.js.map +1 -1
  29. package/dist/providers/openai-responses-shared.d.ts.map +1 -1
  30. package/dist/providers/openai-responses-shared.js +5 -0
  31. package/dist/providers/openai-responses-shared.js.map +1 -1
  32. package/package.json +2 -2
@@ -385,8 +385,8 @@ export const MODELS = {
385
385
  reasoning: true,
386
386
  input: ["text", "image"],
387
387
  cost: {
388
- input: 5,
389
- output: 25,
388
+ input: 5.5,
389
+ output: 27.5,
390
390
  cacheRead: 0.5,
391
391
  cacheWrite: 6.25,
392
392
  },
@@ -402,10 +402,10 @@ export const MODELS = {
402
402
  reasoning: true,
403
403
  input: ["text", "image"],
404
404
  cost: {
405
- input: 5,
406
- output: 25,
407
- cacheRead: 0.5,
408
- cacheWrite: 6.25,
405
+ input: 5.5,
406
+ output: 27.5,
407
+ cacheRead: 0.55,
408
+ cacheWrite: 6.875,
409
409
  },
410
410
  contextWindow: 1000000,
411
411
  maxTokens: 128000,
@@ -419,10 +419,10 @@ export const MODELS = {
419
419
  reasoning: true,
420
420
  input: ["text", "image"],
421
421
  cost: {
422
- input: 5,
423
- output: 25,
424
- cacheRead: 0.5,
425
- cacheWrite: 6.25,
422
+ input: 5.5,
423
+ output: 27.5,
424
+ cacheRead: 0.55,
425
+ cacheWrite: 6.875,
426
426
  },
427
427
  contextWindow: 1000000,
428
428
  maxTokens: 128000,
@@ -436,10 +436,10 @@ export const MODELS = {
436
436
  reasoning: true,
437
437
  input: ["text", "image"],
438
438
  cost: {
439
- input: 3,
440
- output: 15,
441
- cacheRead: 0.3,
442
- cacheWrite: 3.75,
439
+ input: 3.3,
440
+ output: 16.5,
441
+ cacheRead: 0.33,
442
+ cacheWrite: 4.125,
443
443
  },
444
444
  contextWindow: 200000,
445
445
  maxTokens: 64000,
@@ -453,10 +453,10 @@ export const MODELS = {
453
453
  reasoning: true,
454
454
  input: ["text", "image"],
455
455
  cost: {
456
- input: 3,
457
- output: 15,
458
- cacheRead: 0.3,
459
- cacheWrite: 3.75,
456
+ input: 3.3,
457
+ output: 16.5,
458
+ cacheRead: 0.33,
459
+ cacheWrite: 4.125,
460
460
  },
461
461
  contextWindow: 1000000,
462
462
  maxTokens: 64000,
@@ -2698,7 +2698,7 @@ export const MODELS = {
2698
2698
  "github-copilot": {
2699
2699
  "claude-haiku-4.5": {
2700
2700
  id: "claude-haiku-4.5",
2701
- name: "Claude Haiku 4.5",
2701
+ name: "Claude Haiku 4.5 (latest)",
2702
2702
  api: "anthropic-messages",
2703
2703
  provider: "github-copilot",
2704
2704
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2706,17 +2706,17 @@ export const MODELS = {
2706
2706
  reasoning: true,
2707
2707
  input: ["text", "image"],
2708
2708
  cost: {
2709
- input: 0,
2710
- output: 0,
2711
- cacheRead: 0,
2712
- cacheWrite: 0,
2709
+ input: 1,
2710
+ output: 5,
2711
+ cacheRead: 0.1,
2712
+ cacheWrite: 1.25,
2713
2713
  },
2714
- contextWindow: 144000,
2715
- maxTokens: 32000,
2714
+ contextWindow: 200000,
2715
+ maxTokens: 64000,
2716
2716
  },
2717
2717
  "claude-opus-4.5": {
2718
2718
  id: "claude-opus-4.5",
2719
- name: "Claude Opus 4.5",
2719
+ name: "Claude Opus 4.5 (latest)",
2720
2720
  api: "anthropic-messages",
2721
2721
  provider: "github-copilot",
2722
2722
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2724,13 +2724,13 @@ export const MODELS = {
2724
2724
  reasoning: true,
2725
2725
  input: ["text", "image"],
2726
2726
  cost: {
2727
- input: 0,
2728
- output: 0,
2729
- cacheRead: 0,
2730
- cacheWrite: 0,
2727
+ input: 5,
2728
+ output: 25,
2729
+ cacheRead: 0.5,
2730
+ cacheWrite: 6.25,
2731
2731
  },
2732
- contextWindow: 160000,
2733
- maxTokens: 32000,
2732
+ contextWindow: 200000,
2733
+ maxTokens: 64000,
2734
2734
  },
2735
2735
  "claude-opus-4.6": {
2736
2736
  id: "claude-opus-4.6",
@@ -2742,13 +2742,13 @@ export const MODELS = {
2742
2742
  reasoning: true,
2743
2743
  input: ["text", "image"],
2744
2744
  cost: {
2745
- input: 0,
2746
- output: 0,
2747
- cacheRead: 0,
2748
- cacheWrite: 0,
2745
+ input: 5,
2746
+ output: 25,
2747
+ cacheRead: 0.5,
2748
+ cacheWrite: 6.25,
2749
2749
  },
2750
2750
  contextWindow: 1000000,
2751
- maxTokens: 64000,
2751
+ maxTokens: 128000,
2752
2752
  },
2753
2753
  "claude-opus-4.7": {
2754
2754
  id: "claude-opus-4.7",
@@ -2760,17 +2760,53 @@ export const MODELS = {
2760
2760
  reasoning: true,
2761
2761
  input: ["text", "image"],
2762
2762
  cost: {
2763
- input: 0,
2764
- output: 0,
2765
- cacheRead: 0,
2766
- cacheWrite: 0,
2763
+ input: 5,
2764
+ output: 25,
2765
+ cacheRead: 0.5,
2766
+ cacheWrite: 6.25,
2767
+ },
2768
+ contextWindow: 1000000,
2769
+ maxTokens: 128000,
2770
+ },
2771
+ "claude-opus-4.8": {
2772
+ id: "claude-opus-4.8",
2773
+ name: "Claude Opus 4.8",
2774
+ api: "anthropic-messages",
2775
+ provider: "github-copilot",
2776
+ baseUrl: "https://api.individual.githubcopilot.com",
2777
+ 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" },
2778
+ reasoning: true,
2779
+ input: ["text", "image"],
2780
+ cost: {
2781
+ input: 5,
2782
+ output: 25,
2783
+ cacheRead: 0.5,
2784
+ cacheWrite: 6.25,
2767
2785
  },
2768
- contextWindow: 144000,
2786
+ contextWindow: 1000000,
2787
+ maxTokens: 128000,
2788
+ },
2789
+ "claude-sonnet-4": {
2790
+ id: "claude-sonnet-4",
2791
+ name: "Claude Sonnet 4 (latest)",
2792
+ api: "anthropic-messages",
2793
+ provider: "github-copilot",
2794
+ baseUrl: "https://api.individual.githubcopilot.com",
2795
+ 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" },
2796
+ reasoning: true,
2797
+ input: ["text", "image"],
2798
+ cost: {
2799
+ input: 3,
2800
+ output: 15,
2801
+ cacheRead: 0.3,
2802
+ cacheWrite: 3.75,
2803
+ },
2804
+ contextWindow: 200000,
2769
2805
  maxTokens: 64000,
2770
2806
  },
2771
2807
  "claude-sonnet-4.5": {
2772
2808
  id: "claude-sonnet-4.5",
2773
- name: "Claude Sonnet 4.5",
2809
+ name: "Claude Sonnet 4.5 (latest)",
2774
2810
  api: "anthropic-messages",
2775
2811
  provider: "github-copilot",
2776
2812
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2778,13 +2814,13 @@ export const MODELS = {
2778
2814
  reasoning: true,
2779
2815
  input: ["text", "image"],
2780
2816
  cost: {
2781
- input: 0,
2782
- output: 0,
2783
- cacheRead: 0,
2784
- cacheWrite: 0,
2817
+ input: 3,
2818
+ output: 15,
2819
+ cacheRead: 0.3,
2820
+ cacheWrite: 3.75,
2785
2821
  },
2786
- contextWindow: 144000,
2787
- maxTokens: 32000,
2822
+ contextWindow: 200000,
2823
+ maxTokens: 64000,
2788
2824
  },
2789
2825
  "claude-sonnet-4.6": {
2790
2826
  id: "claude-sonnet-4.6",
@@ -2796,13 +2832,13 @@ export const MODELS = {
2796
2832
  reasoning: true,
2797
2833
  input: ["text", "image"],
2798
2834
  cost: {
2799
- input: 0,
2800
- output: 0,
2801
- cacheRead: 0,
2802
- cacheWrite: 0,
2835
+ input: 3,
2836
+ output: 15,
2837
+ cacheRead: 0.3,
2838
+ cacheWrite: 3.75,
2803
2839
  },
2804
2840
  contextWindow: 1000000,
2805
- maxTokens: 32000,
2841
+ maxTokens: 64000,
2806
2842
  },
2807
2843
  "gemini-2.5-pro": {
2808
2844
  id: "gemini-2.5-pro",
@@ -2812,20 +2848,20 @@ export const MODELS = {
2812
2848
  baseUrl: "https://api.individual.githubcopilot.com",
2813
2849
  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" },
2814
2850
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
2815
- reasoning: false,
2851
+ reasoning: true,
2816
2852
  input: ["text", "image"],
2817
2853
  cost: {
2818
- input: 0,
2819
- output: 0,
2820
- cacheRead: 0,
2854
+ input: 1.25,
2855
+ output: 10,
2856
+ cacheRead: 0.125,
2821
2857
  cacheWrite: 0,
2822
2858
  },
2823
- contextWindow: 128000,
2824
- maxTokens: 64000,
2859
+ contextWindow: 1048576,
2860
+ maxTokens: 65536,
2825
2861
  },
2826
2862
  "gemini-3-flash-preview": {
2827
2863
  id: "gemini-3-flash-preview",
2828
- name: "Gemini 3 Flash",
2864
+ name: "Gemini 3 Flash Preview",
2829
2865
  api: "openai-completions",
2830
2866
  provider: "github-copilot",
2831
2867
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2834,13 +2870,13 @@ export const MODELS = {
2834
2870
  reasoning: true,
2835
2871
  input: ["text", "image"],
2836
2872
  cost: {
2837
- input: 0,
2838
- output: 0,
2839
- cacheRead: 0,
2873
+ input: 0.5,
2874
+ output: 3,
2875
+ cacheRead: 0.05,
2840
2876
  cacheWrite: 0,
2841
2877
  },
2842
- contextWindow: 128000,
2843
- maxTokens: 64000,
2878
+ contextWindow: 1048576,
2879
+ maxTokens: 65536,
2844
2880
  },
2845
2881
  "gemini-3.1-pro-preview": {
2846
2882
  id: "gemini-3.1-pro-preview",
@@ -2853,13 +2889,13 @@ export const MODELS = {
2853
2889
  reasoning: true,
2854
2890
  input: ["text", "image"],
2855
2891
  cost: {
2856
- input: 0,
2857
- output: 0,
2858
- cacheRead: 0,
2892
+ input: 2,
2893
+ output: 12,
2894
+ cacheRead: 0.2,
2859
2895
  cacheWrite: 0,
2860
2896
  },
2861
- contextWindow: 128000,
2862
- maxTokens: 64000,
2897
+ contextWindow: 1048576,
2898
+ maxTokens: 65536,
2863
2899
  },
2864
2900
  "gemini-3.5-flash": {
2865
2901
  id: "gemini-3.5-flash",
@@ -2872,13 +2908,13 @@ export const MODELS = {
2872
2908
  reasoning: true,
2873
2909
  input: ["text", "image"],
2874
2910
  cost: {
2875
- input: 0,
2876
- output: 0,
2877
- cacheRead: 0,
2911
+ input: 1.5,
2912
+ output: 9,
2913
+ cacheRead: 0.15,
2878
2914
  cacheWrite: 0,
2879
2915
  },
2880
- contextWindow: 128000,
2881
- maxTokens: 64000,
2916
+ contextWindow: 1048576,
2917
+ maxTokens: 65536,
2882
2918
  },
2883
2919
  "gpt-4.1": {
2884
2920
  id: "gpt-4.1",
@@ -2891,36 +2927,17 @@ export const MODELS = {
2891
2927
  reasoning: false,
2892
2928
  input: ["text", "image"],
2893
2929
  cost: {
2894
- input: 0,
2895
- output: 0,
2896
- cacheRead: 0,
2897
- cacheWrite: 0,
2898
- },
2899
- contextWindow: 128000,
2900
- maxTokens: 16384,
2901
- },
2902
- "gpt-4o": {
2903
- id: "gpt-4o",
2904
- name: "GPT-4o",
2905
- api: "openai-completions",
2906
- provider: "github-copilot",
2907
- baseUrl: "https://api.individual.githubcopilot.com",
2908
- 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" },
2909
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
2910
- reasoning: false,
2911
- input: ["text", "image"],
2912
- cost: {
2913
- input: 0,
2914
- output: 0,
2915
- cacheRead: 0,
2930
+ input: 2,
2931
+ output: 8,
2932
+ cacheRead: 0.5,
2916
2933
  cacheWrite: 0,
2917
2934
  },
2918
- contextWindow: 128000,
2919
- maxTokens: 4096,
2935
+ contextWindow: 1047576,
2936
+ maxTokens: 32768,
2920
2937
  },
2921
2938
  "gpt-5-mini": {
2922
2939
  id: "gpt-5-mini",
2923
- name: "GPT-5-mini",
2940
+ name: "GPT-5 Mini",
2924
2941
  api: "openai-responses",
2925
2942
  provider: "github-copilot",
2926
2943
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2928,13 +2945,13 @@ export const MODELS = {
2928
2945
  reasoning: true,
2929
2946
  input: ["text", "image"],
2930
2947
  cost: {
2931
- input: 0,
2932
- output: 0,
2933
- cacheRead: 0,
2948
+ input: 0.25,
2949
+ output: 2,
2950
+ cacheRead: 0.025,
2934
2951
  cacheWrite: 0,
2935
2952
  },
2936
- contextWindow: 264000,
2937
- maxTokens: 64000,
2953
+ contextWindow: 400000,
2954
+ maxTokens: 128000,
2938
2955
  },
2939
2956
  "gpt-5.2": {
2940
2957
  id: "gpt-5.2",
@@ -2946,17 +2963,17 @@ export const MODELS = {
2946
2963
  reasoning: true,
2947
2964
  input: ["text", "image"],
2948
2965
  cost: {
2949
- input: 0,
2950
- output: 0,
2951
- cacheRead: 0,
2966
+ input: 1.75,
2967
+ output: 14,
2968
+ cacheRead: 0.175,
2952
2969
  cacheWrite: 0,
2953
2970
  },
2954
- contextWindow: 264000,
2955
- maxTokens: 64000,
2971
+ contextWindow: 400000,
2972
+ maxTokens: 128000,
2956
2973
  },
2957
2974
  "gpt-5.2-codex": {
2958
2975
  id: "gpt-5.2-codex",
2959
- name: "GPT-5.2-Codex",
2976
+ name: "GPT-5.2 Codex",
2960
2977
  api: "openai-responses",
2961
2978
  provider: "github-copilot",
2962
2979
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2964,9 +2981,9 @@ export const MODELS = {
2964
2981
  reasoning: true,
2965
2982
  input: ["text", "image"],
2966
2983
  cost: {
2967
- input: 0,
2968
- output: 0,
2969
- cacheRead: 0,
2984
+ input: 1.75,
2985
+ output: 14,
2986
+ cacheRead: 0.175,
2970
2987
  cacheWrite: 0,
2971
2988
  },
2972
2989
  contextWindow: 400000,
@@ -2974,7 +2991,7 @@ export const MODELS = {
2974
2991
  },
2975
2992
  "gpt-5.3-codex": {
2976
2993
  id: "gpt-5.3-codex",
2977
- name: "GPT-5.3-Codex",
2994
+ name: "GPT-5.3 Codex",
2978
2995
  api: "openai-responses",
2979
2996
  provider: "github-copilot",
2980
2997
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2982,9 +2999,9 @@ export const MODELS = {
2982
2999
  reasoning: true,
2983
3000
  input: ["text", "image"],
2984
3001
  cost: {
2985
- input: 0,
2986
- output: 0,
2987
- cacheRead: 0,
3002
+ input: 1.75,
3003
+ output: 14,
3004
+ cacheRead: 0.175,
2988
3005
  cacheWrite: 0,
2989
3006
  },
2990
3007
  contextWindow: 400000,
@@ -3000,17 +3017,17 @@ export const MODELS = {
3000
3017
  reasoning: true,
3001
3018
  input: ["text", "image"],
3002
3019
  cost: {
3003
- input: 0,
3004
- output: 0,
3005
- cacheRead: 0,
3020
+ input: 2.5,
3021
+ output: 15,
3022
+ cacheRead: 0.25,
3006
3023
  cacheWrite: 0,
3007
3024
  },
3008
- contextWindow: 400000,
3025
+ contextWindow: 1050000,
3009
3026
  maxTokens: 128000,
3010
3027
  },
3011
3028
  "gpt-5.4-mini": {
3012
3029
  id: "gpt-5.4-mini",
3013
- name: "GPT-5.4 Mini",
3030
+ name: "GPT-5.4 mini",
3014
3031
  api: "openai-responses",
3015
3032
  provider: "github-copilot",
3016
3033
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3018,9 +3035,27 @@ export const MODELS = {
3018
3035
  reasoning: true,
3019
3036
  input: ["text", "image"],
3020
3037
  cost: {
3021
- input: 0,
3022
- output: 0,
3023
- cacheRead: 0,
3038
+ input: 0.75,
3039
+ output: 4.5,
3040
+ cacheRead: 0.075,
3041
+ cacheWrite: 0,
3042
+ },
3043
+ contextWindow: 400000,
3044
+ maxTokens: 128000,
3045
+ },
3046
+ "gpt-5.4-nano": {
3047
+ id: "gpt-5.4-nano",
3048
+ name: "GPT-5.4 nano",
3049
+ api: "openai-responses",
3050
+ provider: "github-copilot",
3051
+ baseUrl: "https://api.individual.githubcopilot.com",
3052
+ 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" },
3053
+ reasoning: true,
3054
+ input: ["text", "image"],
3055
+ cost: {
3056
+ input: 0.2,
3057
+ output: 1.25,
3058
+ cacheRead: 0.02,
3024
3059
  cacheWrite: 0,
3025
3060
  },
3026
3061
  contextWindow: 400000,
@@ -3036,32 +3071,32 @@ export const MODELS = {
3036
3071
  reasoning: true,
3037
3072
  input: ["text", "image"],
3038
3073
  cost: {
3039
- input: 0,
3040
- output: 0,
3041
- cacheRead: 0,
3074
+ input: 5,
3075
+ output: 30,
3076
+ cacheRead: 0.5,
3042
3077
  cacheWrite: 0,
3043
3078
  },
3044
- contextWindow: 400000,
3079
+ contextWindow: 1050000,
3045
3080
  maxTokens: 128000,
3046
3081
  },
3047
- "grok-code-fast-1": {
3048
- id: "grok-code-fast-1",
3049
- name: "Grok Code Fast 1",
3082
+ "raptor-mini": {
3083
+ id: "raptor-mini",
3084
+ name: "Raptor mini",
3050
3085
  api: "openai-completions",
3051
3086
  provider: "github-copilot",
3052
3087
  baseUrl: "https://api.individual.githubcopilot.com",
3053
3088
  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" },
3054
3089
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
3055
3090
  reasoning: true,
3056
- input: ["text"],
3091
+ input: ["text", "image"],
3057
3092
  cost: {
3058
- input: 0,
3059
- output: 0,
3060
- cacheRead: 0,
3093
+ input: 0.25,
3094
+ output: 2,
3095
+ cacheRead: 0.025,
3061
3096
  cacheWrite: 0,
3062
3097
  },
3063
- contextWindow: 128000,
3064
- maxTokens: 64000,
3098
+ contextWindow: 400000,
3099
+ maxTokens: 128000,
3065
3100
  },
3066
3101
  },
3067
3102
  "google": {
@@ -6140,6 +6175,23 @@ export const MODELS = {
6140
6175
  contextWindow: 1000000,
6141
6176
  maxTokens: 64000,
6142
6177
  },
6178
+ "deepseek-v4-flash": {
6179
+ id: "deepseek-v4-flash",
6180
+ name: "DeepSeek V4 Flash",
6181
+ api: "openai-completions",
6182
+ provider: "opencode",
6183
+ baseUrl: "https://opencode.ai/zen/v1",
6184
+ reasoning: true,
6185
+ input: ["text"],
6186
+ cost: {
6187
+ input: 0.14,
6188
+ output: 0.28,
6189
+ cacheRead: 0.03,
6190
+ cacheWrite: 0,
6191
+ },
6192
+ contextWindow: 1000000,
6193
+ maxTokens: 384000,
6194
+ },
6143
6195
  "deepseek-v4-flash-free": {
6144
6196
  id: "deepseek-v4-flash-free",
6145
6197
  name: "DeepSeek V4 Flash Free",
@@ -6579,8 +6631,8 @@ export const MODELS = {
6579
6631
  cacheRead: 0,
6580
6632
  cacheWrite: 0,
6581
6633
  },
6582
- contextWindow: 1000000,
6583
- maxTokens: 128000,
6634
+ contextWindow: 200000,
6635
+ maxTokens: 32000,
6584
6636
  },
6585
6637
  "minimax-m2.5": {
6586
6638
  id: "minimax-m2.5",
@@ -6616,6 +6668,23 @@ export const MODELS = {
6616
6668
  contextWindow: 204800,
6617
6669
  maxTokens: 131072,
6618
6670
  },
6671
+ "minimax-m3-free": {
6672
+ id: "minimax-m3-free",
6673
+ name: "MiniMax M3 Free",
6674
+ api: "anthropic-messages",
6675
+ provider: "opencode",
6676
+ baseUrl: "https://opencode.ai/zen",
6677
+ reasoning: true,
6678
+ input: ["text", "image"],
6679
+ cost: {
6680
+ input: 0,
6681
+ output: 0,
6682
+ cacheRead: 0,
6683
+ cacheWrite: 0,
6684
+ },
6685
+ contextWindow: 200000,
6686
+ maxTokens: 32000,
6687
+ },
6619
6688
  "nemotron-3-super-free": {
6620
6689
  id: "nemotron-3-super-free",
6621
6690
  name: "Nemotron 3 Super Free",
@@ -6839,22 +6908,22 @@ export const MODELS = {
6839
6908
  contextWindow: 204800,
6840
6909
  maxTokens: 131072,
6841
6910
  },
6842
- "qwen3.5-plus": {
6843
- id: "qwen3.5-plus",
6844
- name: "Qwen3.5 Plus",
6911
+ "minimax-m3": {
6912
+ id: "minimax-m3",
6913
+ name: "MiniMax M3",
6845
6914
  api: "anthropic-messages",
6846
6915
  provider: "opencode-go",
6847
6916
  baseUrl: "https://opencode.ai/zen/go",
6848
6917
  reasoning: true,
6849
6918
  input: ["text", "image"],
6850
6919
  cost: {
6851
- input: 0.2,
6852
- output: 1.2,
6853
- cacheRead: 0.02,
6854
- cacheWrite: 0.25,
6920
+ input: 0.6,
6921
+ output: 2.4,
6922
+ cacheRead: 0.12,
6923
+ cacheWrite: 0,
6855
6924
  },
6856
- contextWindow: 262144,
6857
- maxTokens: 65536,
6925
+ contextWindow: 512000,
6926
+ maxTokens: 131072,
6858
6927
  },
6859
6928
  "qwen3.6-plus": {
6860
6929
  id: "qwen3.6-plus",
@@ -7317,23 +7386,6 @@ export const MODELS = {
7317
7386
  contextWindow: 2000000,
7318
7387
  maxTokens: 30000,
7319
7388
  },
7320
- "baidu/ernie-4.5-21b-a3b": {
7321
- id: "baidu/ernie-4.5-21b-a3b",
7322
- name: "Baidu: ERNIE 4.5 21B A3B",
7323
- api: "openai-completions",
7324
- provider: "openrouter",
7325
- baseUrl: "https://openrouter.ai/api/v1",
7326
- reasoning: false,
7327
- input: ["text"],
7328
- cost: {
7329
- input: 0.07,
7330
- output: 0.28,
7331
- cacheRead: 0,
7332
- cacheWrite: 0,
7333
- },
7334
- contextWindow: 131072,
7335
- maxTokens: 8000,
7336
- },
7337
7389
  "baidu/ernie-4.5-vl-28b-a3b": {
7338
7390
  id: "baidu/ernie-4.5-vl-28b-a3b",
7339
7391
  name: "Baidu: ERNIE 4.5 VL 28B A3B",
@@ -7462,8 +7514,8 @@ export const MODELS = {
7462
7514
  reasoning: false,
7463
7515
  input: ["text"],
7464
7516
  cost: {
7465
- input: 0.2288,
7466
- output: 0.9144,
7517
+ input: 0.20020000000000002,
7518
+ output: 0.8000999999999999,
7467
7519
  cacheRead: 0,
7468
7520
  cacheWrite: 0,
7469
7521
  },
@@ -7564,13 +7616,13 @@ export const MODELS = {
7564
7616
  reasoning: true,
7565
7617
  input: ["text"],
7566
7618
  cost: {
7567
- input: 0.252,
7568
- output: 0.378,
7569
- cacheRead: 0.0252,
7619
+ input: 0.2288,
7620
+ output: 0.3432,
7621
+ cacheRead: 0,
7570
7622
  cacheWrite: 0,
7571
7623
  },
7572
7624
  contextWindow: 131072,
7573
- maxTokens: 65536,
7625
+ maxTokens: 64000,
7574
7626
  },
7575
7627
  "deepseek/deepseek-v3.2-exp": {
7576
7628
  id: "deepseek/deepseek-v3.2-exp",
@@ -7598,30 +7650,13 @@ export const MODELS = {
7598
7650
  reasoning: true,
7599
7651
  input: ["text"],
7600
7652
  cost: {
7601
- input: 0.09999999999999999,
7602
- output: 0.19999999999999998,
7603
- cacheRead: 0.02,
7604
- cacheWrite: 0,
7605
- },
7606
- contextWindow: 1048576,
7607
- maxTokens: 16384,
7608
- },
7609
- "deepseek/deepseek-v4-flash:free": {
7610
- id: "deepseek/deepseek-v4-flash:free",
7611
- name: "DeepSeek: DeepSeek V4 Flash (free)",
7612
- api: "openai-completions",
7613
- provider: "openrouter",
7614
- baseUrl: "https://openrouter.ai/api/v1",
7615
- reasoning: true,
7616
- input: ["text"],
7617
- cost: {
7618
- input: 0,
7619
- output: 0,
7620
- cacheRead: 0,
7653
+ input: 0.0983,
7654
+ output: 0.1966,
7655
+ cacheRead: 0.019700000000000002,
7621
7656
  cacheWrite: 0,
7622
7657
  },
7623
7658
  contextWindow: 1048576,
7624
- maxTokens: 384000,
7659
+ maxTokens: 131072,
7625
7660
  },
7626
7661
  "deepseek/deepseek-v4-pro": {
7627
7662
  id: "deepseek/deepseek-v4-pro",
@@ -7657,40 +7692,6 @@ export const MODELS = {
7657
7692
  contextWindow: 32768,
7658
7693
  maxTokens: 4096,
7659
7694
  },
7660
- "google/gemini-2.0-flash-001": {
7661
- id: "google/gemini-2.0-flash-001",
7662
- name: "Google: Gemini 2.0 Flash",
7663
- api: "openai-completions",
7664
- provider: "openrouter",
7665
- baseUrl: "https://openrouter.ai/api/v1",
7666
- reasoning: false,
7667
- input: ["text", "image"],
7668
- cost: {
7669
- input: 0.09999999999999999,
7670
- output: 0.39999999999999997,
7671
- cacheRead: 0.024999999999999998,
7672
- cacheWrite: 0.08333333333333334,
7673
- },
7674
- contextWindow: 1000000,
7675
- maxTokens: 8192,
7676
- },
7677
- "google/gemini-2.0-flash-lite-001": {
7678
- id: "google/gemini-2.0-flash-lite-001",
7679
- name: "Google: Gemini 2.0 Flash Lite",
7680
- api: "openai-completions",
7681
- provider: "openrouter",
7682
- baseUrl: "https://openrouter.ai/api/v1",
7683
- reasoning: false,
7684
- input: ["text", "image"],
7685
- cost: {
7686
- input: 0.075,
7687
- output: 0.3,
7688
- cacheRead: 0,
7689
- cacheWrite: 0,
7690
- },
7691
- contextWindow: 1048576,
7692
- maxTokens: 8192,
7693
- },
7694
7695
  "google/gemini-2.5-flash": {
7695
7696
  id: "google/gemini-2.5-flash",
7696
7697
  name: "Google: Gemini 2.5 Flash",
@@ -8074,9 +8075,9 @@ export const MODELS = {
8074
8075
  reasoning: true,
8075
8076
  input: ["text"],
8076
8077
  cost: {
8077
- input: 0.075,
8078
- output: 0.625,
8079
- cacheRead: 0.015,
8078
+ input: 0.3,
8079
+ output: 2.5,
8080
+ cacheRead: 0.06,
8080
8081
  cacheWrite: 0,
8081
8082
  },
8082
8083
  contextWindow: 262144,
@@ -8167,6 +8168,23 @@ export const MODELS = {
8167
8168
  contextWindow: 131072,
8168
8169
  maxTokens: 4096,
8169
8170
  },
8171
+ "meta-llama/llama-4-maverick": {
8172
+ id: "meta-llama/llama-4-maverick",
8173
+ name: "Meta: Llama 4 Maverick",
8174
+ api: "openai-completions",
8175
+ provider: "openrouter",
8176
+ baseUrl: "https://openrouter.ai/api/v1",
8177
+ reasoning: false,
8178
+ input: ["text", "image"],
8179
+ cost: {
8180
+ input: 0.15,
8181
+ output: 0.6,
8182
+ cacheRead: 0,
8183
+ cacheWrite: 0,
8184
+ },
8185
+ contextWindow: 1048576,
8186
+ maxTokens: 16384,
8187
+ },
8170
8188
  "meta-llama/llama-4-scout": {
8171
8189
  id: "meta-llama/llama-4-scout",
8172
8190
  name: "Meta: Llama 4 Scout",
@@ -8252,39 +8270,39 @@ export const MODELS = {
8252
8270
  contextWindow: 204800,
8253
8271
  maxTokens: 196608,
8254
8272
  },
8255
- "minimax/minimax-m2.5:free": {
8256
- id: "minimax/minimax-m2.5:free",
8257
- name: "MiniMax: MiniMax M2.5 (free)",
8273
+ "minimax/minimax-m2.7": {
8274
+ id: "minimax/minimax-m2.7",
8275
+ name: "MiniMax: MiniMax M2.7",
8258
8276
  api: "openai-completions",
8259
8277
  provider: "openrouter",
8260
8278
  baseUrl: "https://openrouter.ai/api/v1",
8261
8279
  reasoning: true,
8262
8280
  input: ["text"],
8263
8281
  cost: {
8264
- input: 0,
8265
- output: 0,
8282
+ input: 0.27899999999999997,
8283
+ output: 1.2,
8266
8284
  cacheRead: 0,
8267
8285
  cacheWrite: 0,
8268
8286
  },
8269
- contextWindow: 262144,
8270
- maxTokens: 8192,
8287
+ contextWindow: 204800,
8288
+ maxTokens: 131072,
8271
8289
  },
8272
- "minimax/minimax-m2.7": {
8273
- id: "minimax/minimax-m2.7",
8274
- name: "MiniMax: MiniMax M2.7",
8290
+ "minimax/minimax-m3": {
8291
+ id: "minimax/minimax-m3",
8292
+ name: "MiniMax: MiniMax M3",
8275
8293
  api: "openai-completions",
8276
8294
  provider: "openrouter",
8277
8295
  baseUrl: "https://openrouter.ai/api/v1",
8278
8296
  reasoning: true,
8279
- input: ["text"],
8297
+ input: ["text", "image"],
8280
8298
  cost: {
8281
- input: 0.27899999999999997,
8299
+ input: 0.3,
8282
8300
  output: 1.2,
8283
- cacheRead: 0,
8301
+ cacheRead: 0.06,
8284
8302
  cacheWrite: 0,
8285
8303
  },
8286
- contextWindow: 204800,
8287
- maxTokens: 131072,
8304
+ contextWindow: 1048576,
8305
+ maxTokens: 512000,
8288
8306
  },
8289
8307
  "mistralai/codestral-2508": {
8290
8308
  id: "mistralai/codestral-2508",
@@ -8320,40 +8338,6 @@ export const MODELS = {
8320
8338
  contextWindow: 262144,
8321
8339
  maxTokens: 4096,
8322
8340
  },
8323
- "mistralai/devstral-medium": {
8324
- id: "mistralai/devstral-medium",
8325
- name: "Mistral: Devstral Medium",
8326
- api: "openai-completions",
8327
- provider: "openrouter",
8328
- baseUrl: "https://openrouter.ai/api/v1",
8329
- reasoning: false,
8330
- input: ["text"],
8331
- cost: {
8332
- input: 0.39999999999999997,
8333
- output: 2,
8334
- cacheRead: 0.04,
8335
- cacheWrite: 0,
8336
- },
8337
- contextWindow: 131072,
8338
- maxTokens: 4096,
8339
- },
8340
- "mistralai/devstral-small": {
8341
- id: "mistralai/devstral-small",
8342
- name: "Mistral: Devstral Small 1.1",
8343
- api: "openai-completions",
8344
- provider: "openrouter",
8345
- baseUrl: "https://openrouter.ai/api/v1",
8346
- reasoning: false,
8347
- input: ["text"],
8348
- cost: {
8349
- input: 0.09999999999999999,
8350
- output: 0.3,
8351
- cacheRead: 0.01,
8352
- cacheWrite: 0,
8353
- },
8354
- contextWindow: 131072,
8355
- maxTokens: 4096,
8356
- },
8357
8341
  "mistralai/ministral-14b-2512": {
8358
8342
  id: "mistralai/ministral-14b-2512",
8359
8343
  name: "Mistral: Ministral 3 14B 2512",
@@ -8439,23 +8423,6 @@ export const MODELS = {
8439
8423
  contextWindow: 131072,
8440
8424
  maxTokens: 4096,
8441
8425
  },
8442
- "mistralai/mistral-large-2411": {
8443
- id: "mistralai/mistral-large-2411",
8444
- name: "Mistral Large 2411",
8445
- api: "openai-completions",
8446
- provider: "openrouter",
8447
- baseUrl: "https://openrouter.ai/api/v1",
8448
- reasoning: false,
8449
- input: ["text"],
8450
- cost: {
8451
- input: 2,
8452
- output: 6,
8453
- cacheRead: 0.19999999999999998,
8454
- cacheWrite: 0,
8455
- },
8456
- contextWindow: 131072,
8457
- maxTokens: 4096,
8458
- },
8459
8426
  "mistralai/mistral-large-2512": {
8460
8427
  id: "mistralai/mistral-large-2512",
8461
8428
  name: "Mistral: Mistral Large 3 2512",
@@ -8609,23 +8576,6 @@ export const MODELS = {
8609
8576
  contextWindow: 65536,
8610
8577
  maxTokens: 4096,
8611
8578
  },
8612
- "mistralai/pixtral-large-2411": {
8613
- id: "mistralai/pixtral-large-2411",
8614
- name: "Mistral: Pixtral Large 2411",
8615
- api: "openai-completions",
8616
- provider: "openrouter",
8617
- baseUrl: "https://openrouter.ai/api/v1",
8618
- reasoning: false,
8619
- input: ["text", "image"],
8620
- cost: {
8621
- input: 2,
8622
- output: 6,
8623
- cacheRead: 0.19999999999999998,
8624
- cacheWrite: 0,
8625
- },
8626
- contextWindow: 131072,
8627
- maxTokens: 4096,
8628
- },
8629
8579
  "mistralai/voxtral-small-24b-2507": {
8630
8580
  id: "mistralai/voxtral-small-24b-2507",
8631
8581
  name: "Mistral: Voxtral Small 24B 2507",
@@ -8720,13 +8670,13 @@ export const MODELS = {
8720
8670
  reasoning: true,
8721
8671
  input: ["text", "image"],
8722
8672
  cost: {
8723
- input: 0.73,
8724
- output: 3.49,
8725
- cacheRead: 0.25,
8673
+ input: 0.684,
8674
+ output: 3.42,
8675
+ cacheRead: 0.144,
8726
8676
  cacheWrite: 0,
8727
8677
  },
8728
8678
  contextWindow: 262144,
8729
- maxTokens: 262142,
8679
+ maxTokens: 262144,
8730
8680
  },
8731
8681
  "moonshotai/kimi-k2.6:free": {
8732
8682
  id: "moonshotai/kimi-k2.6:free",
@@ -9170,23 +9120,6 @@ export const MODELS = {
9170
9120
  contextWindow: 128000,
9171
9121
  maxTokens: 16384,
9172
9122
  },
9173
- "openai/gpt-4o-audio-preview": {
9174
- id: "openai/gpt-4o-audio-preview",
9175
- name: "OpenAI: GPT-4o Audio",
9176
- api: "openai-completions",
9177
- provider: "openrouter",
9178
- baseUrl: "https://openrouter.ai/api/v1",
9179
- reasoning: false,
9180
- input: ["text"],
9181
- cost: {
9182
- input: 2.5,
9183
- output: 10,
9184
- cacheRead: 0,
9185
- cacheWrite: 0,
9186
- },
9187
- contextWindow: 128000,
9188
- maxTokens: 16384,
9189
- },
9190
9123
  "openai/gpt-4o-mini": {
9191
9124
  id: "openai/gpt-4o-mini",
9192
9125
  name: "OpenAI: GPT-4o-mini",
@@ -9689,13 +9622,13 @@ export const MODELS = {
9689
9622
  reasoning: true,
9690
9623
  input: ["text"],
9691
9624
  cost: {
9692
- input: 0.03,
9625
+ input: 0.029,
9693
9626
  output: 0.14,
9694
9627
  cacheRead: 0,
9695
9628
  cacheWrite: 0,
9696
9629
  },
9697
9630
  contextWindow: 131072,
9698
- maxTokens: 131072,
9631
+ maxTokens: 4096,
9699
9632
  },
9700
9633
  "openai/gpt-oss-20b:free": {
9701
9634
  id: "openai/gpt-oss-20b:free",
@@ -10003,23 +9936,6 @@ export const MODELS = {
10003
9936
  contextWindow: 131072,
10004
9937
  maxTokens: 16384,
10005
9938
  },
10006
- "qwen/qwen-2.5-7b-instruct": {
10007
- id: "qwen/qwen-2.5-7b-instruct",
10008
- name: "Qwen: Qwen2.5 7B Instruct",
10009
- api: "openai-completions",
10010
- provider: "openrouter",
10011
- baseUrl: "https://openrouter.ai/api/v1",
10012
- reasoning: false,
10013
- input: ["text"],
10014
- cost: {
10015
- input: 0.04,
10016
- output: 0.09999999999999999,
10017
- cacheRead: 0,
10018
- cacheWrite: 0,
10019
- },
10020
- contextWindow: 131072,
10021
- maxTokens: 32768,
10022
- },
10023
9939
  "qwen/qwen-plus": {
10024
9940
  id: "qwen/qwen-plus",
10025
9941
  name: "Qwen: Qwen-Plus",
@@ -10131,13 +10047,13 @@ export const MODELS = {
10131
10047
  reasoning: true,
10132
10048
  input: ["text"],
10133
10049
  cost: {
10134
- input: 0.14950000000000002,
10135
- output: 1.495,
10136
- cacheRead: 0,
10050
+ input: 0.09999999999999999,
10051
+ output: 0.09999999999999999,
10052
+ cacheRead: 0.09999999999999999,
10137
10053
  cacheWrite: 0,
10138
10054
  },
10139
10055
  contextWindow: 262144,
10140
- maxTokens: 4096,
10056
+ maxTokens: 262144,
10141
10057
  },
10142
10058
  "qwen/qwen3-30b-a3b": {
10143
10059
  id: "qwen/qwen3-30b-a3b",
@@ -10165,13 +10081,13 @@ export const MODELS = {
10165
10081
  reasoning: false,
10166
10082
  input: ["text"],
10167
10083
  cost: {
10168
- input: 0.09,
10169
- output: 0.3,
10084
+ input: 0.0428,
10085
+ output: 0.1716,
10170
10086
  cacheRead: 0,
10171
10087
  cacheWrite: 0,
10172
10088
  },
10173
- contextWindow: 262144,
10174
- maxTokens: 262144,
10089
+ contextWindow: 131072,
10090
+ maxTokens: 32000,
10175
10091
  },
10176
10092
  "qwen/qwen3-30b-a3b-thinking-2507": {
10177
10093
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -10573,13 +10489,13 @@ export const MODELS = {
10573
10489
  reasoning: true,
10574
10490
  input: ["text", "image"],
10575
10491
  cost: {
10576
- input: 0.13899999999999998,
10492
+ input: 0.14,
10577
10493
  output: 1,
10578
- cacheRead: 0,
10494
+ cacheRead: 0.049999999999999996,
10579
10495
  cacheWrite: 0,
10580
10496
  },
10581
10497
  contextWindow: 262144,
10582
- maxTokens: 4096,
10498
+ maxTokens: 262144,
10583
10499
  },
10584
10500
  "qwen/qwen3.5-397b-a17b": {
10585
10501
  id: "qwen/qwen3.5-397b-a17b",
@@ -10853,6 +10769,23 @@ export const MODELS = {
10853
10769
  contextWindow: 262144,
10854
10770
  maxTokens: 16384,
10855
10771
  },
10772
+ "stepfun/step-3.7-flash": {
10773
+ id: "stepfun/step-3.7-flash",
10774
+ name: "StepFun: Step 3.7 Flash",
10775
+ api: "openai-completions",
10776
+ provider: "openrouter",
10777
+ baseUrl: "https://openrouter.ai/api/v1",
10778
+ reasoning: true,
10779
+ input: ["text", "image"],
10780
+ cost: {
10781
+ input: 0.19999999999999998,
10782
+ output: 1.15,
10783
+ cacheRead: 0.04,
10784
+ cacheWrite: 0,
10785
+ },
10786
+ contextWindow: 256000,
10787
+ maxTokens: 256000,
10788
+ },
10856
10789
  "tencent/hy3-preview": {
10857
10790
  id: "tencent/hy3-preview",
10858
10791
  name: "Tencent: Hy3 preview",
@@ -10989,40 +10922,6 @@ export const MODELS = {
10989
10922
  contextWindow: 262144,
10990
10923
  maxTokens: 65536,
10991
10924
  },
10992
- "xiaomi/mimo-v2-omni": {
10993
- id: "xiaomi/mimo-v2-omni",
10994
- name: "Xiaomi: MiMo-V2-Omni",
10995
- api: "openai-completions",
10996
- provider: "openrouter",
10997
- baseUrl: "https://openrouter.ai/api/v1",
10998
- reasoning: true,
10999
- input: ["text", "image"],
11000
- cost: {
11001
- input: 0.39999999999999997,
11002
- output: 2,
11003
- cacheRead: 0.08,
11004
- cacheWrite: 0,
11005
- },
11006
- contextWindow: 262144,
11007
- maxTokens: 65536,
11008
- },
11009
- "xiaomi/mimo-v2-pro": {
11010
- id: "xiaomi/mimo-v2-pro",
11011
- name: "Xiaomi: MiMo-V2-Pro",
11012
- api: "openai-completions",
11013
- provider: "openrouter",
11014
- baseUrl: "https://openrouter.ai/api/v1",
11015
- reasoning: true,
11016
- input: ["text"],
11017
- cost: {
11018
- input: 1,
11019
- output: 3,
11020
- cacheRead: 0.19999999999999998,
11021
- cacheWrite: 0,
11022
- },
11023
- contextWindow: 1048576,
11024
- maxTokens: 131072,
11025
- },
11026
10925
  "xiaomi/mimo-v2.5": {
11027
10926
  id: "xiaomi/mimo-v2.5",
11028
10927
  name: "Xiaomi: MiMo-V2.5",
@@ -11220,12 +11119,12 @@ export const MODELS = {
11220
11119
  input: ["text"],
11221
11120
  cost: {
11222
11121
  input: 0.6,
11223
- output: 1.92,
11122
+ output: 2.08,
11224
11123
  cacheRead: 0.12,
11225
11124
  cacheWrite: 0,
11226
11125
  },
11227
11126
  contextWindow: 202752,
11228
- maxTokens: 4096,
11127
+ maxTokens: 16384,
11229
11128
  },
11230
11129
  "z-ai/glm-5-turbo": {
11231
11130
  id: "z-ai/glm-5-turbo",
@@ -11259,7 +11158,7 @@ export const MODELS = {
11259
11158
  cacheWrite: 0,
11260
11159
  },
11261
11160
  contextWindow: 202752,
11262
- maxTokens: 4096,
11161
+ maxTokens: 131072,
11263
11162
  },
11264
11163
  "z-ai/glm-5v-turbo": {
11265
11164
  id: "z-ai/glm-5v-turbo",
@@ -11372,13 +11271,13 @@ export const MODELS = {
11372
11271
  reasoning: true,
11373
11272
  input: ["text", "image"],
11374
11273
  cost: {
11375
- input: 0.73,
11376
- output: 3.49,
11377
- cacheRead: 0.25,
11274
+ input: 0.684,
11275
+ output: 3.42,
11276
+ cacheRead: 0.144,
11378
11277
  cacheWrite: 0,
11379
11278
  },
11380
11279
  contextWindow: 262144,
11381
- maxTokens: 262142,
11280
+ maxTokens: 262144,
11382
11281
  },
11383
11282
  "~openai/gpt-latest": {
11384
11283
  id: "~openai/gpt-latest",
@@ -11435,20 +11334,20 @@ export const MODELS = {
11435
11334
  },
11436
11335
  "alibaba/qwen-3-235b": {
11437
11336
  id: "alibaba/qwen-3-235b",
11438
- name: "Qwen3 235B A22b Instruct 2507",
11337
+ name: "Qwen3 235B A22B",
11439
11338
  api: "anthropic-messages",
11440
11339
  provider: "vercel-ai-gateway",
11441
11340
  baseUrl: "https://ai-gateway.vercel.sh",
11442
- reasoning: false,
11341
+ reasoning: true,
11443
11342
  input: ["text"],
11444
11343
  cost: {
11445
- input: 0.6,
11446
- output: 1.2,
11447
- cacheRead: 0.6,
11344
+ input: 0.22,
11345
+ output: 0.88,
11346
+ cacheRead: 0,
11448
11347
  cacheWrite: 0,
11449
11348
  },
11450
- contextWindow: 131000,
11451
- maxTokens: 40000,
11349
+ contextWindow: 262144,
11350
+ maxTokens: 16384,
11452
11351
  },
11453
11352
  "alibaba/qwen-3-30b": {
11454
11353
  id: "alibaba/qwen-3-30b",
@@ -11524,7 +11423,7 @@ export const MODELS = {
11524
11423
  api: "anthropic-messages",
11525
11424
  provider: "vercel-ai-gateway",
11526
11425
  baseUrl: "https://ai-gateway.vercel.sh",
11527
- reasoning: false,
11426
+ reasoning: true,
11528
11427
  input: ["text"],
11529
11428
  cost: {
11530
11429
  input: 1.5,
@@ -11558,7 +11457,7 @@ export const MODELS = {
11558
11457
  api: "anthropic-messages",
11559
11458
  provider: "vercel-ai-gateway",
11560
11459
  baseUrl: "https://ai-gateway.vercel.sh",
11561
- reasoning: false,
11460
+ reasoning: true,
11562
11461
  input: ["text"],
11563
11462
  cost: {
11564
11463
  input: 0.5,
@@ -11637,6 +11536,40 @@ export const MODELS = {
11637
11536
  contextWindow: 256000,
11638
11537
  maxTokens: 65536,
11639
11538
  },
11539
+ "alibaba/qwen3-next-80b-a3b-instruct": {
11540
+ id: "alibaba/qwen3-next-80b-a3b-instruct",
11541
+ name: "Qwen3 Next 80B A3B Instruct",
11542
+ api: "anthropic-messages",
11543
+ provider: "vercel-ai-gateway",
11544
+ baseUrl: "https://ai-gateway.vercel.sh",
11545
+ reasoning: false,
11546
+ input: ["text"],
11547
+ cost: {
11548
+ input: 0.15,
11549
+ output: 1.2,
11550
+ cacheRead: 0,
11551
+ cacheWrite: 0,
11552
+ },
11553
+ contextWindow: 131072,
11554
+ maxTokens: 32768,
11555
+ },
11556
+ "alibaba/qwen3-next-80b-a3b-thinking": {
11557
+ id: "alibaba/qwen3-next-80b-a3b-thinking",
11558
+ name: "Qwen3 Next 80B A3B Thinking",
11559
+ api: "anthropic-messages",
11560
+ provider: "vercel-ai-gateway",
11561
+ baseUrl: "https://ai-gateway.vercel.sh",
11562
+ reasoning: true,
11563
+ input: ["text"],
11564
+ cost: {
11565
+ input: 0.15,
11566
+ output: 1.2,
11567
+ cacheRead: 0,
11568
+ cacheWrite: 0,
11569
+ },
11570
+ contextWindow: 131072,
11571
+ maxTokens: 32768,
11572
+ },
11640
11573
  "alibaba/qwen3-vl-thinking": {
11641
11574
  id: "alibaba/qwen3-vl-thinking",
11642
11575
  name: "Qwen3 VL 235B A22B Thinking",
@@ -11739,6 +11672,23 @@ export const MODELS = {
11739
11672
  contextWindow: 991000,
11740
11673
  maxTokens: 64000,
11741
11674
  },
11675
+ "alibaba/qwen3.7-plus": {
11676
+ id: "alibaba/qwen3.7-plus",
11677
+ name: "Qwen 3.7 Plus",
11678
+ api: "anthropic-messages",
11679
+ provider: "vercel-ai-gateway",
11680
+ baseUrl: "https://ai-gateway.vercel.sh",
11681
+ reasoning: true,
11682
+ input: ["text", "image"],
11683
+ cost: {
11684
+ input: 0.39999999999999997,
11685
+ output: 1.5999999999999999,
11686
+ cacheRead: 0.08,
11687
+ cacheWrite: 0.5,
11688
+ },
11689
+ contextWindow: 1000000,
11690
+ maxTokens: 64000,
11691
+ },
11742
11692
  "anthropic/claude-3-haiku": {
11743
11693
  id: "anthropic/claude-3-haiku",
11744
11694
  name: "Claude 3 Haiku",
@@ -12037,17 +11987,17 @@ export const MODELS = {
12037
11987
  reasoning: false,
12038
11988
  input: ["text"],
12039
11989
  cost: {
12040
- input: 0.77,
12041
- output: 0.77,
12042
- cacheRead: 0,
11990
+ input: 0.27,
11991
+ output: 1.12,
11992
+ cacheRead: 0.135,
12043
11993
  cacheWrite: 0,
12044
11994
  },
12045
11995
  contextWindow: 163840,
12046
- maxTokens: 16384,
11996
+ maxTokens: 163840,
12047
11997
  },
12048
11998
  "deepseek/deepseek-v3.1": {
12049
11999
  id: "deepseek/deepseek-v3.1",
12050
- name: "DeepSeek-V3.1",
12000
+ name: "DeepSeek V3.1",
12051
12001
  api: "anthropic-messages",
12052
12002
  provider: "vercel-ai-gateway",
12053
12003
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -12085,8 +12035,8 @@ export const MODELS = {
12085
12035
  api: "anthropic-messages",
12086
12036
  provider: "vercel-ai-gateway",
12087
12037
  baseUrl: "https://ai-gateway.vercel.sh",
12088
- reasoning: false,
12089
- input: ["text"],
12038
+ reasoning: true,
12039
+ input: ["text", "image"],
12090
12040
  cost: {
12091
12041
  input: 0.28,
12092
12042
  output: 0.42,
@@ -12102,8 +12052,8 @@ export const MODELS = {
12102
12052
  api: "anthropic-messages",
12103
12053
  provider: "vercel-ai-gateway",
12104
12054
  baseUrl: "https://ai-gateway.vercel.sh",
12105
- reasoning: false,
12106
- input: ["text"],
12055
+ reasoning: true,
12056
+ input: ["text", "image"],
12107
12057
  cost: {
12108
12058
  input: 0.62,
12109
12059
  output: 1.85,
@@ -12120,7 +12070,7 @@ export const MODELS = {
12120
12070
  provider: "vercel-ai-gateway",
12121
12071
  baseUrl: "https://ai-gateway.vercel.sh",
12122
12072
  reasoning: true,
12123
- input: ["text"],
12073
+ input: ["text", "image"],
12124
12074
  cost: {
12125
12075
  input: 0.14,
12126
12076
  output: 0.28,
@@ -12137,7 +12087,7 @@ export const MODELS = {
12137
12087
  provider: "vercel-ai-gateway",
12138
12088
  baseUrl: "https://ai-gateway.vercel.sh",
12139
12089
  reasoning: true,
12140
- input: ["text"],
12090
+ input: ["text", "image"],
12141
12091
  cost: {
12142
12092
  input: 0.435,
12143
12093
  output: 0.87,
@@ -12630,7 +12580,7 @@ export const MODELS = {
12630
12580
  provider: "vercel-ai-gateway",
12631
12581
  baseUrl: "https://ai-gateway.vercel.sh",
12632
12582
  reasoning: true,
12633
- input: ["text"],
12583
+ input: ["text", "image"],
12634
12584
  cost: {
12635
12585
  input: 0.6,
12636
12586
  output: 2.4,
@@ -12674,6 +12624,23 @@ export const MODELS = {
12674
12624
  contextWindow: 204800,
12675
12625
  maxTokens: 131100,
12676
12626
  },
12627
+ "minimax/minimax-m3": {
12628
+ id: "minimax/minimax-m3",
12629
+ name: "MiniMax M3",
12630
+ api: "anthropic-messages",
12631
+ provider: "vercel-ai-gateway",
12632
+ baseUrl: "https://ai-gateway.vercel.sh",
12633
+ reasoning: true,
12634
+ input: ["text", "image"],
12635
+ cost: {
12636
+ input: 0.3,
12637
+ output: 1.2,
12638
+ cacheRead: 0.06,
12639
+ cacheWrite: 0,
12640
+ },
12641
+ contextWindow: 1000000,
12642
+ maxTokens: 1000000,
12643
+ },
12677
12644
  "mistral/codestral": {
12678
12645
  id: "mistral/codestral",
12679
12646
  name: "Mistral Codestral",
@@ -12810,6 +12777,23 @@ export const MODELS = {
12810
12777
  contextWindow: 256000,
12811
12778
  maxTokens: 256000,
12812
12779
  },
12780
+ "mistral/mistral-nemo": {
12781
+ id: "mistral/mistral-nemo",
12782
+ name: "Mistral Nemo 12B",
12783
+ api: "anthropic-messages",
12784
+ provider: "vercel-ai-gateway",
12785
+ baseUrl: "https://ai-gateway.vercel.sh",
12786
+ reasoning: false,
12787
+ input: ["text"],
12788
+ cost: {
12789
+ input: 0.02,
12790
+ output: 0.04,
12791
+ cacheRead: 0,
12792
+ cacheWrite: 0,
12793
+ },
12794
+ contextWindow: 131072,
12795
+ maxTokens: 131072,
12796
+ },
12813
12797
  "mistral/mistral-small": {
12814
12798
  id: "mistral/mistral-small",
12815
12799
  name: "Mistral Small",
@@ -12963,6 +12947,23 @@ export const MODELS = {
12963
12947
  contextWindow: 262000,
12964
12948
  maxTokens: 262000,
12965
12949
  },
12950
+ "nvidia/nemotron-3-super-120b-a12b": {
12951
+ id: "nvidia/nemotron-3-super-120b-a12b",
12952
+ name: "NVIDIA Nemotron 3 Super 120B A12B",
12953
+ api: "anthropic-messages",
12954
+ provider: "vercel-ai-gateway",
12955
+ baseUrl: "https://ai-gateway.vercel.sh",
12956
+ reasoning: true,
12957
+ input: ["text"],
12958
+ cost: {
12959
+ input: 0.15,
12960
+ output: 0.65,
12961
+ cacheRead: 0,
12962
+ cacheWrite: 0,
12963
+ },
12964
+ contextWindow: 256000,
12965
+ maxTokens: 32000,
12966
+ },
12966
12967
  "nvidia/nemotron-nano-12b-v2-vl": {
12967
12968
  id: "nvidia/nemotron-nano-12b-v2-vl",
12968
12969
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -13140,7 +13141,7 @@ export const MODELS = {
13140
13141
  provider: "vercel-ai-gateway",
13141
13142
  baseUrl: "https://ai-gateway.vercel.sh",
13142
13143
  reasoning: true,
13143
- input: ["text"],
13144
+ input: ["text", "image"],
13144
13145
  cost: {
13145
13146
  input: 1.25,
13146
13147
  output: 10,
@@ -13490,6 +13491,23 @@ export const MODELS = {
13490
13491
  contextWindow: 1000000,
13491
13492
  maxTokens: 128000,
13492
13493
  },
13494
+ "openai/gpt-oss-120b": {
13495
+ id: "openai/gpt-oss-120b",
13496
+ name: "GPT OSS 120B",
13497
+ api: "anthropic-messages",
13498
+ provider: "vercel-ai-gateway",
13499
+ baseUrl: "https://ai-gateway.vercel.sh",
13500
+ reasoning: true,
13501
+ input: ["text"],
13502
+ cost: {
13503
+ input: 0.35,
13504
+ output: 0.75,
13505
+ cacheRead: 0.25,
13506
+ cacheWrite: 0,
13507
+ },
13508
+ contextWindow: 131072,
13509
+ maxTokens: 131000,
13510
+ },
13493
13511
  "openai/gpt-oss-20b": {
13494
13512
  id: "openai/gpt-oss-20b",
13495
13513
  name: "GPT OSS 20B",
@@ -13660,6 +13678,40 @@ export const MODELS = {
13660
13678
  contextWindow: 200000,
13661
13679
  maxTokens: 8000,
13662
13680
  },
13681
+ "stepfun/step-3.5-flash": {
13682
+ id: "stepfun/step-3.5-flash",
13683
+ name: "StepFun 3.5 Flash",
13684
+ api: "anthropic-messages",
13685
+ provider: "vercel-ai-gateway",
13686
+ baseUrl: "https://ai-gateway.vercel.sh",
13687
+ reasoning: true,
13688
+ input: ["text"],
13689
+ cost: {
13690
+ input: 0.09,
13691
+ output: 0.3,
13692
+ cacheRead: 0,
13693
+ cacheWrite: 0.02,
13694
+ },
13695
+ contextWindow: 262114,
13696
+ maxTokens: 262114,
13697
+ },
13698
+ "stepfun/step-3.7-flash": {
13699
+ id: "stepfun/step-3.7-flash",
13700
+ name: "Step 3.7 Flash",
13701
+ api: "anthropic-messages",
13702
+ provider: "vercel-ai-gateway",
13703
+ baseUrl: "https://ai-gateway.vercel.sh",
13704
+ reasoning: true,
13705
+ input: ["text", "image"],
13706
+ cost: {
13707
+ input: 0.19999999999999998,
13708
+ output: 1.15,
13709
+ cacheRead: 0.04,
13710
+ cacheWrite: 0,
13711
+ },
13712
+ contextWindow: 256000,
13713
+ maxTokens: 256000,
13714
+ },
13663
13715
  "xai/grok-4.1-fast-non-reasoning": {
13664
13716
  id: "xai/grok-4.1-fast-non-reasoning",
13665
13717
  name: "Grok 4.1 Fast Non-Reasoning",
@@ -13938,7 +13990,7 @@ export const MODELS = {
13938
13990
  api: "anthropic-messages",
13939
13991
  provider: "vercel-ai-gateway",
13940
13992
  baseUrl: "https://ai-gateway.vercel.sh",
13941
- reasoning: false,
13993
+ reasoning: true,
13942
13994
  input: ["text", "image"],
13943
13995
  cost: {
13944
13996
  input: 0.6,
@@ -14092,7 +14144,7 @@ export const MODELS = {
14092
14144
  provider: "vercel-ai-gateway",
14093
14145
  baseUrl: "https://ai-gateway.vercel.sh",
14094
14146
  reasoning: true,
14095
- input: ["text"],
14147
+ input: ["text", "image"],
14096
14148
  cost: {
14097
14149
  input: 1.4,
14098
14150
  output: 4.4,