@dreb/ai 2.21.0 → 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 +370 -227
  2. package/dist/models.generated.d.ts.map +1 -1
  3. package/dist/models.generated.js +564 -427
  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
@@ -240,6 +240,23 @@ export const MODELS = {
240
240
  contextWindow: 1000000,
241
241
  maxTokens: 128000,
242
242
  },
243
+ "au.anthropic.claude-opus-4-8": {
244
+ id: "au.anthropic.claude-opus-4-8",
245
+ name: "Claude Opus 4.8 (AU)",
246
+ api: "bedrock-converse-stream",
247
+ provider: "amazon-bedrock",
248
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
249
+ reasoning: true,
250
+ input: ["text", "image"],
251
+ cost: {
252
+ input: 5,
253
+ output: 25,
254
+ cacheRead: 0.5,
255
+ cacheWrite: 6.25,
256
+ },
257
+ contextWindow: 1000000,
258
+ maxTokens: 128000,
259
+ },
243
260
  "au.anthropic.claude-sonnet-4-5-20250929-v1:0": {
244
261
  id: "au.anthropic.claude-sonnet-4-5-20250929-v1:0",
245
262
  name: "Claude Sonnet 4.5 (AU)",
@@ -368,8 +385,8 @@ export const MODELS = {
368
385
  reasoning: true,
369
386
  input: ["text", "image"],
370
387
  cost: {
371
- input: 5,
372
- output: 25,
388
+ input: 5.5,
389
+ output: 27.5,
373
390
  cacheRead: 0.5,
374
391
  cacheWrite: 6.25,
375
392
  },
@@ -385,10 +402,27 @@ export const MODELS = {
385
402
  reasoning: true,
386
403
  input: ["text", "image"],
387
404
  cost: {
388
- input: 5,
389
- output: 25,
390
- cacheRead: 0.5,
391
- cacheWrite: 6.25,
405
+ input: 5.5,
406
+ output: 27.5,
407
+ cacheRead: 0.55,
408
+ cacheWrite: 6.875,
409
+ },
410
+ contextWindow: 1000000,
411
+ maxTokens: 128000,
412
+ },
413
+ "eu.anthropic.claude-opus-4-8": {
414
+ id: "eu.anthropic.claude-opus-4-8",
415
+ name: "Claude Opus 4.8 (EU)",
416
+ api: "bedrock-converse-stream",
417
+ provider: "amazon-bedrock",
418
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
419
+ reasoning: true,
420
+ input: ["text", "image"],
421
+ cost: {
422
+ input: 5.5,
423
+ output: 27.5,
424
+ cacheRead: 0.55,
425
+ cacheWrite: 6.875,
392
426
  },
393
427
  contextWindow: 1000000,
394
428
  maxTokens: 128000,
@@ -402,10 +436,10 @@ export const MODELS = {
402
436
  reasoning: true,
403
437
  input: ["text", "image"],
404
438
  cost: {
405
- input: 3,
406
- output: 15,
407
- cacheRead: 0.3,
408
- cacheWrite: 3.75,
439
+ input: 3.3,
440
+ output: 16.5,
441
+ cacheRead: 0.33,
442
+ cacheWrite: 4.125,
409
443
  },
410
444
  contextWindow: 200000,
411
445
  maxTokens: 64000,
@@ -419,10 +453,10 @@ export const MODELS = {
419
453
  reasoning: true,
420
454
  input: ["text", "image"],
421
455
  cost: {
422
- input: 3,
423
- output: 15,
424
- cacheRead: 0.3,
425
- cacheWrite: 3.75,
456
+ input: 3.3,
457
+ output: 16.5,
458
+ cacheRead: 0.33,
459
+ cacheWrite: 4.125,
426
460
  },
427
461
  contextWindow: 1000000,
428
462
  maxTokens: 64000,
@@ -495,6 +529,23 @@ export const MODELS = {
495
529
  contextWindow: 1000000,
496
530
  maxTokens: 128000,
497
531
  },
532
+ "global.anthropic.claude-opus-4-8": {
533
+ id: "global.anthropic.claude-opus-4-8",
534
+ name: "Claude Opus 4.8 (Global)",
535
+ api: "bedrock-converse-stream",
536
+ provider: "amazon-bedrock",
537
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
538
+ reasoning: true,
539
+ input: ["text", "image"],
540
+ cost: {
541
+ input: 5,
542
+ output: 25,
543
+ cacheRead: 0.5,
544
+ cacheWrite: 6.25,
545
+ },
546
+ contextWindow: 1000000,
547
+ maxTokens: 128000,
548
+ },
498
549
  "global.anthropic.claude-sonnet-4-5-20250929-v1:0": {
499
550
  id: "global.anthropic.claude-sonnet-4-5-20250929-v1:0",
500
551
  name: "Claude Sonnet 4.5 (Global)",
@@ -580,6 +631,23 @@ export const MODELS = {
580
631
  contextWindow: 1000000,
581
632
  maxTokens: 128000,
582
633
  },
634
+ "jp.anthropic.claude-opus-4-8": {
635
+ id: "jp.anthropic.claude-opus-4-8",
636
+ name: "Claude Opus 4.8 (JP)",
637
+ api: "bedrock-converse-stream",
638
+ provider: "amazon-bedrock",
639
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
640
+ reasoning: true,
641
+ input: ["text", "image"],
642
+ cost: {
643
+ input: 5,
644
+ output: 25,
645
+ cacheRead: 0.5,
646
+ cacheWrite: 6.25,
647
+ },
648
+ contextWindow: 1000000,
649
+ maxTokens: 128000,
650
+ },
583
651
  "jp.anthropic.claude-sonnet-4-5-20250929-v1:0": {
584
652
  id: "jp.anthropic.claude-sonnet-4-5-20250929-v1:0",
585
653
  name: "Claude Sonnet 4.5 (JP)",
@@ -1277,6 +1345,23 @@ export const MODELS = {
1277
1345
  contextWindow: 1000000,
1278
1346
  maxTokens: 128000,
1279
1347
  },
1348
+ "us.anthropic.claude-opus-4-8": {
1349
+ id: "us.anthropic.claude-opus-4-8",
1350
+ name: "Claude Opus 4.8 (US)",
1351
+ api: "bedrock-converse-stream",
1352
+ provider: "amazon-bedrock",
1353
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1354
+ reasoning: true,
1355
+ input: ["text", "image"],
1356
+ cost: {
1357
+ input: 5,
1358
+ output: 25,
1359
+ cacheRead: 0.5,
1360
+ cacheWrite: 6.25,
1361
+ },
1362
+ contextWindow: 1000000,
1363
+ maxTokens: 128000,
1364
+ },
1280
1365
  "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
1281
1366
  id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
1282
1367
  name: "Claude Sonnet 4.5 (US)",
@@ -2613,7 +2698,7 @@ export const MODELS = {
2613
2698
  "github-copilot": {
2614
2699
  "claude-haiku-4.5": {
2615
2700
  id: "claude-haiku-4.5",
2616
- name: "Claude Haiku 4.5",
2701
+ name: "Claude Haiku 4.5 (latest)",
2617
2702
  api: "anthropic-messages",
2618
2703
  provider: "github-copilot",
2619
2704
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2621,17 +2706,17 @@ export const MODELS = {
2621
2706
  reasoning: true,
2622
2707
  input: ["text", "image"],
2623
2708
  cost: {
2624
- input: 0,
2625
- output: 0,
2626
- cacheRead: 0,
2627
- cacheWrite: 0,
2709
+ input: 1,
2710
+ output: 5,
2711
+ cacheRead: 0.1,
2712
+ cacheWrite: 1.25,
2628
2713
  },
2629
- contextWindow: 144000,
2630
- maxTokens: 32000,
2714
+ contextWindow: 200000,
2715
+ maxTokens: 64000,
2631
2716
  },
2632
2717
  "claude-opus-4.5": {
2633
2718
  id: "claude-opus-4.5",
2634
- name: "Claude Opus 4.5",
2719
+ name: "Claude Opus 4.5 (latest)",
2635
2720
  api: "anthropic-messages",
2636
2721
  provider: "github-copilot",
2637
2722
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2639,13 +2724,13 @@ export const MODELS = {
2639
2724
  reasoning: true,
2640
2725
  input: ["text", "image"],
2641
2726
  cost: {
2642
- input: 0,
2643
- output: 0,
2644
- cacheRead: 0,
2645
- cacheWrite: 0,
2727
+ input: 5,
2728
+ output: 25,
2729
+ cacheRead: 0.5,
2730
+ cacheWrite: 6.25,
2646
2731
  },
2647
- contextWindow: 160000,
2648
- maxTokens: 32000,
2732
+ contextWindow: 200000,
2733
+ maxTokens: 64000,
2649
2734
  },
2650
2735
  "claude-opus-4.6": {
2651
2736
  id: "claude-opus-4.6",
@@ -2657,13 +2742,13 @@ export const MODELS = {
2657
2742
  reasoning: true,
2658
2743
  input: ["text", "image"],
2659
2744
  cost: {
2660
- input: 0,
2661
- output: 0,
2662
- cacheRead: 0,
2663
- cacheWrite: 0,
2745
+ input: 5,
2746
+ output: 25,
2747
+ cacheRead: 0.5,
2748
+ cacheWrite: 6.25,
2664
2749
  },
2665
2750
  contextWindow: 1000000,
2666
- maxTokens: 64000,
2751
+ maxTokens: 128000,
2667
2752
  },
2668
2753
  "claude-opus-4.7": {
2669
2754
  id: "claude-opus-4.7",
@@ -2675,17 +2760,53 @@ export const MODELS = {
2675
2760
  reasoning: true,
2676
2761
  input: ["text", "image"],
2677
2762
  cost: {
2678
- input: 0,
2679
- output: 0,
2680
- cacheRead: 0,
2681
- 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,
2785
+ },
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,
2682
2803
  },
2683
- contextWindow: 144000,
2804
+ contextWindow: 200000,
2684
2805
  maxTokens: 64000,
2685
2806
  },
2686
2807
  "claude-sonnet-4.5": {
2687
2808
  id: "claude-sonnet-4.5",
2688
- name: "Claude Sonnet 4.5",
2809
+ name: "Claude Sonnet 4.5 (latest)",
2689
2810
  api: "anthropic-messages",
2690
2811
  provider: "github-copilot",
2691
2812
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2693,13 +2814,13 @@ export const MODELS = {
2693
2814
  reasoning: true,
2694
2815
  input: ["text", "image"],
2695
2816
  cost: {
2696
- input: 0,
2697
- output: 0,
2698
- cacheRead: 0,
2699
- cacheWrite: 0,
2817
+ input: 3,
2818
+ output: 15,
2819
+ cacheRead: 0.3,
2820
+ cacheWrite: 3.75,
2700
2821
  },
2701
- contextWindow: 144000,
2702
- maxTokens: 32000,
2822
+ contextWindow: 200000,
2823
+ maxTokens: 64000,
2703
2824
  },
2704
2825
  "claude-sonnet-4.6": {
2705
2826
  id: "claude-sonnet-4.6",
@@ -2711,13 +2832,13 @@ export const MODELS = {
2711
2832
  reasoning: true,
2712
2833
  input: ["text", "image"],
2713
2834
  cost: {
2714
- input: 0,
2715
- output: 0,
2716
- cacheRead: 0,
2717
- cacheWrite: 0,
2835
+ input: 3,
2836
+ output: 15,
2837
+ cacheRead: 0.3,
2838
+ cacheWrite: 3.75,
2718
2839
  },
2719
2840
  contextWindow: 1000000,
2720
- maxTokens: 32000,
2841
+ maxTokens: 64000,
2721
2842
  },
2722
2843
  "gemini-2.5-pro": {
2723
2844
  id: "gemini-2.5-pro",
@@ -2727,20 +2848,20 @@ export const MODELS = {
2727
2848
  baseUrl: "https://api.individual.githubcopilot.com",
2728
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" },
2729
2850
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
2730
- reasoning: false,
2851
+ reasoning: true,
2731
2852
  input: ["text", "image"],
2732
2853
  cost: {
2733
- input: 0,
2734
- output: 0,
2735
- cacheRead: 0,
2854
+ input: 1.25,
2855
+ output: 10,
2856
+ cacheRead: 0.125,
2736
2857
  cacheWrite: 0,
2737
2858
  },
2738
- contextWindow: 128000,
2739
- maxTokens: 64000,
2859
+ contextWindow: 1048576,
2860
+ maxTokens: 65536,
2740
2861
  },
2741
2862
  "gemini-3-flash-preview": {
2742
2863
  id: "gemini-3-flash-preview",
2743
- name: "Gemini 3 Flash",
2864
+ name: "Gemini 3 Flash Preview",
2744
2865
  api: "openai-completions",
2745
2866
  provider: "github-copilot",
2746
2867
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2749,13 +2870,13 @@ export const MODELS = {
2749
2870
  reasoning: true,
2750
2871
  input: ["text", "image"],
2751
2872
  cost: {
2752
- input: 0,
2753
- output: 0,
2754
- cacheRead: 0,
2873
+ input: 0.5,
2874
+ output: 3,
2875
+ cacheRead: 0.05,
2755
2876
  cacheWrite: 0,
2756
2877
  },
2757
- contextWindow: 128000,
2758
- maxTokens: 64000,
2878
+ contextWindow: 1048576,
2879
+ maxTokens: 65536,
2759
2880
  },
2760
2881
  "gemini-3.1-pro-preview": {
2761
2882
  id: "gemini-3.1-pro-preview",
@@ -2768,13 +2889,13 @@ export const MODELS = {
2768
2889
  reasoning: true,
2769
2890
  input: ["text", "image"],
2770
2891
  cost: {
2771
- input: 0,
2772
- output: 0,
2773
- cacheRead: 0,
2892
+ input: 2,
2893
+ output: 12,
2894
+ cacheRead: 0.2,
2774
2895
  cacheWrite: 0,
2775
2896
  },
2776
- contextWindow: 128000,
2777
- maxTokens: 64000,
2897
+ contextWindow: 1048576,
2898
+ maxTokens: 65536,
2778
2899
  },
2779
2900
  "gemini-3.5-flash": {
2780
2901
  id: "gemini-3.5-flash",
@@ -2787,13 +2908,13 @@ export const MODELS = {
2787
2908
  reasoning: true,
2788
2909
  input: ["text", "image"],
2789
2910
  cost: {
2790
- input: 0,
2791
- output: 0,
2792
- cacheRead: 0,
2911
+ input: 1.5,
2912
+ output: 9,
2913
+ cacheRead: 0.15,
2793
2914
  cacheWrite: 0,
2794
2915
  },
2795
- contextWindow: 128000,
2796
- maxTokens: 64000,
2916
+ contextWindow: 1048576,
2917
+ maxTokens: 65536,
2797
2918
  },
2798
2919
  "gpt-4.1": {
2799
2920
  id: "gpt-4.1",
@@ -2806,36 +2927,17 @@ export const MODELS = {
2806
2927
  reasoning: false,
2807
2928
  input: ["text", "image"],
2808
2929
  cost: {
2809
- input: 0,
2810
- output: 0,
2811
- cacheRead: 0,
2812
- cacheWrite: 0,
2813
- },
2814
- contextWindow: 128000,
2815
- maxTokens: 16384,
2816
- },
2817
- "gpt-4o": {
2818
- id: "gpt-4o",
2819
- name: "GPT-4o",
2820
- api: "openai-completions",
2821
- provider: "github-copilot",
2822
- baseUrl: "https://api.individual.githubcopilot.com",
2823
- 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" },
2824
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
2825
- reasoning: false,
2826
- input: ["text", "image"],
2827
- cost: {
2828
- input: 0,
2829
- output: 0,
2830
- cacheRead: 0,
2930
+ input: 2,
2931
+ output: 8,
2932
+ cacheRead: 0.5,
2831
2933
  cacheWrite: 0,
2832
2934
  },
2833
- contextWindow: 128000,
2834
- maxTokens: 4096,
2935
+ contextWindow: 1047576,
2936
+ maxTokens: 32768,
2835
2937
  },
2836
2938
  "gpt-5-mini": {
2837
2939
  id: "gpt-5-mini",
2838
- name: "GPT-5-mini",
2940
+ name: "GPT-5 Mini",
2839
2941
  api: "openai-responses",
2840
2942
  provider: "github-copilot",
2841
2943
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2843,13 +2945,13 @@ export const MODELS = {
2843
2945
  reasoning: true,
2844
2946
  input: ["text", "image"],
2845
2947
  cost: {
2846
- input: 0,
2847
- output: 0,
2848
- cacheRead: 0,
2948
+ input: 0.25,
2949
+ output: 2,
2950
+ cacheRead: 0.025,
2849
2951
  cacheWrite: 0,
2850
2952
  },
2851
- contextWindow: 264000,
2852
- maxTokens: 64000,
2953
+ contextWindow: 400000,
2954
+ maxTokens: 128000,
2853
2955
  },
2854
2956
  "gpt-5.2": {
2855
2957
  id: "gpt-5.2",
@@ -2861,17 +2963,17 @@ export const MODELS = {
2861
2963
  reasoning: true,
2862
2964
  input: ["text", "image"],
2863
2965
  cost: {
2864
- input: 0,
2865
- output: 0,
2866
- cacheRead: 0,
2966
+ input: 1.75,
2967
+ output: 14,
2968
+ cacheRead: 0.175,
2867
2969
  cacheWrite: 0,
2868
2970
  },
2869
- contextWindow: 264000,
2870
- maxTokens: 64000,
2971
+ contextWindow: 400000,
2972
+ maxTokens: 128000,
2871
2973
  },
2872
2974
  "gpt-5.2-codex": {
2873
2975
  id: "gpt-5.2-codex",
2874
- name: "GPT-5.2-Codex",
2976
+ name: "GPT-5.2 Codex",
2875
2977
  api: "openai-responses",
2876
2978
  provider: "github-copilot",
2877
2979
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2879,9 +2981,9 @@ export const MODELS = {
2879
2981
  reasoning: true,
2880
2982
  input: ["text", "image"],
2881
2983
  cost: {
2882
- input: 0,
2883
- output: 0,
2884
- cacheRead: 0,
2984
+ input: 1.75,
2985
+ output: 14,
2986
+ cacheRead: 0.175,
2885
2987
  cacheWrite: 0,
2886
2988
  },
2887
2989
  contextWindow: 400000,
@@ -2889,7 +2991,7 @@ export const MODELS = {
2889
2991
  },
2890
2992
  "gpt-5.3-codex": {
2891
2993
  id: "gpt-5.3-codex",
2892
- name: "GPT-5.3-Codex",
2994
+ name: "GPT-5.3 Codex",
2893
2995
  api: "openai-responses",
2894
2996
  provider: "github-copilot",
2895
2997
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2897,9 +2999,9 @@ export const MODELS = {
2897
2999
  reasoning: true,
2898
3000
  input: ["text", "image"],
2899
3001
  cost: {
2900
- input: 0,
2901
- output: 0,
2902
- cacheRead: 0,
3002
+ input: 1.75,
3003
+ output: 14,
3004
+ cacheRead: 0.175,
2903
3005
  cacheWrite: 0,
2904
3006
  },
2905
3007
  contextWindow: 400000,
@@ -2915,17 +3017,17 @@ export const MODELS = {
2915
3017
  reasoning: true,
2916
3018
  input: ["text", "image"],
2917
3019
  cost: {
2918
- input: 0,
2919
- output: 0,
2920
- cacheRead: 0,
3020
+ input: 2.5,
3021
+ output: 15,
3022
+ cacheRead: 0.25,
2921
3023
  cacheWrite: 0,
2922
3024
  },
2923
- contextWindow: 400000,
3025
+ contextWindow: 1050000,
2924
3026
  maxTokens: 128000,
2925
3027
  },
2926
3028
  "gpt-5.4-mini": {
2927
3029
  id: "gpt-5.4-mini",
2928
- name: "GPT-5.4 Mini",
3030
+ name: "GPT-5.4 mini",
2929
3031
  api: "openai-responses",
2930
3032
  provider: "github-copilot",
2931
3033
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -2933,9 +3035,27 @@ export const MODELS = {
2933
3035
  reasoning: true,
2934
3036
  input: ["text", "image"],
2935
3037
  cost: {
2936
- input: 0,
2937
- output: 0,
2938
- 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,
2939
3059
  cacheWrite: 0,
2940
3060
  },
2941
3061
  contextWindow: 400000,
@@ -2951,32 +3071,32 @@ export const MODELS = {
2951
3071
  reasoning: true,
2952
3072
  input: ["text", "image"],
2953
3073
  cost: {
2954
- input: 0,
2955
- output: 0,
2956
- cacheRead: 0,
3074
+ input: 5,
3075
+ output: 30,
3076
+ cacheRead: 0.5,
2957
3077
  cacheWrite: 0,
2958
3078
  },
2959
- contextWindow: 400000,
3079
+ contextWindow: 1050000,
2960
3080
  maxTokens: 128000,
2961
3081
  },
2962
- "grok-code-fast-1": {
2963
- id: "grok-code-fast-1",
2964
- name: "Grok Code Fast 1",
3082
+ "raptor-mini": {
3083
+ id: "raptor-mini",
3084
+ name: "Raptor mini",
2965
3085
  api: "openai-completions",
2966
3086
  provider: "github-copilot",
2967
3087
  baseUrl: "https://api.individual.githubcopilot.com",
2968
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" },
2969
3089
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
2970
3090
  reasoning: true,
2971
- input: ["text"],
3091
+ input: ["text", "image"],
2972
3092
  cost: {
2973
- input: 0,
2974
- output: 0,
2975
- cacheRead: 0,
3093
+ input: 0.25,
3094
+ output: 2,
3095
+ cacheRead: 0.025,
2976
3096
  cacheWrite: 0,
2977
3097
  },
2978
- contextWindow: 128000,
2979
- maxTokens: 64000,
3098
+ contextWindow: 400000,
3099
+ maxTokens: 128000,
2980
3100
  },
2981
3101
  },
2982
3102
  "google": {
@@ -6055,6 +6175,23 @@ export const MODELS = {
6055
6175
  contextWindow: 1000000,
6056
6176
  maxTokens: 64000,
6057
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
+ },
6058
6195
  "deepseek-v4-flash-free": {
6059
6196
  id: "deepseek-v4-flash-free",
6060
6197
  name: "DeepSeek V4 Flash Free",
@@ -6494,8 +6631,8 @@ export const MODELS = {
6494
6631
  cacheRead: 0,
6495
6632
  cacheWrite: 0,
6496
6633
  },
6497
- contextWindow: 1000000,
6498
- maxTokens: 128000,
6634
+ contextWindow: 200000,
6635
+ maxTokens: 32000,
6499
6636
  },
6500
6637
  "minimax-m2.5": {
6501
6638
  id: "minimax-m2.5",
@@ -6531,6 +6668,23 @@ export const MODELS = {
6531
6668
  contextWindow: 204800,
6532
6669
  maxTokens: 131072,
6533
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
+ },
6534
6688
  "nemotron-3-super-free": {
6535
6689
  id: "nemotron-3-super-free",
6536
6690
  name: "Nemotron 3 Super Free",
@@ -6754,22 +6908,22 @@ export const MODELS = {
6754
6908
  contextWindow: 204800,
6755
6909
  maxTokens: 131072,
6756
6910
  },
6757
- "qwen3.5-plus": {
6758
- id: "qwen3.5-plus",
6759
- name: "Qwen3.5 Plus",
6911
+ "minimax-m3": {
6912
+ id: "minimax-m3",
6913
+ name: "MiniMax M3",
6760
6914
  api: "anthropic-messages",
6761
6915
  provider: "opencode-go",
6762
6916
  baseUrl: "https://opencode.ai/zen/go",
6763
6917
  reasoning: true,
6764
6918
  input: ["text", "image"],
6765
6919
  cost: {
6766
- input: 0.2,
6767
- output: 1.2,
6768
- cacheRead: 0.02,
6769
- cacheWrite: 0.25,
6920
+ input: 0.6,
6921
+ output: 2.4,
6922
+ cacheRead: 0.12,
6923
+ cacheWrite: 0,
6770
6924
  },
6771
- contextWindow: 262144,
6772
- maxTokens: 65536,
6925
+ contextWindow: 512000,
6926
+ maxTokens: 131072,
6773
6927
  },
6774
6928
  "qwen3.6-plus": {
6775
6929
  id: "qwen3.6-plus",
@@ -7232,23 +7386,6 @@ export const MODELS = {
7232
7386
  contextWindow: 2000000,
7233
7387
  maxTokens: 30000,
7234
7388
  },
7235
- "baidu/ernie-4.5-21b-a3b": {
7236
- id: "baidu/ernie-4.5-21b-a3b",
7237
- name: "Baidu: ERNIE 4.5 21B A3B",
7238
- api: "openai-completions",
7239
- provider: "openrouter",
7240
- baseUrl: "https://openrouter.ai/api/v1",
7241
- reasoning: false,
7242
- input: ["text"],
7243
- cost: {
7244
- input: 0.07,
7245
- output: 0.28,
7246
- cacheRead: 0,
7247
- cacheWrite: 0,
7248
- },
7249
- contextWindow: 131072,
7250
- maxTokens: 8000,
7251
- },
7252
7389
  "baidu/ernie-4.5-vl-28b-a3b": {
7253
7390
  id: "baidu/ernie-4.5-vl-28b-a3b",
7254
7391
  name: "Baidu: ERNIE 4.5 VL 28B A3B",
@@ -7377,8 +7514,8 @@ export const MODELS = {
7377
7514
  reasoning: false,
7378
7515
  input: ["text"],
7379
7516
  cost: {
7380
- input: 0.2288,
7381
- output: 0.9144,
7517
+ input: 0.20020000000000002,
7518
+ output: 0.8000999999999999,
7382
7519
  cacheRead: 0,
7383
7520
  cacheWrite: 0,
7384
7521
  },
@@ -7479,13 +7616,13 @@ export const MODELS = {
7479
7616
  reasoning: true,
7480
7617
  input: ["text"],
7481
7618
  cost: {
7482
- input: 0.252,
7483
- output: 0.378,
7484
- cacheRead: 0.0252,
7619
+ input: 0.2288,
7620
+ output: 0.3432,
7621
+ cacheRead: 0,
7485
7622
  cacheWrite: 0,
7486
7623
  },
7487
7624
  contextWindow: 131072,
7488
- maxTokens: 65536,
7625
+ maxTokens: 64000,
7489
7626
  },
7490
7627
  "deepseek/deepseek-v3.2-exp": {
7491
7628
  id: "deepseek/deepseek-v3.2-exp",
@@ -7513,30 +7650,13 @@ export const MODELS = {
7513
7650
  reasoning: true,
7514
7651
  input: ["text"],
7515
7652
  cost: {
7516
- input: 0.09999999999999999,
7517
- output: 0.19999999999999998,
7518
- cacheRead: 0.02,
7519
- cacheWrite: 0,
7520
- },
7521
- contextWindow: 1048576,
7522
- maxTokens: 16384,
7523
- },
7524
- "deepseek/deepseek-v4-flash:free": {
7525
- id: "deepseek/deepseek-v4-flash:free",
7526
- name: "DeepSeek: DeepSeek V4 Flash (free)",
7527
- api: "openai-completions",
7528
- provider: "openrouter",
7529
- baseUrl: "https://openrouter.ai/api/v1",
7530
- reasoning: true,
7531
- input: ["text"],
7532
- cost: {
7533
- input: 0,
7534
- output: 0,
7535
- cacheRead: 0,
7653
+ input: 0.0983,
7654
+ output: 0.1966,
7655
+ cacheRead: 0.019700000000000002,
7536
7656
  cacheWrite: 0,
7537
7657
  },
7538
7658
  contextWindow: 1048576,
7539
- maxTokens: 384000,
7659
+ maxTokens: 131072,
7540
7660
  },
7541
7661
  "deepseek/deepseek-v4-pro": {
7542
7662
  id: "deepseek/deepseek-v4-pro",
@@ -7572,40 +7692,6 @@ export const MODELS = {
7572
7692
  contextWindow: 32768,
7573
7693
  maxTokens: 4096,
7574
7694
  },
7575
- "google/gemini-2.0-flash-001": {
7576
- id: "google/gemini-2.0-flash-001",
7577
- name: "Google: Gemini 2.0 Flash",
7578
- api: "openai-completions",
7579
- provider: "openrouter",
7580
- baseUrl: "https://openrouter.ai/api/v1",
7581
- reasoning: false,
7582
- input: ["text", "image"],
7583
- cost: {
7584
- input: 0.09999999999999999,
7585
- output: 0.39999999999999997,
7586
- cacheRead: 0.024999999999999998,
7587
- cacheWrite: 0.08333333333333334,
7588
- },
7589
- contextWindow: 1000000,
7590
- maxTokens: 8192,
7591
- },
7592
- "google/gemini-2.0-flash-lite-001": {
7593
- id: "google/gemini-2.0-flash-lite-001",
7594
- name: "Google: Gemini 2.0 Flash Lite",
7595
- api: "openai-completions",
7596
- provider: "openrouter",
7597
- baseUrl: "https://openrouter.ai/api/v1",
7598
- reasoning: false,
7599
- input: ["text", "image"],
7600
- cost: {
7601
- input: 0.075,
7602
- output: 0.3,
7603
- cacheRead: 0,
7604
- cacheWrite: 0,
7605
- },
7606
- contextWindow: 1048576,
7607
- maxTokens: 8192,
7608
- },
7609
7695
  "google/gemini-2.5-flash": {
7610
7696
  id: "google/gemini-2.5-flash",
7611
7697
  name: "Google: Gemini 2.5 Flash",
@@ -7989,9 +8075,9 @@ export const MODELS = {
7989
8075
  reasoning: true,
7990
8076
  input: ["text"],
7991
8077
  cost: {
7992
- input: 0.075,
7993
- output: 0.625,
7994
- cacheRead: 0.015,
8078
+ input: 0.3,
8079
+ output: 2.5,
8080
+ cacheRead: 0.06,
7995
8081
  cacheWrite: 0,
7996
8082
  },
7997
8083
  contextWindow: 262144,
@@ -8082,6 +8168,23 @@ export const MODELS = {
8082
8168
  contextWindow: 131072,
8083
8169
  maxTokens: 4096,
8084
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
+ },
8085
8188
  "meta-llama/llama-4-scout": {
8086
8189
  id: "meta-llama/llama-4-scout",
8087
8190
  name: "Meta: Llama 4 Scout",
@@ -8167,39 +8270,39 @@ export const MODELS = {
8167
8270
  contextWindow: 204800,
8168
8271
  maxTokens: 196608,
8169
8272
  },
8170
- "minimax/minimax-m2.5:free": {
8171
- id: "minimax/minimax-m2.5:free",
8172
- name: "MiniMax: MiniMax M2.5 (free)",
8273
+ "minimax/minimax-m2.7": {
8274
+ id: "minimax/minimax-m2.7",
8275
+ name: "MiniMax: MiniMax M2.7",
8173
8276
  api: "openai-completions",
8174
8277
  provider: "openrouter",
8175
8278
  baseUrl: "https://openrouter.ai/api/v1",
8176
8279
  reasoning: true,
8177
8280
  input: ["text"],
8178
8281
  cost: {
8179
- input: 0,
8180
- output: 0,
8282
+ input: 0.27899999999999997,
8283
+ output: 1.2,
8181
8284
  cacheRead: 0,
8182
8285
  cacheWrite: 0,
8183
8286
  },
8184
- contextWindow: 262144,
8185
- maxTokens: 8192,
8287
+ contextWindow: 204800,
8288
+ maxTokens: 131072,
8186
8289
  },
8187
- "minimax/minimax-m2.7": {
8188
- id: "minimax/minimax-m2.7",
8189
- name: "MiniMax: MiniMax M2.7",
8290
+ "minimax/minimax-m3": {
8291
+ id: "minimax/minimax-m3",
8292
+ name: "MiniMax: MiniMax M3",
8190
8293
  api: "openai-completions",
8191
8294
  provider: "openrouter",
8192
8295
  baseUrl: "https://openrouter.ai/api/v1",
8193
8296
  reasoning: true,
8194
- input: ["text"],
8297
+ input: ["text", "image"],
8195
8298
  cost: {
8196
- input: 0.27899999999999997,
8299
+ input: 0.3,
8197
8300
  output: 1.2,
8198
- cacheRead: 0,
8301
+ cacheRead: 0.06,
8199
8302
  cacheWrite: 0,
8200
8303
  },
8201
- contextWindow: 204800,
8202
- maxTokens: 131072,
8304
+ contextWindow: 1048576,
8305
+ maxTokens: 512000,
8203
8306
  },
8204
8307
  "mistralai/codestral-2508": {
8205
8308
  id: "mistralai/codestral-2508",
@@ -8235,40 +8338,6 @@ export const MODELS = {
8235
8338
  contextWindow: 262144,
8236
8339
  maxTokens: 4096,
8237
8340
  },
8238
- "mistralai/devstral-medium": {
8239
- id: "mistralai/devstral-medium",
8240
- name: "Mistral: Devstral Medium",
8241
- api: "openai-completions",
8242
- provider: "openrouter",
8243
- baseUrl: "https://openrouter.ai/api/v1",
8244
- reasoning: false,
8245
- input: ["text"],
8246
- cost: {
8247
- input: 0.39999999999999997,
8248
- output: 2,
8249
- cacheRead: 0.04,
8250
- cacheWrite: 0,
8251
- },
8252
- contextWindow: 131072,
8253
- maxTokens: 4096,
8254
- },
8255
- "mistralai/devstral-small": {
8256
- id: "mistralai/devstral-small",
8257
- name: "Mistral: Devstral Small 1.1",
8258
- api: "openai-completions",
8259
- provider: "openrouter",
8260
- baseUrl: "https://openrouter.ai/api/v1",
8261
- reasoning: false,
8262
- input: ["text"],
8263
- cost: {
8264
- input: 0.09999999999999999,
8265
- output: 0.3,
8266
- cacheRead: 0.01,
8267
- cacheWrite: 0,
8268
- },
8269
- contextWindow: 131072,
8270
- maxTokens: 4096,
8271
- },
8272
8341
  "mistralai/ministral-14b-2512": {
8273
8342
  id: "mistralai/ministral-14b-2512",
8274
8343
  name: "Mistral: Ministral 3 14B 2512",
@@ -8354,23 +8423,6 @@ export const MODELS = {
8354
8423
  contextWindow: 131072,
8355
8424
  maxTokens: 4096,
8356
8425
  },
8357
- "mistralai/mistral-large-2411": {
8358
- id: "mistralai/mistral-large-2411",
8359
- name: "Mistral Large 2411",
8360
- api: "openai-completions",
8361
- provider: "openrouter",
8362
- baseUrl: "https://openrouter.ai/api/v1",
8363
- reasoning: false,
8364
- input: ["text"],
8365
- cost: {
8366
- input: 2,
8367
- output: 6,
8368
- cacheRead: 0.19999999999999998,
8369
- cacheWrite: 0,
8370
- },
8371
- contextWindow: 131072,
8372
- maxTokens: 4096,
8373
- },
8374
8426
  "mistralai/mistral-large-2512": {
8375
8427
  id: "mistralai/mistral-large-2512",
8376
8428
  name: "Mistral: Mistral Large 3 2512",
@@ -8524,23 +8576,6 @@ export const MODELS = {
8524
8576
  contextWindow: 65536,
8525
8577
  maxTokens: 4096,
8526
8578
  },
8527
- "mistralai/pixtral-large-2411": {
8528
- id: "mistralai/pixtral-large-2411",
8529
- name: "Mistral: Pixtral Large 2411",
8530
- api: "openai-completions",
8531
- provider: "openrouter",
8532
- baseUrl: "https://openrouter.ai/api/v1",
8533
- reasoning: false,
8534
- input: ["text", "image"],
8535
- cost: {
8536
- input: 2,
8537
- output: 6,
8538
- cacheRead: 0.19999999999999998,
8539
- cacheWrite: 0,
8540
- },
8541
- contextWindow: 131072,
8542
- maxTokens: 4096,
8543
- },
8544
8579
  "mistralai/voxtral-small-24b-2507": {
8545
8580
  id: "mistralai/voxtral-small-24b-2507",
8546
8581
  name: "Mistral: Voxtral Small 24B 2507",
@@ -8635,13 +8670,13 @@ export const MODELS = {
8635
8670
  reasoning: true,
8636
8671
  input: ["text", "image"],
8637
8672
  cost: {
8638
- input: 0.73,
8639
- output: 3.49,
8640
- cacheRead: 0.25,
8673
+ input: 0.684,
8674
+ output: 3.42,
8675
+ cacheRead: 0.144,
8641
8676
  cacheWrite: 0,
8642
8677
  },
8643
8678
  contextWindow: 262144,
8644
- maxTokens: 262142,
8679
+ maxTokens: 262144,
8645
8680
  },
8646
8681
  "moonshotai/kimi-k2.6:free": {
8647
8682
  id: "moonshotai/kimi-k2.6:free",
@@ -9085,23 +9120,6 @@ export const MODELS = {
9085
9120
  contextWindow: 128000,
9086
9121
  maxTokens: 16384,
9087
9122
  },
9088
- "openai/gpt-4o-audio-preview": {
9089
- id: "openai/gpt-4o-audio-preview",
9090
- name: "OpenAI: GPT-4o Audio",
9091
- api: "openai-completions",
9092
- provider: "openrouter",
9093
- baseUrl: "https://openrouter.ai/api/v1",
9094
- reasoning: false,
9095
- input: ["text"],
9096
- cost: {
9097
- input: 2.5,
9098
- output: 10,
9099
- cacheRead: 0,
9100
- cacheWrite: 0,
9101
- },
9102
- contextWindow: 128000,
9103
- maxTokens: 16384,
9104
- },
9105
9123
  "openai/gpt-4o-mini": {
9106
9124
  id: "openai/gpt-4o-mini",
9107
9125
  name: "OpenAI: GPT-4o-mini",
@@ -9604,13 +9622,13 @@ export const MODELS = {
9604
9622
  reasoning: true,
9605
9623
  input: ["text"],
9606
9624
  cost: {
9607
- input: 0.03,
9625
+ input: 0.029,
9608
9626
  output: 0.14,
9609
9627
  cacheRead: 0,
9610
9628
  cacheWrite: 0,
9611
9629
  },
9612
9630
  contextWindow: 131072,
9613
- maxTokens: 131072,
9631
+ maxTokens: 4096,
9614
9632
  },
9615
9633
  "openai/gpt-oss-20b:free": {
9616
9634
  id: "openai/gpt-oss-20b:free",
@@ -9918,23 +9936,6 @@ export const MODELS = {
9918
9936
  contextWindow: 131072,
9919
9937
  maxTokens: 16384,
9920
9938
  },
9921
- "qwen/qwen-2.5-7b-instruct": {
9922
- id: "qwen/qwen-2.5-7b-instruct",
9923
- name: "Qwen: Qwen2.5 7B Instruct",
9924
- api: "openai-completions",
9925
- provider: "openrouter",
9926
- baseUrl: "https://openrouter.ai/api/v1",
9927
- reasoning: false,
9928
- input: ["text"],
9929
- cost: {
9930
- input: 0.04,
9931
- output: 0.09999999999999999,
9932
- cacheRead: 0,
9933
- cacheWrite: 0,
9934
- },
9935
- contextWindow: 131072,
9936
- maxTokens: 32768,
9937
- },
9938
9939
  "qwen/qwen-plus": {
9939
9940
  id: "qwen/qwen-plus",
9940
9941
  name: "Qwen: Qwen-Plus",
@@ -10046,13 +10047,13 @@ export const MODELS = {
10046
10047
  reasoning: true,
10047
10048
  input: ["text"],
10048
10049
  cost: {
10049
- input: 0.14950000000000002,
10050
- output: 1.495,
10051
- cacheRead: 0,
10050
+ input: 0.09999999999999999,
10051
+ output: 0.09999999999999999,
10052
+ cacheRead: 0.09999999999999999,
10052
10053
  cacheWrite: 0,
10053
10054
  },
10054
10055
  contextWindow: 262144,
10055
- maxTokens: 4096,
10056
+ maxTokens: 262144,
10056
10057
  },
10057
10058
  "qwen/qwen3-30b-a3b": {
10058
10059
  id: "qwen/qwen3-30b-a3b",
@@ -10080,13 +10081,13 @@ export const MODELS = {
10080
10081
  reasoning: false,
10081
10082
  input: ["text"],
10082
10083
  cost: {
10083
- input: 0.09,
10084
- output: 0.3,
10084
+ input: 0.0428,
10085
+ output: 0.1716,
10085
10086
  cacheRead: 0,
10086
10087
  cacheWrite: 0,
10087
10088
  },
10088
- contextWindow: 262144,
10089
- maxTokens: 262144,
10089
+ contextWindow: 131072,
10090
+ maxTokens: 32000,
10090
10091
  },
10091
10092
  "qwen/qwen3-30b-a3b-thinking-2507": {
10092
10093
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -10488,13 +10489,13 @@ export const MODELS = {
10488
10489
  reasoning: true,
10489
10490
  input: ["text", "image"],
10490
10491
  cost: {
10491
- input: 0.13899999999999998,
10492
+ input: 0.14,
10492
10493
  output: 1,
10493
- cacheRead: 0,
10494
+ cacheRead: 0.049999999999999996,
10494
10495
  cacheWrite: 0,
10495
10496
  },
10496
10497
  contextWindow: 262144,
10497
- maxTokens: 4096,
10498
+ maxTokens: 262144,
10498
10499
  },
10499
10500
  "qwen/qwen3.5-397b-a17b": {
10500
10501
  id: "qwen/qwen3.5-397b-a17b",
@@ -10768,6 +10769,23 @@ export const MODELS = {
10768
10769
  contextWindow: 262144,
10769
10770
  maxTokens: 16384,
10770
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
+ },
10771
10789
  "tencent/hy3-preview": {
10772
10790
  id: "tencent/hy3-preview",
10773
10791
  name: "Tencent: Hy3 preview",
@@ -10904,40 +10922,6 @@ export const MODELS = {
10904
10922
  contextWindow: 262144,
10905
10923
  maxTokens: 65536,
10906
10924
  },
10907
- "xiaomi/mimo-v2-omni": {
10908
- id: "xiaomi/mimo-v2-omni",
10909
- name: "Xiaomi: MiMo-V2-Omni",
10910
- api: "openai-completions",
10911
- provider: "openrouter",
10912
- baseUrl: "https://openrouter.ai/api/v1",
10913
- reasoning: true,
10914
- input: ["text", "image"],
10915
- cost: {
10916
- input: 0.39999999999999997,
10917
- output: 2,
10918
- cacheRead: 0.08,
10919
- cacheWrite: 0,
10920
- },
10921
- contextWindow: 262144,
10922
- maxTokens: 65536,
10923
- },
10924
- "xiaomi/mimo-v2-pro": {
10925
- id: "xiaomi/mimo-v2-pro",
10926
- name: "Xiaomi: MiMo-V2-Pro",
10927
- api: "openai-completions",
10928
- provider: "openrouter",
10929
- baseUrl: "https://openrouter.ai/api/v1",
10930
- reasoning: true,
10931
- input: ["text"],
10932
- cost: {
10933
- input: 1,
10934
- output: 3,
10935
- cacheRead: 0.19999999999999998,
10936
- cacheWrite: 0,
10937
- },
10938
- contextWindow: 1048576,
10939
- maxTokens: 131072,
10940
- },
10941
10925
  "xiaomi/mimo-v2.5": {
10942
10926
  id: "xiaomi/mimo-v2.5",
10943
10927
  name: "Xiaomi: MiMo-V2.5",
@@ -11135,12 +11119,12 @@ export const MODELS = {
11135
11119
  input: ["text"],
11136
11120
  cost: {
11137
11121
  input: 0.6,
11138
- output: 1.92,
11122
+ output: 2.08,
11139
11123
  cacheRead: 0.12,
11140
11124
  cacheWrite: 0,
11141
11125
  },
11142
11126
  contextWindow: 202752,
11143
- maxTokens: 4096,
11127
+ maxTokens: 16384,
11144
11128
  },
11145
11129
  "z-ai/glm-5-turbo": {
11146
11130
  id: "z-ai/glm-5-turbo",
@@ -11174,7 +11158,7 @@ export const MODELS = {
11174
11158
  cacheWrite: 0,
11175
11159
  },
11176
11160
  contextWindow: 202752,
11177
- maxTokens: 4096,
11161
+ maxTokens: 131072,
11178
11162
  },
11179
11163
  "z-ai/glm-5v-turbo": {
11180
11164
  id: "z-ai/glm-5v-turbo",
@@ -11287,13 +11271,13 @@ export const MODELS = {
11287
11271
  reasoning: true,
11288
11272
  input: ["text", "image"],
11289
11273
  cost: {
11290
- input: 0.73,
11291
- output: 3.49,
11292
- cacheRead: 0.25,
11274
+ input: 0.684,
11275
+ output: 3.42,
11276
+ cacheRead: 0.144,
11293
11277
  cacheWrite: 0,
11294
11278
  },
11295
11279
  contextWindow: 262144,
11296
- maxTokens: 262142,
11280
+ maxTokens: 262144,
11297
11281
  },
11298
11282
  "~openai/gpt-latest": {
11299
11283
  id: "~openai/gpt-latest",
@@ -11350,20 +11334,20 @@ export const MODELS = {
11350
11334
  },
11351
11335
  "alibaba/qwen-3-235b": {
11352
11336
  id: "alibaba/qwen-3-235b",
11353
- name: "Qwen3 235B A22b Instruct 2507",
11337
+ name: "Qwen3 235B A22B",
11354
11338
  api: "anthropic-messages",
11355
11339
  provider: "vercel-ai-gateway",
11356
11340
  baseUrl: "https://ai-gateway.vercel.sh",
11357
- reasoning: false,
11341
+ reasoning: true,
11358
11342
  input: ["text"],
11359
11343
  cost: {
11360
- input: 0.6,
11361
- output: 1.2,
11362
- cacheRead: 0.6,
11344
+ input: 0.22,
11345
+ output: 0.88,
11346
+ cacheRead: 0,
11363
11347
  cacheWrite: 0,
11364
11348
  },
11365
- contextWindow: 131000,
11366
- maxTokens: 40000,
11349
+ contextWindow: 262144,
11350
+ maxTokens: 16384,
11367
11351
  },
11368
11352
  "alibaba/qwen-3-30b": {
11369
11353
  id: "alibaba/qwen-3-30b",
@@ -11439,7 +11423,7 @@ export const MODELS = {
11439
11423
  api: "anthropic-messages",
11440
11424
  provider: "vercel-ai-gateway",
11441
11425
  baseUrl: "https://ai-gateway.vercel.sh",
11442
- reasoning: false,
11426
+ reasoning: true,
11443
11427
  input: ["text"],
11444
11428
  cost: {
11445
11429
  input: 1.5,
@@ -11473,7 +11457,7 @@ export const MODELS = {
11473
11457
  api: "anthropic-messages",
11474
11458
  provider: "vercel-ai-gateway",
11475
11459
  baseUrl: "https://ai-gateway.vercel.sh",
11476
- reasoning: false,
11460
+ reasoning: true,
11477
11461
  input: ["text"],
11478
11462
  cost: {
11479
11463
  input: 0.5,
@@ -11552,6 +11536,40 @@ export const MODELS = {
11552
11536
  contextWindow: 256000,
11553
11537
  maxTokens: 65536,
11554
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
+ },
11555
11573
  "alibaba/qwen3-vl-thinking": {
11556
11574
  id: "alibaba/qwen3-vl-thinking",
11557
11575
  name: "Qwen3 VL 235B A22B Thinking",
@@ -11654,6 +11672,23 @@ export const MODELS = {
11654
11672
  contextWindow: 991000,
11655
11673
  maxTokens: 64000,
11656
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
+ },
11657
11692
  "anthropic/claude-3-haiku": {
11658
11693
  id: "anthropic/claude-3-haiku",
11659
11694
  name: "Claude 3 Haiku",
@@ -11952,17 +11987,17 @@ export const MODELS = {
11952
11987
  reasoning: false,
11953
11988
  input: ["text"],
11954
11989
  cost: {
11955
- input: 0.77,
11956
- output: 0.77,
11957
- cacheRead: 0,
11990
+ input: 0.27,
11991
+ output: 1.12,
11992
+ cacheRead: 0.135,
11958
11993
  cacheWrite: 0,
11959
11994
  },
11960
11995
  contextWindow: 163840,
11961
- maxTokens: 16384,
11996
+ maxTokens: 163840,
11962
11997
  },
11963
11998
  "deepseek/deepseek-v3.1": {
11964
11999
  id: "deepseek/deepseek-v3.1",
11965
- name: "DeepSeek-V3.1",
12000
+ name: "DeepSeek V3.1",
11966
12001
  api: "anthropic-messages",
11967
12002
  provider: "vercel-ai-gateway",
11968
12003
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -12000,8 +12035,8 @@ export const MODELS = {
12000
12035
  api: "anthropic-messages",
12001
12036
  provider: "vercel-ai-gateway",
12002
12037
  baseUrl: "https://ai-gateway.vercel.sh",
12003
- reasoning: false,
12004
- input: ["text"],
12038
+ reasoning: true,
12039
+ input: ["text", "image"],
12005
12040
  cost: {
12006
12041
  input: 0.28,
12007
12042
  output: 0.42,
@@ -12017,8 +12052,8 @@ export const MODELS = {
12017
12052
  api: "anthropic-messages",
12018
12053
  provider: "vercel-ai-gateway",
12019
12054
  baseUrl: "https://ai-gateway.vercel.sh",
12020
- reasoning: false,
12021
- input: ["text"],
12055
+ reasoning: true,
12056
+ input: ["text", "image"],
12022
12057
  cost: {
12023
12058
  input: 0.62,
12024
12059
  output: 1.85,
@@ -12035,7 +12070,7 @@ export const MODELS = {
12035
12070
  provider: "vercel-ai-gateway",
12036
12071
  baseUrl: "https://ai-gateway.vercel.sh",
12037
12072
  reasoning: true,
12038
- input: ["text"],
12073
+ input: ["text", "image"],
12039
12074
  cost: {
12040
12075
  input: 0.14,
12041
12076
  output: 0.28,
@@ -12052,7 +12087,7 @@ export const MODELS = {
12052
12087
  provider: "vercel-ai-gateway",
12053
12088
  baseUrl: "https://ai-gateway.vercel.sh",
12054
12089
  reasoning: true,
12055
- input: ["text"],
12090
+ input: ["text", "image"],
12056
12091
  cost: {
12057
12092
  input: 0.435,
12058
12093
  output: 0.87,
@@ -12545,7 +12580,7 @@ export const MODELS = {
12545
12580
  provider: "vercel-ai-gateway",
12546
12581
  baseUrl: "https://ai-gateway.vercel.sh",
12547
12582
  reasoning: true,
12548
- input: ["text"],
12583
+ input: ["text", "image"],
12549
12584
  cost: {
12550
12585
  input: 0.6,
12551
12586
  output: 2.4,
@@ -12589,6 +12624,23 @@ export const MODELS = {
12589
12624
  contextWindow: 204800,
12590
12625
  maxTokens: 131100,
12591
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
+ },
12592
12644
  "mistral/codestral": {
12593
12645
  id: "mistral/codestral",
12594
12646
  name: "Mistral Codestral",
@@ -12725,6 +12777,23 @@ export const MODELS = {
12725
12777
  contextWindow: 256000,
12726
12778
  maxTokens: 256000,
12727
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
+ },
12728
12797
  "mistral/mistral-small": {
12729
12798
  id: "mistral/mistral-small",
12730
12799
  name: "Mistral Small",
@@ -12878,6 +12947,23 @@ export const MODELS = {
12878
12947
  contextWindow: 262000,
12879
12948
  maxTokens: 262000,
12880
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
+ },
12881
12967
  "nvidia/nemotron-nano-12b-v2-vl": {
12882
12968
  id: "nvidia/nemotron-nano-12b-v2-vl",
12883
12969
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -13055,7 +13141,7 @@ export const MODELS = {
13055
13141
  provider: "vercel-ai-gateway",
13056
13142
  baseUrl: "https://ai-gateway.vercel.sh",
13057
13143
  reasoning: true,
13058
- input: ["text"],
13144
+ input: ["text", "image"],
13059
13145
  cost: {
13060
13146
  input: 1.25,
13061
13147
  output: 10,
@@ -13405,6 +13491,23 @@ export const MODELS = {
13405
13491
  contextWindow: 1000000,
13406
13492
  maxTokens: 128000,
13407
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
+ },
13408
13511
  "openai/gpt-oss-20b": {
13409
13512
  id: "openai/gpt-oss-20b",
13410
13513
  name: "GPT OSS 20B",
@@ -13575,6 +13678,40 @@ export const MODELS = {
13575
13678
  contextWindow: 200000,
13576
13679
  maxTokens: 8000,
13577
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
+ },
13578
13715
  "xai/grok-4.1-fast-non-reasoning": {
13579
13716
  id: "xai/grok-4.1-fast-non-reasoning",
13580
13717
  name: "Grok 4.1 Fast Non-Reasoning",
@@ -13853,7 +13990,7 @@ export const MODELS = {
13853
13990
  api: "anthropic-messages",
13854
13991
  provider: "vercel-ai-gateway",
13855
13992
  baseUrl: "https://ai-gateway.vercel.sh",
13856
- reasoning: false,
13993
+ reasoning: true,
13857
13994
  input: ["text", "image"],
13858
13995
  cost: {
13859
13996
  input: 0.6,
@@ -14007,7 +14144,7 @@ export const MODELS = {
14007
14144
  provider: "vercel-ai-gateway",
14008
14145
  baseUrl: "https://ai-gateway.vercel.sh",
14009
14146
  reasoning: true,
14010
- input: ["text"],
14147
+ input: ["text", "image"],
14011
14148
  cost: {
14012
14149
  input: 1.4,
14013
14150
  output: 4.4,