@mariozechner/pi-ai 0.68.0 → 0.68.1

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 (22) hide show
  1. package/README.md +2 -0
  2. package/dist/env-api-keys.d.ts.map +1 -1
  3. package/dist/env-api-keys.js +1 -0
  4. package/dist/env-api-keys.js.map +1 -1
  5. package/dist/models.generated.d.ts +447 -51
  6. package/dist/models.generated.d.ts.map +1 -1
  7. package/dist/models.generated.js +468 -74
  8. package/dist/models.generated.js.map +1 -1
  9. package/dist/providers/amazon-bedrock.d.ts.map +1 -1
  10. package/dist/providers/amazon-bedrock.js +49 -9
  11. package/dist/providers/amazon-bedrock.js.map +1 -1
  12. package/dist/providers/anthropic.d.ts.map +1 -1
  13. package/dist/providers/anthropic.js +140 -8
  14. package/dist/providers/anthropic.js.map +1 -1
  15. package/dist/types.d.ts +1 -1
  16. package/dist/types.d.ts.map +1 -1
  17. package/dist/types.js.map +1 -1
  18. package/dist/utils/json-parse.d.ts +8 -1
  19. package/dist/utils/json-parse.d.ts.map +1 -1
  20. package/dist/utils/json-parse.js +89 -5
  21. package/dist/utils/json-parse.js.map +1 -1
  22. package/package.json +1 -1
package/dist/models.generated.js CHANGED
@@ -381,7 +381,7 @@ export const MODELS = {
381
381
  name: "Claude Haiku 4.5 (EU)",
382
382
  api: "bedrock-converse-stream",
383
383
  provider: "amazon-bedrock",
384
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
384
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
385
385
  reasoning: true,
386
386
  input: ["text", "image"],
387
387
  cost: {
@@ -398,7 +398,7 @@ export const MODELS = {
398
398
  name: "Claude Opus 4.5 (EU)",
399
399
  api: "bedrock-converse-stream",
400
400
  provider: "amazon-bedrock",
401
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
401
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
402
402
  reasoning: true,
403
403
  input: ["text", "image"],
404
404
  cost: {
@@ -415,7 +415,7 @@ export const MODELS = {
415
415
  name: "Claude Opus 4.6 (EU)",
416
416
  api: "bedrock-converse-stream",
417
417
  provider: "amazon-bedrock",
418
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
418
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
419
419
  reasoning: true,
420
420
  input: ["text", "image"],
421
421
  cost: {
@@ -432,7 +432,7 @@ export const MODELS = {
432
432
  name: "Claude Opus 4.7 (EU)",
433
433
  api: "bedrock-converse-stream",
434
434
  provider: "amazon-bedrock",
435
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
435
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
436
436
  reasoning: true,
437
437
  input: ["text", "image"],
438
438
  cost: {
@@ -449,7 +449,7 @@ export const MODELS = {
449
449
  name: "Claude Sonnet 4 (EU)",
450
450
  api: "bedrock-converse-stream",
451
451
  provider: "amazon-bedrock",
452
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
452
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
453
453
  reasoning: true,
454
454
  input: ["text", "image"],
455
455
  cost: {
@@ -466,7 +466,7 @@ export const MODELS = {
466
466
  name: "Claude Sonnet 4.5 (EU)",
467
467
  api: "bedrock-converse-stream",
468
468
  provider: "amazon-bedrock",
469
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
469
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
470
470
  reasoning: true,
471
471
  input: ["text", "image"],
472
472
  cost: {
@@ -483,7 +483,7 @@ export const MODELS = {
483
483
  name: "Claude Sonnet 4.6 (EU)",
484
484
  api: "bedrock-converse-stream",
485
485
  provider: "amazon-bedrock",
486
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
486
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
487
487
  reasoning: true,
488
488
  input: ["text", "image"],
489
489
  cost: {
@@ -2695,6 +2695,314 @@ export const MODELS = {
2695
2695
  maxTokens: 40000,
2696
2696
  },
2697
2697
  },
2698
+ "fireworks": {
2699
+ "accounts/fireworks/models/deepseek-v3p1": {
2700
+ id: "accounts/fireworks/models/deepseek-v3p1",
2701
+ name: "DeepSeek V3.1",
2702
+ api: "anthropic-messages",
2703
+ provider: "fireworks",
2704
+ baseUrl: "https://api.fireworks.ai/inference",
2705
+ reasoning: true,
2706
+ input: ["text"],
2707
+ cost: {
2708
+ input: 0.56,
2709
+ output: 1.68,
2710
+ cacheRead: 0,
2711
+ cacheWrite: 0,
2712
+ },
2713
+ contextWindow: 163840,
2714
+ maxTokens: 163840,
2715
+ },
2716
+ "accounts/fireworks/models/deepseek-v3p2": {
2717
+ id: "accounts/fireworks/models/deepseek-v3p2",
2718
+ name: "DeepSeek V3.2",
2719
+ api: "anthropic-messages",
2720
+ provider: "fireworks",
2721
+ baseUrl: "https://api.fireworks.ai/inference",
2722
+ reasoning: true,
2723
+ input: ["text"],
2724
+ cost: {
2725
+ input: 0.56,
2726
+ output: 1.68,
2727
+ cacheRead: 0.28,
2728
+ cacheWrite: 0,
2729
+ },
2730
+ contextWindow: 160000,
2731
+ maxTokens: 160000,
2732
+ },
2733
+ "accounts/fireworks/models/glm-4p5": {
2734
+ id: "accounts/fireworks/models/glm-4p5",
2735
+ name: "GLM 4.5",
2736
+ api: "anthropic-messages",
2737
+ provider: "fireworks",
2738
+ baseUrl: "https://api.fireworks.ai/inference",
2739
+ reasoning: true,
2740
+ input: ["text"],
2741
+ cost: {
2742
+ input: 0.55,
2743
+ output: 2.19,
2744
+ cacheRead: 0,
2745
+ cacheWrite: 0,
2746
+ },
2747
+ contextWindow: 131072,
2748
+ maxTokens: 131072,
2749
+ },
2750
+ "accounts/fireworks/models/glm-4p5-air": {
2751
+ id: "accounts/fireworks/models/glm-4p5-air",
2752
+ name: "GLM 4.5 Air",
2753
+ api: "anthropic-messages",
2754
+ provider: "fireworks",
2755
+ baseUrl: "https://api.fireworks.ai/inference",
2756
+ reasoning: true,
2757
+ input: ["text"],
2758
+ cost: {
2759
+ input: 0.22,
2760
+ output: 0.88,
2761
+ cacheRead: 0,
2762
+ cacheWrite: 0,
2763
+ },
2764
+ contextWindow: 131072,
2765
+ maxTokens: 131072,
2766
+ },
2767
+ "accounts/fireworks/models/glm-4p7": {
2768
+ id: "accounts/fireworks/models/glm-4p7",
2769
+ name: "GLM 4.7",
2770
+ api: "anthropic-messages",
2771
+ provider: "fireworks",
2772
+ baseUrl: "https://api.fireworks.ai/inference",
2773
+ reasoning: true,
2774
+ input: ["text"],
2775
+ cost: {
2776
+ input: 0.6,
2777
+ output: 2.2,
2778
+ cacheRead: 0.3,
2779
+ cacheWrite: 0,
2780
+ },
2781
+ contextWindow: 198000,
2782
+ maxTokens: 198000,
2783
+ },
2784
+ "accounts/fireworks/models/glm-5": {
2785
+ id: "accounts/fireworks/models/glm-5",
2786
+ name: "GLM 5",
2787
+ api: "anthropic-messages",
2788
+ provider: "fireworks",
2789
+ baseUrl: "https://api.fireworks.ai/inference",
2790
+ reasoning: true,
2791
+ input: ["text"],
2792
+ cost: {
2793
+ input: 1,
2794
+ output: 3.2,
2795
+ cacheRead: 0.5,
2796
+ cacheWrite: 0,
2797
+ },
2798
+ contextWindow: 202752,
2799
+ maxTokens: 131072,
2800
+ },
2801
+ "accounts/fireworks/models/glm-5p1": {
2802
+ id: "accounts/fireworks/models/glm-5p1",
2803
+ name: "GLM 5.1",
2804
+ api: "anthropic-messages",
2805
+ provider: "fireworks",
2806
+ baseUrl: "https://api.fireworks.ai/inference",
2807
+ reasoning: true,
2808
+ input: ["text"],
2809
+ cost: {
2810
+ input: 1.4,
2811
+ output: 4.4,
2812
+ cacheRead: 0.26,
2813
+ cacheWrite: 0,
2814
+ },
2815
+ contextWindow: 202800,
2816
+ maxTokens: 131072,
2817
+ },
2818
+ "accounts/fireworks/models/gpt-oss-120b": {
2819
+ id: "accounts/fireworks/models/gpt-oss-120b",
2820
+ name: "GPT OSS 120B",
2821
+ api: "anthropic-messages",
2822
+ provider: "fireworks",
2823
+ baseUrl: "https://api.fireworks.ai/inference",
2824
+ reasoning: true,
2825
+ input: ["text"],
2826
+ cost: {
2827
+ input: 0.15,
2828
+ output: 0.6,
2829
+ cacheRead: 0,
2830
+ cacheWrite: 0,
2831
+ },
2832
+ contextWindow: 131072,
2833
+ maxTokens: 32768,
2834
+ },
2835
+ "accounts/fireworks/models/gpt-oss-20b": {
2836
+ id: "accounts/fireworks/models/gpt-oss-20b",
2837
+ name: "GPT OSS 20B",
2838
+ api: "anthropic-messages",
2839
+ provider: "fireworks",
2840
+ baseUrl: "https://api.fireworks.ai/inference",
2841
+ reasoning: true,
2842
+ input: ["text"],
2843
+ cost: {
2844
+ input: 0.05,
2845
+ output: 0.2,
2846
+ cacheRead: 0,
2847
+ cacheWrite: 0,
2848
+ },
2849
+ contextWindow: 131072,
2850
+ maxTokens: 32768,
2851
+ },
2852
+ "accounts/fireworks/models/kimi-k2-instruct": {
2853
+ id: "accounts/fireworks/models/kimi-k2-instruct",
2854
+ name: "Kimi K2 Instruct",
2855
+ api: "anthropic-messages",
2856
+ provider: "fireworks",
2857
+ baseUrl: "https://api.fireworks.ai/inference",
2858
+ reasoning: false,
2859
+ input: ["text"],
2860
+ cost: {
2861
+ input: 1,
2862
+ output: 3,
2863
+ cacheRead: 0,
2864
+ cacheWrite: 0,
2865
+ },
2866
+ contextWindow: 128000,
2867
+ maxTokens: 16384,
2868
+ },
2869
+ "accounts/fireworks/models/kimi-k2-thinking": {
2870
+ id: "accounts/fireworks/models/kimi-k2-thinking",
2871
+ name: "Kimi K2 Thinking",
2872
+ api: "anthropic-messages",
2873
+ provider: "fireworks",
2874
+ baseUrl: "https://api.fireworks.ai/inference",
2875
+ reasoning: true,
2876
+ input: ["text"],
2877
+ cost: {
2878
+ input: 0.6,
2879
+ output: 2.5,
2880
+ cacheRead: 0.3,
2881
+ cacheWrite: 0,
2882
+ },
2883
+ contextWindow: 256000,
2884
+ maxTokens: 256000,
2885
+ },
2886
+ "accounts/fireworks/models/kimi-k2p5": {
2887
+ id: "accounts/fireworks/models/kimi-k2p5",
2888
+ name: "Kimi K2.5",
2889
+ api: "anthropic-messages",
2890
+ provider: "fireworks",
2891
+ baseUrl: "https://api.fireworks.ai/inference",
2892
+ reasoning: true,
2893
+ input: ["text", "image"],
2894
+ cost: {
2895
+ input: 0.6,
2896
+ output: 3,
2897
+ cacheRead: 0.1,
2898
+ cacheWrite: 0,
2899
+ },
2900
+ contextWindow: 256000,
2901
+ maxTokens: 256000,
2902
+ },
2903
+ "accounts/fireworks/models/kimi-k2p6": {
2904
+ id: "accounts/fireworks/models/kimi-k2p6",
2905
+ name: "Kimi K2.6",
2906
+ api: "anthropic-messages",
2907
+ provider: "fireworks",
2908
+ baseUrl: "https://api.fireworks.ai/inference",
2909
+ reasoning: true,
2910
+ input: ["text", "image"],
2911
+ cost: {
2912
+ input: 0.95,
2913
+ output: 4,
2914
+ cacheRead: 0.16,
2915
+ cacheWrite: 0,
2916
+ },
2917
+ contextWindow: 262000,
2918
+ maxTokens: 262000,
2919
+ },
2920
+ "accounts/fireworks/models/minimax-m2p1": {
2921
+ id: "accounts/fireworks/models/minimax-m2p1",
2922
+ name: "MiniMax-M2.1",
2923
+ api: "anthropic-messages",
2924
+ provider: "fireworks",
2925
+ baseUrl: "https://api.fireworks.ai/inference",
2926
+ reasoning: true,
2927
+ input: ["text"],
2928
+ cost: {
2929
+ input: 0.3,
2930
+ output: 1.2,
2931
+ cacheRead: 0.03,
2932
+ cacheWrite: 0,
2933
+ },
2934
+ contextWindow: 200000,
2935
+ maxTokens: 200000,
2936
+ },
2937
+ "accounts/fireworks/models/minimax-m2p5": {
2938
+ id: "accounts/fireworks/models/minimax-m2p5",
2939
+ name: "MiniMax-M2.5",
2940
+ api: "anthropic-messages",
2941
+ provider: "fireworks",
2942
+ baseUrl: "https://api.fireworks.ai/inference",
2943
+ reasoning: true,
2944
+ input: ["text"],
2945
+ cost: {
2946
+ input: 0.3,
2947
+ output: 1.2,
2948
+ cacheRead: 0.03,
2949
+ cacheWrite: 0,
2950
+ },
2951
+ contextWindow: 196608,
2952
+ maxTokens: 196608,
2953
+ },
2954
+ "accounts/fireworks/models/minimax-m2p7": {
2955
+ id: "accounts/fireworks/models/minimax-m2p7",
2956
+ name: "MiniMax-M2.7",
2957
+ api: "anthropic-messages",
2958
+ provider: "fireworks",
2959
+ baseUrl: "https://api.fireworks.ai/inference",
2960
+ reasoning: true,
2961
+ input: ["text"],
2962
+ cost: {
2963
+ input: 0.3,
2964
+ output: 1.2,
2965
+ cacheRead: 0.03,
2966
+ cacheWrite: 0,
2967
+ },
2968
+ contextWindow: 196608,
2969
+ maxTokens: 196608,
2970
+ },
2971
+ "accounts/fireworks/models/qwen3p6-plus": {
2972
+ id: "accounts/fireworks/models/qwen3p6-plus",
2973
+ name: "Qwen 3.6 Plus",
2974
+ api: "anthropic-messages",
2975
+ provider: "fireworks",
2976
+ baseUrl: "https://api.fireworks.ai/inference",
2977
+ reasoning: true,
2978
+ input: ["text", "image"],
2979
+ cost: {
2980
+ input: 0.5,
2981
+ output: 3,
2982
+ cacheRead: 0.1,
2983
+ cacheWrite: 0,
2984
+ },
2985
+ contextWindow: 128000,
2986
+ maxTokens: 8192,
2987
+ },
2988
+ "accounts/fireworks/routers/kimi-k2p5-turbo": {
2989
+ id: "accounts/fireworks/routers/kimi-k2p5-turbo",
2990
+ name: "Kimi K2.5 Turbo (firepass)",
2991
+ api: "anthropic-messages",
2992
+ provider: "fireworks",
2993
+ baseUrl: "https://api.fireworks.ai/inference",
2994
+ reasoning: true,
2995
+ input: ["text", "image"],
2996
+ cost: {
2997
+ input: 0,
2998
+ output: 0,
2999
+ cacheRead: 0,
3000
+ cacheWrite: 0,
3001
+ },
3002
+ contextWindow: 256000,
3003
+ maxTokens: 256000,
3004
+ },
3005
+ },
2698
3006
  "github-copilot": {
2699
3007
  "claude-haiku-4.5": {
2700
3008
  id: "claude-haiku-4.5",
@@ -4694,6 +5002,24 @@ export const MODELS = {
4694
5002
  contextWindow: 262144,
4695
5003
  maxTokens: 262144,
4696
5004
  },
5005
+ "moonshotai/Kimi-K2.6": {
5006
+ id: "moonshotai/Kimi-K2.6",
5007
+ name: "Kimi-K2.6",
5008
+ api: "openai-completions",
5009
+ provider: "huggingface",
5010
+ baseUrl: "https://router.huggingface.co/v1",
5011
+ compat: { "supportsDeveloperRole": false },
5012
+ reasoning: true,
5013
+ input: ["text", "image"],
5014
+ cost: {
5015
+ input: 0.95,
5016
+ output: 4,
5017
+ cacheRead: 0.16,
5018
+ cacheWrite: 0,
5019
+ },
5020
+ contextWindow: 262144,
5021
+ maxTokens: 262144,
5022
+ },
4697
5023
  "zai-org/GLM-4.7": {
4698
5024
  id: "zai-org/GLM-4.7",
4699
5025
  name: "GLM-4.7",
@@ -4768,6 +5094,23 @@ export const MODELS = {
4768
5094
  },
4769
5095
  },
4770
5096
  "kimi-coding": {
5097
+ "k2p6": {
5098
+ id: "k2p6",
5099
+ name: "Kimi K2.6",
5100
+ api: "anthropic-messages",
5101
+ provider: "kimi-coding",
5102
+ baseUrl: "https://api.kimi.com/coding",
5103
+ reasoning: true,
5104
+ input: ["text", "image"],
5105
+ cost: {
5106
+ input: 0,
5107
+ output: 0,
5108
+ cacheRead: 0,
5109
+ cacheWrite: 0,
5110
+ },
5111
+ contextWindow: 262144,
5112
+ maxTokens: 32768,
5113
+ },
4771
5114
  "kimi-for-coding": {
4772
5115
  id: "kimi-for-coding",
4773
5116
  name: "Kimi For Coding",
@@ -6650,6 +6993,40 @@ export const MODELS = {
6650
6993
  contextWindow: 262144,
6651
6994
  maxTokens: 65536,
6652
6995
  },
6996
+ "kimi-k2.6": {
6997
+ id: "kimi-k2.6",
6998
+ name: "Kimi K2.6",
6999
+ api: "openai-completions",
7000
+ provider: "opencode",
7001
+ baseUrl: "https://opencode.ai/zen/v1",
7002
+ reasoning: true,
7003
+ input: ["text", "image"],
7004
+ cost: {
7005
+ input: 0.95,
7006
+ output: 4,
7007
+ cacheRead: 0.16,
7008
+ cacheWrite: 0,
7009
+ },
7010
+ contextWindow: 262144,
7011
+ maxTokens: 65536,
7012
+ },
7013
+ "ling-2.6-flash-free": {
7014
+ id: "ling-2.6-flash-free",
7015
+ name: "Ling 2.6 Flash Free",
7016
+ api: "openai-completions",
7017
+ provider: "opencode",
7018
+ baseUrl: "https://opencode.ai/zen/v1",
7019
+ reasoning: false,
7020
+ input: ["text"],
7021
+ cost: {
7022
+ input: 0,
7023
+ output: 0,
7024
+ cacheRead: 0,
7025
+ cacheWrite: 0,
7026
+ },
7027
+ contextWindow: 262100,
7028
+ maxTokens: 32800,
7029
+ },
6653
7030
  "minimax-m2.5": {
6654
7031
  id: "minimax-m2.5",
6655
7032
  name: "MiniMax M2.5",
@@ -6684,6 +7061,23 @@ export const MODELS = {
6684
7061
  contextWindow: 204800,
6685
7062
  maxTokens: 131072,
6686
7063
  },
7064
+ "minimax-m2.7": {
7065
+ id: "minimax-m2.7",
7066
+ name: "MiniMax M2.7",
7067
+ api: "openai-completions",
7068
+ provider: "opencode",
7069
+ baseUrl: "https://opencode.ai/zen/v1",
7070
+ reasoning: true,
7071
+ input: ["text"],
7072
+ cost: {
7073
+ input: 0.3,
7074
+ output: 1.2,
7075
+ cacheRead: 0.06,
7076
+ cacheWrite: 0,
7077
+ },
7078
+ contextWindow: 204800,
7079
+ maxTokens: 131072,
7080
+ },
6687
7081
  "nemotron-3-super-free": {
6688
7082
  id: "nemotron-3-super-free",
6689
7083
  name: "Nemotron 3 Super Free",
@@ -6792,16 +7186,16 @@ export const MODELS = {
6792
7186
  },
6793
7187
  "kimi-k2.6": {
6794
7188
  id: "kimi-k2.6",
6795
- name: "Kimi K2.6",
7189
+ name: "Kimi K2.6 (3x limits)",
6796
7190
  api: "openai-completions",
6797
7191
  provider: "opencode-go",
6798
7192
  baseUrl: "https://opencode.ai/zen/go/v1",
6799
7193
  reasoning: true,
6800
7194
  input: ["text", "image"],
6801
7195
  cost: {
6802
- input: 0.95,
6803
- output: 4,
6804
- cacheRead: 0.16,
7196
+ input: 0.32,
7197
+ output: 1.34,
7198
+ cacheRead: 0.054,
6805
7199
  cacheWrite: 0,
6806
7200
  },
6807
7201
  contextWindow: 262144,
@@ -7874,13 +8268,13 @@ export const MODELS = {
7874
8268
  reasoning: true,
7875
8269
  input: ["text", "image"],
7876
8270
  cost: {
7877
- input: 0.08,
8271
+ input: 0.07,
7878
8272
  output: 0.35,
7879
- cacheRead: 0.01,
8273
+ cacheRead: 0.04,
7880
8274
  cacheWrite: 0,
7881
8275
  },
7882
8276
  contextWindow: 262144,
7883
- maxTokens: 4096,
8277
+ maxTokens: 262144,
7884
8278
  },
7885
8279
  "google/gemma-4-26b-a4b-it:free": {
7886
8280
  id: "google/gemma-4-26b-a4b-it:free",
@@ -7950,6 +8344,23 @@ export const MODELS = {
7950
8344
  contextWindow: 128000,
7951
8345
  maxTokens: 50000,
7952
8346
  },
8347
+ "inclusionai/ling-2.6-flash:free": {
8348
+ id: "inclusionai/ling-2.6-flash:free",
8349
+ name: "inclusionAI: Ling-2.6-flash (free)",
8350
+ api: "openai-completions",
8351
+ provider: "openrouter",
8352
+ baseUrl: "https://openrouter.ai/api/v1",
8353
+ reasoning: false,
8354
+ input: ["text"],
8355
+ cost: {
8356
+ input: 0,
8357
+ output: 0,
8358
+ cacheRead: 0,
8359
+ cacheWrite: 0,
8360
+ },
8361
+ contextWindow: 262144,
8362
+ maxTokens: 32768,
8363
+ },
7953
8364
  "kwaipilot/kat-coder-pro-v2": {
7954
8365
  id: "kwaipilot/kat-coder-pro-v2",
7955
8366
  name: "Kwaipilot: KAT-Coder-Pro V2",
@@ -8027,13 +8438,13 @@ export const MODELS = {
8027
8438
  reasoning: false,
8028
8439
  input: ["text"],
8029
8440
  cost: {
8030
- input: 0.12,
8031
- output: 0.38,
8441
+ input: 0.09999999999999999,
8442
+ output: 0.32,
8032
8443
  cacheRead: 0,
8033
8444
  cacheWrite: 0,
8034
8445
  },
8035
8446
  contextWindow: 131072,
8036
- maxTokens: 131072,
8447
+ maxTokens: 16384,
8037
8448
  },
8038
8449
  "meta-llama/llama-3.3-70b-instruct:free": {
8039
8450
  id: "meta-llama/llama-3.3-70b-instruct:free",
@@ -8622,9 +9033,9 @@ export const MODELS = {
8622
9033
  reasoning: true,
8623
9034
  input: ["text", "image"],
8624
9035
  cost: {
8625
- input: 0,
8626
- output: 0,
8627
- cacheRead: 0,
9036
+ input: 0.7999999999999999,
9037
+ output: 3.5,
9038
+ cacheRead: 0.19999999999999998,
8628
9039
  cacheWrite: 0,
8629
9040
  },
8630
9041
  contextWindow: 262144,
@@ -9140,40 +9551,6 @@ export const MODELS = {
9140
9551
  contextWindow: 400000,
9141
9552
  maxTokens: 128000,
9142
9553
  },
9143
- "openai/gpt-5-image": {
9144
- id: "openai/gpt-5-image",
9145
- name: "OpenAI: GPT-5 Image",
9146
- api: "openai-completions",
9147
- provider: "openrouter",
9148
- baseUrl: "https://openrouter.ai/api/v1",
9149
- reasoning: true,
9150
- input: ["text", "image"],
9151
- cost: {
9152
- input: 10,
9153
- output: 10,
9154
- cacheRead: 1.25,
9155
- cacheWrite: 0,
9156
- },
9157
- contextWindow: 400000,
9158
- maxTokens: 128000,
9159
- },
9160
- "openai/gpt-5-image-mini": {
9161
- id: "openai/gpt-5-image-mini",
9162
- name: "OpenAI: GPT-5 Image Mini",
9163
- api: "openai-completions",
9164
- provider: "openrouter",
9165
- baseUrl: "https://openrouter.ai/api/v1",
9166
- reasoning: true,
9167
- input: ["text", "image"],
9168
- cost: {
9169
- input: 2.5,
9170
- output: 2,
9171
- cacheRead: 0.25,
9172
- cacheWrite: 0,
9173
- },
9174
- contextWindow: 400000,
9175
- maxTokens: 128000,
9176
- },
9177
9554
  "openai/gpt-5-mini": {
9178
9555
  id: "openai/gpt-5-mini",
9179
9556
  name: "OpenAI: GPT-5 Mini",
@@ -9769,23 +10146,6 @@ export const MODELS = {
9769
10146
  contextWindow: 2000000,
9770
10147
  maxTokens: 4096,
9771
10148
  },
9772
- "openrouter/elephant-alpha": {
9773
- id: "openrouter/elephant-alpha",
9774
- name: "Elephant",
9775
- api: "openai-completions",
9776
- provider: "openrouter",
9777
- baseUrl: "https://openrouter.ai/api/v1",
9778
- reasoning: false,
9779
- input: ["text"],
9780
- cost: {
9781
- input: 0,
9782
- output: 0,
9783
- cacheRead: 0,
9784
- cacheWrite: 0,
9785
- },
9786
- contextWindow: 262144,
9787
- maxTokens: 32768,
9788
- },
9789
10149
  "openrouter/free": {
9790
10150
  id: "openrouter/free",
9791
10151
  name: "Free Models Router",
@@ -10171,7 +10531,7 @@ export const MODELS = {
10171
10531
  cost: {
10172
10532
  input: 0.15,
10173
10533
  output: 0.7999999999999999,
10174
- cacheRead: 0.12,
10534
+ cacheRead: 0.11,
10175
10535
  cacheWrite: 0,
10176
10536
  },
10177
10537
  contextWindow: 262144,
@@ -11092,8 +11452,8 @@ export const MODELS = {
11092
11452
  cacheRead: 0.119,
11093
11453
  cacheWrite: 0,
11094
11454
  },
11095
- contextWindow: 80000,
11096
- maxTokens: 131072,
11455
+ contextWindow: 202752,
11456
+ maxTokens: 4096,
11097
11457
  },
11098
11458
  "z-ai/glm-5-turbo": {
11099
11459
  id: "z-ai/glm-5-turbo",
@@ -11146,6 +11506,23 @@ export const MODELS = {
11146
11506
  contextWindow: 202752,
11147
11507
  maxTokens: 131072,
11148
11508
  },
11509
+ "~anthropic/claude-opus-latest": {
11510
+ id: "~anthropic/claude-opus-latest",
11511
+ name: "Anthropic: Claude Opus Latest",
11512
+ api: "openai-completions",
11513
+ provider: "openrouter",
11514
+ baseUrl: "https://openrouter.ai/api/v1",
11515
+ reasoning: true,
11516
+ input: ["text", "image"],
11517
+ cost: {
11518
+ input: 5,
11519
+ output: 25,
11520
+ cacheRead: 0.5,
11521
+ cacheWrite: 6.25,
11522
+ },
11523
+ contextWindow: 1000000,
11524
+ maxTokens: 128000,
11525
+ },
11149
11526
  },
11150
11527
  "vercel-ai-gateway": {
11151
11528
  "alibaba/qwen-3-14b": {
@@ -11216,6 +11593,23 @@ export const MODELS = {
11216
11593
  contextWindow: 128000,
11217
11594
  maxTokens: 8192,
11218
11595
  },
11596
+ "alibaba/qwen-3.6-max-preview": {
11597
+ id: "alibaba/qwen-3.6-max-preview",
11598
+ name: "Qwen 3.6 Max Preview",
11599
+ api: "anthropic-messages",
11600
+ provider: "vercel-ai-gateway",
11601
+ baseUrl: "https://ai-gateway.vercel.sh",
11602
+ reasoning: true,
11603
+ input: ["text", "image"],
11604
+ cost: {
11605
+ input: 1.3,
11606
+ output: 7.8,
11607
+ cacheRead: 0.26,
11608
+ cacheWrite: 1.625,
11609
+ },
11610
+ contextWindow: 240000,
11611
+ maxTokens: 64000,
11612
+ },
11219
11613
  "alibaba/qwen3-235b-a22b-thinking": {
11220
11614
  id: "alibaba/qwen3-235b-a22b-thinking",
11221
11615
  name: "Qwen3 235B A22B Thinking 2507",