@caupulican/pi-ai 0.81.39 → 0.81.41

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 (126) hide show
  1. package/dist/env-api-keys.d.ts +5 -0
  2. package/dist/env-api-keys.d.ts.map +1 -1
  3. package/dist/env-api-keys.js +20 -4
  4. package/dist/env-api-keys.js.map +1 -1
  5. package/dist/image-models.generated.d.ts +15 -0
  6. package/dist/image-models.generated.d.ts.map +1 -1
  7. package/dist/image-models.generated.js +15 -0
  8. package/dist/image-models.generated.js.map +1 -1
  9. package/dist/index.d.ts +3 -0
  10. package/dist/index.d.ts.map +1 -1
  11. package/dist/index.js +3 -0
  12. package/dist/index.js.map +1 -1
  13. package/dist/models.d.ts +7 -0
  14. package/dist/models.d.ts.map +1 -1
  15. package/dist/models.generated.d.ts +1847 -423
  16. package/dist/models.generated.d.ts.map +1 -1
  17. package/dist/models.generated.js +1511 -500
  18. package/dist/models.generated.js.map +1 -1
  19. package/dist/models.js +9 -0
  20. package/dist/models.js.map +1 -1
  21. package/dist/providers/amazon-bedrock.d.ts.map +1 -1
  22. package/dist/providers/amazon-bedrock.js +135 -79
  23. package/dist/providers/amazon-bedrock.js.map +1 -1
  24. package/dist/providers/anthropic.d.ts.map +1 -1
  25. package/dist/providers/anthropic.js +79 -42
  26. package/dist/providers/anthropic.js.map +1 -1
  27. package/dist/providers/azure-openai-responses.d.ts.map +1 -1
  28. package/dist/providers/azure-openai-responses.js +13 -6
  29. package/dist/providers/azure-openai-responses.js.map +1 -1
  30. package/dist/providers/bedrock-sso.d.ts +6 -0
  31. package/dist/providers/bedrock-sso.d.ts.map +1 -0
  32. package/dist/providers/bedrock-sso.js +39 -0
  33. package/dist/providers/bedrock-sso.js.map +1 -0
  34. package/dist/providers/faux.d.ts +1 -0
  35. package/dist/providers/faux.d.ts.map +1 -1
  36. package/dist/providers/faux.js +1 -0
  37. package/dist/providers/faux.js.map +1 -1
  38. package/dist/providers/images/openrouter.d.ts.map +1 -1
  39. package/dist/providers/images/openrouter.js +8 -3
  40. package/dist/providers/images/openrouter.js.map +1 -1
  41. package/dist/providers/mistral.d.ts.map +1 -1
  42. package/dist/providers/mistral.js +20 -15
  43. package/dist/providers/mistral.js.map +1 -1
  44. package/dist/providers/openai-codex-account.d.ts +32 -0
  45. package/dist/providers/openai-codex-account.d.ts.map +1 -0
  46. package/dist/providers/openai-codex-account.js +157 -0
  47. package/dist/providers/openai-codex-account.js.map +1 -0
  48. package/dist/providers/openai-codex-auth.d.ts +11 -0
  49. package/dist/providers/openai-codex-auth.d.ts.map +1 -0
  50. package/dist/providers/openai-codex-auth.js +43 -0
  51. package/dist/providers/openai-codex-auth.js.map +1 -0
  52. package/dist/providers/openai-codex-responses.d.ts +1 -0
  53. package/dist/providers/openai-codex-responses.d.ts.map +1 -1
  54. package/dist/providers/openai-codex-responses.js +97 -72
  55. package/dist/providers/openai-codex-responses.js.map +1 -1
  56. package/dist/providers/openai-completions.d.ts.map +1 -1
  57. package/dist/providers/openai-completions.js +35 -15
  58. package/dist/providers/openai-completions.js.map +1 -1
  59. package/dist/providers/openai-responses-shared.d.ts +5 -0
  60. package/dist/providers/openai-responses-shared.d.ts.map +1 -1
  61. package/dist/providers/openai-responses-shared.js +42 -5
  62. package/dist/providers/openai-responses-shared.js.map +1 -1
  63. package/dist/providers/openai-responses.d.ts +1 -0
  64. package/dist/providers/openai-responses.d.ts.map +1 -1
  65. package/dist/providers/openai-responses.js +28 -9
  66. package/dist/providers/openai-responses.js.map +1 -1
  67. package/dist/providers/openrouter-cache.d.ts +3 -0
  68. package/dist/providers/openrouter-cache.d.ts.map +1 -0
  69. package/dist/providers/openrouter-cache.js +4 -0
  70. package/dist/providers/openrouter-cache.js.map +1 -0
  71. package/dist/providers/simple-options.d.ts.map +1 -1
  72. package/dist/providers/simple-options.js +2 -0
  73. package/dist/providers/simple-options.js.map +1 -1
  74. package/dist/stream.d.ts.map +1 -1
  75. package/dist/stream.js +12 -4
  76. package/dist/stream.js.map +1 -1
  77. package/dist/types.d.ts +27 -4
  78. package/dist/types.d.ts.map +1 -1
  79. package/dist/types.js.map +1 -1
  80. package/dist/utils/abort-signals.d.ts +1 -0
  81. package/dist/utils/abort-signals.d.ts.map +1 -1
  82. package/dist/utils/abort-signals.js +7 -2
  83. package/dist/utils/abort-signals.js.map +1 -1
  84. package/dist/utils/error-body.d.ts.map +1 -1
  85. package/dist/utils/error-body.js +8 -3
  86. package/dist/utils/error-body.js.map +1 -1
  87. package/dist/utils/oauth/index.d.ts +3 -0
  88. package/dist/utils/oauth/index.d.ts.map +1 -1
  89. package/dist/utils/oauth/index.js +9 -0
  90. package/dist/utils/oauth/index.js.map +1 -1
  91. package/dist/utils/oauth/kimi-coding.d.ts +5 -0
  92. package/dist/utils/oauth/kimi-coding.d.ts.map +1 -0
  93. package/dist/utils/oauth/kimi-coding.js +173 -0
  94. package/dist/utils/oauth/kimi-coding.js.map +1 -0
  95. package/dist/utils/oauth/openai-codex.d.ts.map +1 -1
  96. package/dist/utils/oauth/openai-codex.js +2 -21
  97. package/dist/utils/oauth/openai-codex.js.map +1 -1
  98. package/dist/utils/oauth/openrouter.d.ts +5 -0
  99. package/dist/utils/oauth/openrouter.d.ts.map +1 -0
  100. package/dist/utils/oauth/openrouter.js +197 -0
  101. package/dist/utils/oauth/openrouter.js.map +1 -0
  102. package/dist/utils/oauth/xai.d.ts +6 -0
  103. package/dist/utils/oauth/xai.d.ts.map +1 -0
  104. package/dist/utils/oauth/xai.js +154 -0
  105. package/dist/utils/oauth/xai.js.map +1 -0
  106. package/dist/utils/overflow.d.ts +1 -0
  107. package/dist/utils/overflow.d.ts.map +1 -1
  108. package/dist/utils/overflow.js +3 -0
  109. package/dist/utils/overflow.js.map +1 -1
  110. package/dist/utils/provider-retry.d.ts +11 -0
  111. package/dist/utils/provider-retry.d.ts.map +1 -0
  112. package/dist/utils/provider-retry.js +66 -0
  113. package/dist/utils/provider-retry.js.map +1 -0
  114. package/dist/utils/tool-names.d.ts +6 -1
  115. package/dist/utils/tool-names.d.ts.map +1 -1
  116. package/dist/utils/tool-names.js +22 -9
  117. package/dist/utils/tool-names.js.map +1 -1
  118. package/dist/utils/tool-repair/text-protocol.d.ts +3 -0
  119. package/dist/utils/tool-repair/text-protocol.d.ts.map +1 -1
  120. package/dist/utils/tool-repair/text-protocol.js +10 -0
  121. package/dist/utils/tool-repair/text-protocol.js.map +1 -1
  122. package/dist/utils/uuid.d.ts +3 -0
  123. package/dist/utils/uuid.d.ts.map +1 -0
  124. package/dist/utils/uuid.js +48 -0
  125. package/dist/utils/uuid.js.map +1 -0
  126. package/package.json +2 -2
@@ -80,6 +80,7 @@ export const MODELS = {
80
80
  provider: "amazon-bedrock",
81
81
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
82
82
  reasoning: true,
83
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
83
84
  input: ["text", "image"],
84
85
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
85
86
  cost: {
@@ -152,7 +153,7 @@ export const MODELS = {
152
153
  provider: "amazon-bedrock",
153
154
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
154
155
  reasoning: true,
155
- thinkingLevelMap: { "xhigh": "max" },
156
+ thinkingLevelMap: { "max": "max" },
156
157
  input: ["text", "image"],
157
158
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
158
159
  cost: {
@@ -171,7 +172,7 @@ export const MODELS = {
171
172
  provider: "amazon-bedrock",
172
173
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
173
174
  reasoning: true,
174
- thinkingLevelMap: { "xhigh": "xhigh" },
175
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
175
176
  input: ["text", "image"],
176
177
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
177
178
  cost: {
@@ -190,7 +191,7 @@ export const MODELS = {
190
191
  provider: "amazon-bedrock",
191
192
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
192
193
  reasoning: true,
193
- thinkingLevelMap: { "xhigh": "xhigh" },
194
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
194
195
  input: ["text", "image"],
195
196
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
196
197
  cost: {
@@ -227,6 +228,7 @@ export const MODELS = {
227
228
  provider: "amazon-bedrock",
228
229
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
229
230
  reasoning: true,
231
+ thinkingLevelMap: { "max": "max" },
230
232
  input: ["text", "image"],
231
233
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
232
234
  cost: {
@@ -245,6 +247,7 @@ export const MODELS = {
245
247
  provider: "amazon-bedrock",
246
248
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
247
249
  reasoning: true,
250
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
248
251
  input: ["text", "image"],
249
252
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
250
253
  cost: {
@@ -281,7 +284,7 @@ export const MODELS = {
281
284
  provider: "amazon-bedrock",
282
285
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
283
286
  reasoning: true,
284
- thinkingLevelMap: { "xhigh": "max" },
287
+ thinkingLevelMap: { "max": "max" },
285
288
  input: ["text", "image"],
286
289
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
287
290
  cost: {
@@ -300,7 +303,26 @@ export const MODELS = {
300
303
  provider: "amazon-bedrock",
301
304
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
302
305
  reasoning: true,
303
- thinkingLevelMap: { "xhigh": "xhigh" },
306
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
307
+ input: ["text", "image"],
308
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
309
+ cost: {
310
+ input: 5,
311
+ output: 25,
312
+ cacheRead: 0.5,
313
+ cacheWrite: 6.25,
314
+ },
315
+ contextWindow: 1000000,
316
+ maxTokens: 128000,
317
+ },
318
+ "au.anthropic.claude-opus-5": {
319
+ id: "au.anthropic.claude-opus-5",
320
+ name: "Claude Opus 5 (AU)",
321
+ api: "bedrock-converse-stream",
322
+ provider: "amazon-bedrock",
323
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
324
+ reasoning: true,
325
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
304
326
  input: ["text", "image"],
305
327
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
306
328
  cost: {
@@ -337,6 +359,7 @@ export const MODELS = {
337
359
  provider: "amazon-bedrock",
338
360
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
339
361
  reasoning: true,
362
+ thinkingLevelMap: { "max": "max" },
340
363
  input: ["text", "image"],
341
364
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
342
365
  cost: {
@@ -355,6 +378,7 @@ export const MODELS = {
355
378
  provider: "amazon-bedrock",
356
379
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
357
380
  reasoning: true,
381
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
358
382
  input: ["text", "image"],
359
383
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
360
384
  cost: {
@@ -424,6 +448,7 @@ export const MODELS = {
424
448
  provider: "amazon-bedrock",
425
449
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
426
450
  reasoning: true,
451
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
427
452
  input: ["text", "image"],
428
453
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
429
454
  cost: {
@@ -478,7 +503,7 @@ export const MODELS = {
478
503
  provider: "amazon-bedrock",
479
504
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
480
505
  reasoning: true,
481
- thinkingLevelMap: { "xhigh": "max" },
506
+ thinkingLevelMap: { "max": "max" },
482
507
  input: ["text", "image"],
483
508
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
484
509
  cost: {
@@ -497,7 +522,7 @@ export const MODELS = {
497
522
  provider: "amazon-bedrock",
498
523
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
499
524
  reasoning: true,
500
- thinkingLevelMap: { "xhigh": "xhigh" },
525
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
501
526
  input: ["text", "image"],
502
527
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
503
528
  cost: {
@@ -516,7 +541,26 @@ export const MODELS = {
516
541
  provider: "amazon-bedrock",
517
542
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
518
543
  reasoning: true,
519
- thinkingLevelMap: { "xhigh": "xhigh" },
544
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
545
+ input: ["text", "image"],
546
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
547
+ cost: {
548
+ input: 5.5,
549
+ output: 27.5,
550
+ cacheRead: 0.55,
551
+ cacheWrite: 6.875,
552
+ },
553
+ contextWindow: 1000000,
554
+ maxTokens: 128000,
555
+ },
556
+ "eu.anthropic.claude-opus-5": {
557
+ id: "eu.anthropic.claude-opus-5",
558
+ name: "Claude Opus 5 (EU)",
559
+ api: "bedrock-converse-stream",
560
+ provider: "amazon-bedrock",
561
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
562
+ reasoning: true,
563
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
520
564
  input: ["text", "image"],
521
565
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
522
566
  cost: {
@@ -553,6 +597,7 @@ export const MODELS = {
553
597
  provider: "amazon-bedrock",
554
598
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
555
599
  reasoning: true,
600
+ thinkingLevelMap: { "max": "max" },
556
601
  input: ["text", "image"],
557
602
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
558
603
  cost: {
@@ -571,6 +616,7 @@ export const MODELS = {
571
616
  provider: "amazon-bedrock",
572
617
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
573
618
  reasoning: true,
619
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
574
620
  input: ["text", "image"],
575
621
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
576
622
  cost: {
@@ -589,6 +635,7 @@ export const MODELS = {
589
635
  provider: "amazon-bedrock",
590
636
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
591
637
  reasoning: true,
638
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
592
639
  input: ["text", "image"],
593
640
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
594
641
  cost: {
@@ -643,7 +690,7 @@ export const MODELS = {
643
690
  provider: "amazon-bedrock",
644
691
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
645
692
  reasoning: true,
646
- thinkingLevelMap: { "xhigh": "max" },
693
+ thinkingLevelMap: { "max": "max" },
647
694
  input: ["text", "image"],
648
695
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
649
696
  cost: {
@@ -662,7 +709,7 @@ export const MODELS = {
662
709
  provider: "amazon-bedrock",
663
710
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
664
711
  reasoning: true,
665
- thinkingLevelMap: { "xhigh": "xhigh" },
712
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
666
713
  input: ["text", "image"],
667
714
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
668
715
  cost: {
@@ -681,7 +728,26 @@ export const MODELS = {
681
728
  provider: "amazon-bedrock",
682
729
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
683
730
  reasoning: true,
684
- thinkingLevelMap: { "xhigh": "xhigh" },
731
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
732
+ input: ["text", "image"],
733
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
734
+ cost: {
735
+ input: 5,
736
+ output: 25,
737
+ cacheRead: 0.5,
738
+ cacheWrite: 6.25,
739
+ },
740
+ contextWindow: 1000000,
741
+ maxTokens: 128000,
742
+ },
743
+ "global.anthropic.claude-opus-5": {
744
+ id: "global.anthropic.claude-opus-5",
745
+ name: "Claude Opus 5 (Global)",
746
+ api: "bedrock-converse-stream",
747
+ provider: "amazon-bedrock",
748
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
749
+ reasoning: true,
750
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
685
751
  input: ["text", "image"],
686
752
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
687
753
  cost: {
@@ -718,6 +784,7 @@ export const MODELS = {
718
784
  provider: "amazon-bedrock",
719
785
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
720
786
  reasoning: true,
787
+ thinkingLevelMap: { "max": "max" },
721
788
  input: ["text", "image"],
722
789
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
723
790
  cost: {
@@ -736,6 +803,7 @@ export const MODELS = {
736
803
  provider: "amazon-bedrock",
737
804
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
738
805
  reasoning: true,
806
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
739
807
  input: ["text", "image"],
740
808
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
741
809
  cost: {
@@ -808,7 +876,7 @@ export const MODELS = {
808
876
  provider: "amazon-bedrock",
809
877
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
810
878
  reasoning: true,
811
- thinkingLevelMap: { "xhigh": "xhigh" },
879
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
812
880
  input: ["text", "image"],
813
881
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
814
882
  cost: {
@@ -827,7 +895,26 @@ export const MODELS = {
827
895
  provider: "amazon-bedrock",
828
896
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
829
897
  reasoning: true,
830
- thinkingLevelMap: { "xhigh": "xhigh" },
898
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
899
+ input: ["text", "image"],
900
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
901
+ cost: {
902
+ input: 5,
903
+ output: 25,
904
+ cacheRead: 0.5,
905
+ cacheWrite: 6.25,
906
+ },
907
+ contextWindow: 1000000,
908
+ maxTokens: 128000,
909
+ },
910
+ "jp.anthropic.claude-opus-5": {
911
+ id: "jp.anthropic.claude-opus-5",
912
+ name: "Claude Opus 5 (JP)",
913
+ api: "bedrock-converse-stream",
914
+ provider: "amazon-bedrock",
915
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
916
+ reasoning: true,
917
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
831
918
  input: ["text", "image"],
832
919
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
833
920
  cost: {
@@ -864,6 +951,7 @@ export const MODELS = {
864
951
  provider: "amazon-bedrock",
865
952
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
866
953
  reasoning: true,
954
+ thinkingLevelMap: { "max": "max" },
867
955
  input: ["text", "image"],
868
956
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
869
957
  cost: {
@@ -882,6 +970,7 @@ export const MODELS = {
882
970
  provider: "amazon-bedrock",
883
971
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
884
972
  reasoning: true,
973
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
885
974
  input: ["text", "image"],
886
975
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
887
976
  cost: {
@@ -1616,6 +1705,7 @@ export const MODELS = {
1616
1705
  provider: "amazon-bedrock",
1617
1706
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1618
1707
  reasoning: true,
1708
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
1619
1709
  input: ["text", "image"],
1620
1710
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
1621
1711
  cost: {
@@ -1688,7 +1778,7 @@ export const MODELS = {
1688
1778
  provider: "amazon-bedrock",
1689
1779
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1690
1780
  reasoning: true,
1691
- thinkingLevelMap: { "xhigh": "max" },
1781
+ thinkingLevelMap: { "max": "max" },
1692
1782
  input: ["text", "image"],
1693
1783
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
1694
1784
  cost: {
@@ -1707,7 +1797,7 @@ export const MODELS = {
1707
1797
  provider: "amazon-bedrock",
1708
1798
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1709
1799
  reasoning: true,
1710
- thinkingLevelMap: { "xhigh": "xhigh" },
1800
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
1711
1801
  input: ["text", "image"],
1712
1802
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
1713
1803
  cost: {
@@ -1726,7 +1816,26 @@ export const MODELS = {
1726
1816
  provider: "amazon-bedrock",
1727
1817
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1728
1818
  reasoning: true,
1729
- thinkingLevelMap: { "xhigh": "xhigh" },
1819
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
1820
+ input: ["text", "image"],
1821
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
1822
+ cost: {
1823
+ input: 5,
1824
+ output: 25,
1825
+ cacheRead: 0.5,
1826
+ cacheWrite: 6.25,
1827
+ },
1828
+ contextWindow: 1000000,
1829
+ maxTokens: 128000,
1830
+ },
1831
+ "us.anthropic.claude-opus-5": {
1832
+ id: "us.anthropic.claude-opus-5",
1833
+ name: "Claude Opus 5 (US)",
1834
+ api: "bedrock-converse-stream",
1835
+ provider: "amazon-bedrock",
1836
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1837
+ reasoning: true,
1838
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
1730
1839
  input: ["text", "image"],
1731
1840
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
1732
1841
  cost: {
@@ -1763,6 +1872,7 @@ export const MODELS = {
1763
1872
  provider: "amazon-bedrock",
1764
1873
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1765
1874
  reasoning: true,
1875
+ thinkingLevelMap: { "max": "max" },
1766
1876
  input: ["text", "image"],
1767
1877
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
1768
1878
  cost: {
@@ -1781,6 +1891,7 @@ export const MODELS = {
1781
1891
  provider: "amazon-bedrock",
1782
1892
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1783
1893
  reasoning: true,
1894
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
1784
1895
  input: ["text", "image"],
1785
1896
  supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
1786
1897
  cost: {
@@ -1956,7 +2067,9 @@ export const MODELS = {
1956
2067
  api: "anthropic-messages",
1957
2068
  provider: "anthropic",
1958
2069
  baseUrl: "https://api.anthropic.com",
2070
+ compat: { "forceAdaptiveThinking": true },
1959
2071
  reasoning: true,
2072
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
1960
2073
  input: ["text", "image"],
1961
2074
  cost: {
1962
2075
  input: 10,
@@ -2077,7 +2190,7 @@ export const MODELS = {
2077
2190
  baseUrl: "https://api.anthropic.com",
2078
2191
  compat: { "forceAdaptiveThinking": true },
2079
2192
  reasoning: true,
2080
- thinkingLevelMap: { "xhigh": "max" },
2193
+ thinkingLevelMap: { "max": "max" },
2081
2194
  input: ["text", "image"],
2082
2195
  cost: {
2083
2196
  input: 5,
@@ -2096,7 +2209,7 @@ export const MODELS = {
2096
2209
  baseUrl: "https://api.anthropic.com",
2097
2210
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
2098
2211
  reasoning: true,
2099
- thinkingLevelMap: { "xhigh": "xhigh" },
2212
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
2100
2213
  input: ["text", "image"],
2101
2214
  cost: {
2102
2215
  input: 5,
@@ -2115,7 +2228,26 @@ export const MODELS = {
2115
2228
  baseUrl: "https://api.anthropic.com",
2116
2229
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
2117
2230
  reasoning: true,
2118
- thinkingLevelMap: { "xhigh": "xhigh" },
2231
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
2232
+ input: ["text", "image"],
2233
+ cost: {
2234
+ input: 5,
2235
+ output: 25,
2236
+ cacheRead: 0.5,
2237
+ cacheWrite: 6.25,
2238
+ },
2239
+ contextWindow: 1000000,
2240
+ maxTokens: 128000,
2241
+ },
2242
+ "claude-opus-5": {
2243
+ id: "claude-opus-5",
2244
+ name: "Claude Opus 5",
2245
+ api: "anthropic-messages",
2246
+ provider: "anthropic",
2247
+ baseUrl: "https://api.anthropic.com",
2248
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
2249
+ reasoning: true,
2250
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
2119
2251
  input: ["text", "image"],
2120
2252
  cost: {
2121
2253
  input: 5,
@@ -2168,6 +2300,7 @@ export const MODELS = {
2168
2300
  baseUrl: "https://api.anthropic.com",
2169
2301
  compat: { "forceAdaptiveThinking": true },
2170
2302
  reasoning: true,
2303
+ thinkingLevelMap: { "max": "max" },
2171
2304
  input: ["text", "image"],
2172
2305
  cost: {
2173
2306
  input: 3,
@@ -2184,7 +2317,9 @@ export const MODELS = {
2184
2317
  api: "anthropic-messages",
2185
2318
  provider: "anthropic",
2186
2319
  baseUrl: "https://api.anthropic.com",
2320
+ compat: { "forceAdaptiveThinking": true },
2187
2321
  reasoning: true,
2322
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
2188
2323
  input: ["text", "image"],
2189
2324
  cost: {
2190
2325
  input: 2,
@@ -2403,24 +2538,6 @@ export const MODELS = {
2403
2538
  contextWindow: 128000,
2404
2539
  maxTokens: 16384,
2405
2540
  },
2406
- "gpt-5-codex": {
2407
- id: "gpt-5-codex",
2408
- name: "GPT-5-Codex",
2409
- api: "azure-openai-responses",
2410
- provider: "azure-openai-responses",
2411
- baseUrl: "",
2412
- reasoning: true,
2413
- thinkingLevelMap: { "off": null },
2414
- input: ["text", "image"],
2415
- cost: {
2416
- input: 1.25,
2417
- output: 10,
2418
- cacheRead: 0.125,
2419
- cacheWrite: 0,
2420
- },
2421
- contextWindow: 400000,
2422
- maxTokens: 128000,
2423
- },
2424
2541
  "gpt-5-mini": {
2425
2542
  id: "gpt-5-mini",
2426
2543
  name: "GPT-5 Mini",
@@ -2493,24 +2610,6 @@ export const MODELS = {
2493
2610
  contextWindow: 400000,
2494
2611
  maxTokens: 128000,
2495
2612
  },
2496
- "gpt-5.1-chat-latest": {
2497
- id: "gpt-5.1-chat-latest",
2498
- name: "GPT-5.1 Chat",
2499
- api: "azure-openai-responses",
2500
- provider: "azure-openai-responses",
2501
- baseUrl: "",
2502
- reasoning: true,
2503
- thinkingLevelMap: { "off": null },
2504
- input: ["text", "image"],
2505
- cost: {
2506
- input: 1.25,
2507
- output: 10,
2508
- cacheRead: 0.125,
2509
- cacheWrite: 0,
2510
- },
2511
- contextWindow: 128000,
2512
- maxTokens: 16384,
2513
- },
2514
2613
  "gpt-5.1-codex": {
2515
2614
  id: "gpt-5.1-codex",
2516
2615
  name: "GPT-5.1 Codex",
@@ -2522,9 +2621,9 @@ export const MODELS = {
2522
2621
  input: ["text", "image"],
2523
2622
  cost: {
2524
2623
  input: 1.25,
2525
- output: 10,
2624
+ output: 5,
2526
2625
  cacheRead: 0.125,
2527
- cacheWrite: 0,
2626
+ cacheWrite: 1.25,
2528
2627
  },
2529
2628
  contextWindow: 400000,
2530
2629
  maxTokens: 128000,
@@ -2547,24 +2646,6 @@ export const MODELS = {
2547
2646
  contextWindow: 400000,
2548
2647
  maxTokens: 128000,
2549
2648
  },
2550
- "gpt-5.1-codex-mini": {
2551
- id: "gpt-5.1-codex-mini",
2552
- name: "GPT-5.1 Codex mini",
2553
- api: "azure-openai-responses",
2554
- provider: "azure-openai-responses",
2555
- baseUrl: "",
2556
- reasoning: true,
2557
- thinkingLevelMap: { "off": null },
2558
- input: ["text", "image"],
2559
- cost: {
2560
- input: 0.25,
2561
- output: 2,
2562
- cacheRead: 0.025,
2563
- cacheWrite: 0,
2564
- },
2565
- contextWindow: 400000,
2566
- maxTokens: 128000,
2567
- },
2568
2649
  "gpt-5.2": {
2569
2650
  id: "gpt-5.2",
2570
2651
  name: "GPT-5.2",
@@ -2601,24 +2682,6 @@ export const MODELS = {
2601
2682
  contextWindow: 128000,
2602
2683
  maxTokens: 16384,
2603
2684
  },
2604
- "gpt-5.2-codex": {
2605
- id: "gpt-5.2-codex",
2606
- name: "GPT-5.2 Codex",
2607
- api: "azure-openai-responses",
2608
- provider: "azure-openai-responses",
2609
- baseUrl: "",
2610
- reasoning: true,
2611
- thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
2612
- input: ["text", "image"],
2613
- cost: {
2614
- input: 1.75,
2615
- output: 14,
2616
- cacheRead: 0.175,
2617
- cacheWrite: 0,
2618
- },
2619
- contextWindow: 400000,
2620
- maxTokens: 128000,
2621
- },
2622
2685
  "gpt-5.2-pro": {
2623
2686
  id: "gpt-5.2-pro",
2624
2687
  name: "GPT-5.2 Pro",
@@ -2867,23 +2930,6 @@ export const MODELS = {
2867
2930
  contextWindow: 200000,
2868
2931
  maxTokens: 100000,
2869
2932
  },
2870
- "o3-deep-research": {
2871
- id: "o3-deep-research",
2872
- name: "o3-deep-research",
2873
- api: "azure-openai-responses",
2874
- provider: "azure-openai-responses",
2875
- baseUrl: "",
2876
- reasoning: true,
2877
- input: ["text", "image"],
2878
- cost: {
2879
- input: 10,
2880
- output: 40,
2881
- cacheRead: 2.5,
2882
- cacheWrite: 0,
2883
- },
2884
- contextWindow: 200000,
2885
- maxTokens: 100000,
2886
- },
2887
2933
  "o3-mini": {
2888
2934
  id: "o3-mini",
2889
2935
  name: "o3-mini",
@@ -2935,23 +2981,6 @@ export const MODELS = {
2935
2981
  contextWindow: 200000,
2936
2982
  maxTokens: 100000,
2937
2983
  },
2938
- "o4-mini-deep-research": {
2939
- id: "o4-mini-deep-research",
2940
- name: "o4-mini-deep-research",
2941
- api: "azure-openai-responses",
2942
- provider: "azure-openai-responses",
2943
- baseUrl: "",
2944
- reasoning: true,
2945
- input: ["text", "image"],
2946
- cost: {
2947
- input: 2,
2948
- output: 8,
2949
- cacheRead: 0.5,
2950
- cacheWrite: 0,
2951
- },
2952
- contextWindow: 200000,
2953
- maxTokens: 100000,
2954
- },
2955
2984
  },
2956
2985
  "cerebras": {
2957
2986
  "gemma-4-31b": {
@@ -3115,7 +3144,9 @@ export const MODELS = {
3115
3144
  api: "anthropic-messages",
3116
3145
  provider: "cloudflare-ai-gateway",
3117
3146
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3147
+ compat: { "forceAdaptiveThinking": true },
3118
3148
  reasoning: true,
3149
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
3119
3150
  input: ["text", "image"],
3120
3151
  cost: {
3121
3152
  input: 10,
@@ -3202,7 +3233,7 @@ export const MODELS = {
3202
3233
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3203
3234
  compat: { "forceAdaptiveThinking": true },
3204
3235
  reasoning: true,
3205
- thinkingLevelMap: { "xhigh": "max" },
3236
+ thinkingLevelMap: { "max": "max" },
3206
3237
  input: ["text", "image"],
3207
3238
  cost: {
3208
3239
  input: 5,
@@ -3221,7 +3252,7 @@ export const MODELS = {
3221
3252
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3222
3253
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
3223
3254
  reasoning: true,
3224
- thinkingLevelMap: { "xhigh": "xhigh" },
3255
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
3225
3256
  input: ["text", "image"],
3226
3257
  cost: {
3227
3258
  input: 5,
@@ -3240,7 +3271,7 @@ export const MODELS = {
3240
3271
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3241
3272
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
3242
3273
  reasoning: true,
3243
- thinkingLevelMap: { "xhigh": "xhigh" },
3274
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
3244
3275
  input: ["text", "image"],
3245
3276
  cost: {
3246
3277
  input: 5,
@@ -3251,19 +3282,38 @@ export const MODELS = {
3251
3282
  contextWindow: 1000000,
3252
3283
  maxTokens: 128000,
3253
3284
  },
3254
- "claude-sonnet-4": {
3255
- id: "claude-sonnet-4",
3256
- name: "Claude Sonnet 4 (latest)",
3285
+ "claude-opus-5": {
3286
+ id: "claude-opus-5",
3287
+ name: "Claude Opus 5",
3257
3288
  api: "anthropic-messages",
3258
3289
  provider: "cloudflare-ai-gateway",
3259
3290
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3291
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
3260
3292
  reasoning: true,
3293
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
3261
3294
  input: ["text", "image"],
3262
3295
  cost: {
3263
- input: 3,
3264
- output: 15,
3265
- cacheRead: 0.3,
3266
- cacheWrite: 3.75,
3296
+ input: 5,
3297
+ output: 25,
3298
+ cacheRead: 0.5,
3299
+ cacheWrite: 6.25,
3300
+ },
3301
+ contextWindow: 1000000,
3302
+ maxTokens: 128000,
3303
+ },
3304
+ "claude-sonnet-4": {
3305
+ id: "claude-sonnet-4",
3306
+ name: "Claude Sonnet 4 (latest)",
3307
+ api: "anthropic-messages",
3308
+ provider: "cloudflare-ai-gateway",
3309
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3310
+ reasoning: true,
3311
+ input: ["text", "image"],
3312
+ cost: {
3313
+ input: 3,
3314
+ output: 15,
3315
+ cacheRead: 0.3,
3316
+ cacheWrite: 3.75,
3267
3317
  },
3268
3318
  contextWindow: 200000,
3269
3319
  maxTokens: 64000,
@@ -3293,6 +3343,7 @@ export const MODELS = {
3293
3343
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3294
3344
  compat: { "forceAdaptiveThinking": true },
3295
3345
  reasoning: true,
3346
+ thinkingLevelMap: { "max": "max" },
3296
3347
  input: ["text", "image"],
3297
3348
  cost: {
3298
3349
  input: 3,
@@ -3309,7 +3360,9 @@ export const MODELS = {
3309
3360
  api: "anthropic-messages",
3310
3361
  provider: "cloudflare-ai-gateway",
3311
3362
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3363
+ compat: { "forceAdaptiveThinking": true },
3312
3364
  reasoning: true,
3365
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
3313
3366
  input: ["text", "image"],
3314
3367
  cost: {
3315
3368
  input: 2,
@@ -4191,7 +4244,7 @@ export const MODELS = {
4191
4244
  baseUrl: "https://api.fireworks.ai/inference",
4192
4245
  compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4193
4246
  reasoning: true,
4194
- input: ["text"],
4247
+ input: ["text", "image"],
4195
4248
  cost: {
4196
4249
  input: 0.3,
4197
4250
  output: 1.2,
@@ -4356,12 +4409,13 @@ export const MODELS = {
4356
4409
  "claude-fable-5": {
4357
4410
  id: "claude-fable-5",
4358
4411
  name: "Claude Fable 5",
4359
- api: "openai-completions",
4412
+ api: "anthropic-messages",
4360
4413
  provider: "github-copilot",
4361
4414
  baseUrl: "https://api.individual.githubcopilot.com",
4362
4415
  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" },
4363
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4416
+ compat: { "forceAdaptiveThinking": true },
4364
4417
  reasoning: true,
4418
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
4365
4419
  input: ["text", "image"],
4366
4420
  cost: {
4367
4421
  input: 10,
@@ -4418,7 +4472,7 @@ export const MODELS = {
4418
4472
  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" },
4419
4473
  compat: { "forceAdaptiveThinking": true },
4420
4474
  reasoning: true,
4421
- thinkingLevelMap: { "xhigh": "max" },
4475
+ thinkingLevelMap: { "max": "max" },
4422
4476
  input: ["text", "image"],
4423
4477
  cost: {
4424
4478
  input: 5,
@@ -4438,7 +4492,7 @@ export const MODELS = {
4438
4492
  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" },
4439
4493
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
4440
4494
  reasoning: true,
4441
- thinkingLevelMap: { "xhigh": "xhigh" },
4495
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
4442
4496
  input: ["text", "image"],
4443
4497
  cost: {
4444
4498
  input: 5,
@@ -4458,7 +4512,7 @@ export const MODELS = {
4458
4512
  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" },
4459
4513
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
4460
4514
  reasoning: true,
4461
- thinkingLevelMap: { "xhigh": "xhigh" },
4515
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
4462
4516
  input: ["text", "image"],
4463
4517
  cost: {
4464
4518
  input: 5,
@@ -4469,6 +4523,26 @@ export const MODELS = {
4469
4523
  contextWindow: 200000,
4470
4524
  maxTokens: 64000,
4471
4525
  },
4526
+ "claude-opus-5": {
4527
+ id: "claude-opus-5",
4528
+ name: "Claude Opus 5",
4529
+ api: "anthropic-messages",
4530
+ provider: "github-copilot",
4531
+ baseUrl: "https://api.individual.githubcopilot.com",
4532
+ 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" },
4533
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
4534
+ reasoning: true,
4535
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
4536
+ input: ["text", "image"],
4537
+ cost: {
4538
+ input: 5,
4539
+ output: 25,
4540
+ cacheRead: 0.5,
4541
+ cacheWrite: 6.25,
4542
+ },
4543
+ contextWindow: 1000000,
4544
+ maxTokens: 64000,
4545
+ },
4472
4546
  "claude-sonnet-4": {
4473
4547
  id: "claude-sonnet-4",
4474
4548
  name: "Claude Sonnet 4 (latest)",
@@ -4516,6 +4590,7 @@ export const MODELS = {
4516
4590
  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" },
4517
4591
  compat: { "forceAdaptiveThinking": true },
4518
4592
  reasoning: true,
4593
+ thinkingLevelMap: { "max": "max" },
4519
4594
  input: ["text", "image"],
4520
4595
  cost: {
4521
4596
  input: 3,
@@ -4529,12 +4604,13 @@ export const MODELS = {
4529
4604
  "claude-sonnet-5": {
4530
4605
  id: "claude-sonnet-5",
4531
4606
  name: "Claude Sonnet 5",
4532
- api: "openai-completions",
4607
+ api: "anthropic-messages",
4533
4608
  provider: "github-copilot",
4534
4609
  baseUrl: "https://api.individual.githubcopilot.com",
4535
4610
  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" },
4536
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4611
+ compat: { "forceAdaptiveThinking": true },
4537
4612
  reasoning: true,
4613
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
4538
4614
  input: ["text", "image"],
4539
4615
  cost: {
4540
4616
  input: 2,
@@ -4871,11 +4947,10 @@ export const MODELS = {
4871
4947
  "mai-code-1-flash-picker": {
4872
4948
  id: "mai-code-1-flash-picker",
4873
4949
  name: "MAI-Code-1-Flash",
4874
- api: "openai-completions",
4950
+ api: "openai-responses",
4875
4951
  provider: "github-copilot",
4876
4952
  baseUrl: "https://api.individual.githubcopilot.com",
4877
4953
  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" },
4878
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4879
4954
  reasoning: true,
4880
4955
  input: ["text"],
4881
4956
  cost: {
@@ -4889,6 +4964,40 @@ export const MODELS = {
4889
4964
  },
4890
4965
  },
4891
4966
  "google": {
4967
+ "deep-research-max-preview-04-2026": {
4968
+ id: "deep-research-max-preview-04-2026",
4969
+ name: "Deep Research Max Preview (Apr-21-2026)",
4970
+ api: "google-generative-ai",
4971
+ provider: "google",
4972
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4973
+ reasoning: true,
4974
+ input: ["text", "image"],
4975
+ cost: {
4976
+ input: 2,
4977
+ output: 12,
4978
+ cacheRead: 0.2,
4979
+ cacheWrite: 0,
4980
+ },
4981
+ contextWindow: 131072,
4982
+ maxTokens: 65536,
4983
+ },
4984
+ "deep-research-preview-04-2026": {
4985
+ id: "deep-research-preview-04-2026",
4986
+ name: "Deep Research Preview (Apr-21-2026)",
4987
+ api: "google-generative-ai",
4988
+ provider: "google",
4989
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4990
+ reasoning: true,
4991
+ input: ["text", "image"],
4992
+ cost: {
4993
+ input: 2,
4994
+ output: 12,
4995
+ cacheRead: 0.2,
4996
+ cacheWrite: 0,
4997
+ },
4998
+ contextWindow: 131072,
4999
+ maxTokens: 65536,
5000
+ },
4892
5001
  "gemini-2.0-flash": {
4893
5002
  id: "gemini-2.0-flash",
4894
5003
  name: "Gemini 2.0 Flash",
@@ -4923,6 +5032,23 @@ export const MODELS = {
4923
5032
  contextWindow: 1048576,
4924
5033
  maxTokens: 8192,
4925
5034
  },
5035
+ "gemini-2.5-computer-use-preview-10-2025": {
5036
+ id: "gemini-2.5-computer-use-preview-10-2025",
5037
+ name: "Gemini 2.5 Computer Use Preview 10-2025",
5038
+ api: "google-generative-ai",
5039
+ provider: "google",
5040
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5041
+ reasoning: true,
5042
+ input: ["text", "image"],
5043
+ cost: {
5044
+ input: 1.25,
5045
+ output: 10,
5046
+ cacheRead: 0,
5047
+ cacheWrite: 0,
5048
+ },
5049
+ contextWindow: 131072,
5050
+ maxTokens: 65536,
5051
+ },
4926
5052
  "gemini-2.5-flash": {
4927
5053
  id: "gemini-2.5-flash",
4928
5054
  name: "Gemini 2.5 Flash",
@@ -5028,6 +5154,24 @@ export const MODELS = {
5028
5154
  contextWindow: 1048576,
5029
5155
  maxTokens: 65536,
5030
5156
  },
5157
+ "gemini-3.1-flash-lite-image": {
5158
+ id: "gemini-3.1-flash-lite-image",
5159
+ name: "Nano Banana 2 Lite",
5160
+ api: "google-generative-ai",
5161
+ provider: "google",
5162
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5163
+ reasoning: true,
5164
+ thinkingLevelMap: { "off": null },
5165
+ input: ["text", "image"],
5166
+ cost: {
5167
+ input: 0.25,
5168
+ output: 30,
5169
+ cacheRead: 0,
5170
+ cacheWrite: 0,
5171
+ },
5172
+ contextWindow: 65536,
5173
+ maxTokens: 65536,
5174
+ },
5031
5175
  "gemini-3.1-flash-lite-preview": {
5032
5176
  id: "gemini-3.1-flash-lite-preview",
5033
5177
  name: "Gemini 3.1 Flash Lite Preview",
@@ -5046,6 +5190,24 @@ export const MODELS = {
5046
5190
  contextWindow: 1048576,
5047
5191
  maxTokens: 65536,
5048
5192
  },
5193
+ "gemini-3.1-flash-live-preview": {
5194
+ id: "gemini-3.1-flash-live-preview",
5195
+ name: "Gemini 3.1 Flash Live Preview",
5196
+ api: "google-generative-ai",
5197
+ provider: "google",
5198
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5199
+ reasoning: true,
5200
+ thinkingLevelMap: { "off": null },
5201
+ input: ["text", "image"],
5202
+ cost: {
5203
+ input: 0.75,
5204
+ output: 4.5,
5205
+ cacheRead: 0,
5206
+ cacheWrite: 0,
5207
+ },
5208
+ contextWindow: 131072,
5209
+ maxTokens: 65536,
5210
+ },
5049
5211
  "gemini-3.1-pro-preview": {
5050
5212
  id: "gemini-3.1-pro-preview",
5051
5213
  name: "Gemini 3.1 Pro Preview",
@@ -5100,6 +5262,42 @@ export const MODELS = {
5100
5262
  contextWindow: 1048576,
5101
5263
  maxTokens: 65536,
5102
5264
  },
5265
+ "gemini-3.5-flash-lite": {
5266
+ id: "gemini-3.5-flash-lite",
5267
+ name: "Gemini 3.5 Flash Lite",
5268
+ api: "google-generative-ai",
5269
+ provider: "google",
5270
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5271
+ reasoning: true,
5272
+ thinkingLevelMap: { "off": null },
5273
+ input: ["text", "image"],
5274
+ cost: {
5275
+ input: 0.3,
5276
+ output: 2.5,
5277
+ cacheRead: 0.03,
5278
+ cacheWrite: 0,
5279
+ },
5280
+ contextWindow: 1048576,
5281
+ maxTokens: 65536,
5282
+ },
5283
+ "gemini-3.6-flash": {
5284
+ id: "gemini-3.6-flash",
5285
+ name: "Gemini 3.6 Flash",
5286
+ api: "google-generative-ai",
5287
+ provider: "google",
5288
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5289
+ reasoning: true,
5290
+ thinkingLevelMap: { "off": null },
5291
+ input: ["text", "image"],
5292
+ cost: {
5293
+ input: 1.5,
5294
+ output: 7.5,
5295
+ cacheRead: 0.15,
5296
+ cacheWrite: 0,
5297
+ },
5298
+ contextWindow: 1048576,
5299
+ maxTokens: 65536,
5300
+ },
5103
5301
  "gemini-flash-latest": {
5104
5302
  id: "gemini-flash-latest",
5105
5303
  name: "Gemini Flash Latest",
@@ -5134,6 +5332,23 @@ export const MODELS = {
5134
5332
  contextWindow: 1048576,
5135
5333
  maxTokens: 65536,
5136
5334
  },
5335
+ "gemini-robotics-er-1.6-preview": {
5336
+ id: "gemini-robotics-er-1.6-preview",
5337
+ name: "Gemini Robotics-ER 1.6 Preview",
5338
+ api: "google-generative-ai",
5339
+ provider: "google",
5340
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
5341
+ reasoning: true,
5342
+ input: ["text", "image"],
5343
+ cost: {
5344
+ input: 1,
5345
+ output: 5,
5346
+ cacheRead: 0,
5347
+ cacheWrite: 0,
5348
+ },
5349
+ contextWindow: 131072,
5350
+ maxTokens: 65536,
5351
+ },
5137
5352
  "gemma-4-26b-a4b-it": {
5138
5353
  id: "gemma-4-26b-a4b-it",
5139
5354
  name: "Gemma 4 26B A4B IT",
@@ -5899,6 +6114,24 @@ export const MODELS = {
5899
6114
  contextWindow: 262144,
5900
6115
  maxTokens: 4096,
5901
6116
  },
6117
+ "XiaomiMiMo/MiMo-V2.5": {
6118
+ id: "XiaomiMiMo/MiMo-V2.5",
6119
+ name: "MiMo-V2.5",
6120
+ api: "openai-completions",
6121
+ provider: "huggingface",
6122
+ baseUrl: "https://router.huggingface.co/v1",
6123
+ compat: { "supportsDeveloperRole": false },
6124
+ reasoning: true,
6125
+ input: ["text"],
6126
+ cost: {
6127
+ input: 0.4,
6128
+ output: 2,
6129
+ cacheRead: 0,
6130
+ cacheWrite: 0,
6131
+ },
6132
+ contextWindow: 262144,
6133
+ maxTokens: 131072,
6134
+ },
5902
6135
  "XiaomiMiMo/MiMo-V2.5-Pro": {
5903
6136
  id: "XiaomiMiMo/MiMo-V2.5-Pro",
5904
6137
  name: "MiMo-V2.5-Pro",
@@ -6412,6 +6645,7 @@ export const MODELS = {
6412
6645
  provider: "kimi-coding",
6413
6646
  baseUrl: "https://api.kimi.com/coding",
6414
6647
  headers: { "User-Agent": "KimiCLI/1.5" },
6648
+ compat: { "authFormat": "bearer" },
6415
6649
  reasoning: true,
6416
6650
  input: ["text", "image"],
6417
6651
  cost: {
@@ -6423,6 +6657,25 @@ export const MODELS = {
6423
6657
  contextWindow: 1048576,
6424
6658
  maxTokens: 131072,
6425
6659
  },
6660
+ "k3-256k": {
6661
+ id: "k3-256k",
6662
+ name: "Kimi K3-256K",
6663
+ api: "anthropic-messages",
6664
+ provider: "kimi-coding",
6665
+ baseUrl: "https://api.kimi.com/coding",
6666
+ headers: { "User-Agent": "KimiCLI/1.5" },
6667
+ compat: { "authFormat": "bearer" },
6668
+ reasoning: true,
6669
+ input: ["text", "image"],
6670
+ cost: {
6671
+ input: 0,
6672
+ output: 0,
6673
+ cacheRead: 0,
6674
+ cacheWrite: 0,
6675
+ },
6676
+ contextWindow: 262144,
6677
+ maxTokens: 131072,
6678
+ },
6426
6679
  "kimi-for-coding": {
6427
6680
  id: "kimi-for-coding",
6428
6681
  name: "Kimi K2.7 Code",
@@ -6430,6 +6683,7 @@ export const MODELS = {
6430
6683
  provider: "kimi-coding",
6431
6684
  baseUrl: "https://api.kimi.com/coding",
6432
6685
  headers: { "User-Agent": "KimiCLI/1.5" },
6686
+ compat: { "authFormat": "bearer" },
6433
6687
  reasoning: true,
6434
6688
  input: ["text", "image"],
6435
6689
  cost: {
@@ -6448,6 +6702,7 @@ export const MODELS = {
6448
6702
  provider: "kimi-coding",
6449
6703
  baseUrl: "https://api.kimi.com/coding",
6450
6704
  headers: { "User-Agent": "KimiCLI/1.5" },
6705
+ compat: { "authFormat": "bearer" },
6451
6706
  reasoning: true,
6452
6707
  input: ["text", "image"],
6453
6708
  cost: {
@@ -7635,24 +7890,6 @@ export const MODELS = {
7635
7890
  contextWindow: 128000,
7636
7891
  maxTokens: 16384,
7637
7892
  },
7638
- "gpt-5-codex": {
7639
- id: "gpt-5-codex",
7640
- name: "GPT-5-Codex",
7641
- api: "openai-responses",
7642
- provider: "openai",
7643
- baseUrl: "https://api.openai.com/v1",
7644
- reasoning: true,
7645
- thinkingLevelMap: { "off": null },
7646
- input: ["text", "image"],
7647
- cost: {
7648
- input: 1.25,
7649
- output: 10,
7650
- cacheRead: 0.125,
7651
- cacheWrite: 0,
7652
- },
7653
- contextWindow: 400000,
7654
- maxTokens: 128000,
7655
- },
7656
7893
  "gpt-5-mini": {
7657
7894
  id: "gpt-5-mini",
7658
7895
  name: "GPT-5 Mini",
@@ -7725,24 +7962,6 @@ export const MODELS = {
7725
7962
  contextWindow: 400000,
7726
7963
  maxTokens: 128000,
7727
7964
  },
7728
- "gpt-5.1-chat-latest": {
7729
- id: "gpt-5.1-chat-latest",
7730
- name: "GPT-5.1 Chat",
7731
- api: "openai-responses",
7732
- provider: "openai",
7733
- baseUrl: "https://api.openai.com/v1",
7734
- reasoning: true,
7735
- thinkingLevelMap: { "off": null },
7736
- input: ["text", "image"],
7737
- cost: {
7738
- input: 1.25,
7739
- output: 10,
7740
- cacheRead: 0.125,
7741
- cacheWrite: 0,
7742
- },
7743
- contextWindow: 128000,
7744
- maxTokens: 16384,
7745
- },
7746
7965
  "gpt-5.1-codex": {
7747
7966
  id: "gpt-5.1-codex",
7748
7967
  name: "GPT-5.1 Codex",
@@ -7754,9 +7973,9 @@ export const MODELS = {
7754
7973
  input: ["text", "image"],
7755
7974
  cost: {
7756
7975
  input: 1.25,
7757
- output: 10,
7976
+ output: 5,
7758
7977
  cacheRead: 0.125,
7759
- cacheWrite: 0,
7978
+ cacheWrite: 1.25,
7760
7979
  },
7761
7980
  contextWindow: 400000,
7762
7981
  maxTokens: 128000,
@@ -7779,24 +7998,6 @@ export const MODELS = {
7779
7998
  contextWindow: 400000,
7780
7999
  maxTokens: 128000,
7781
8000
  },
7782
- "gpt-5.1-codex-mini": {
7783
- id: "gpt-5.1-codex-mini",
7784
- name: "GPT-5.1 Codex mini",
7785
- api: "openai-responses",
7786
- provider: "openai",
7787
- baseUrl: "https://api.openai.com/v1",
7788
- reasoning: true,
7789
- thinkingLevelMap: { "off": null },
7790
- input: ["text", "image"],
7791
- cost: {
7792
- input: 0.25,
7793
- output: 2,
7794
- cacheRead: 0.025,
7795
- cacheWrite: 0,
7796
- },
7797
- contextWindow: 400000,
7798
- maxTokens: 128000,
7799
- },
7800
8001
  "gpt-5.2": {
7801
8002
  id: "gpt-5.2",
7802
8003
  name: "GPT-5.2",
@@ -7833,24 +8034,6 @@ export const MODELS = {
7833
8034
  contextWindow: 128000,
7834
8035
  maxTokens: 16384,
7835
8036
  },
7836
- "gpt-5.2-codex": {
7837
- id: "gpt-5.2-codex",
7838
- name: "GPT-5.2 Codex",
7839
- api: "openai-responses",
7840
- provider: "openai",
7841
- baseUrl: "https://api.openai.com/v1",
7842
- reasoning: true,
7843
- thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
7844
- input: ["text", "image"],
7845
- cost: {
7846
- input: 1.75,
7847
- output: 14,
7848
- cacheRead: 0.175,
7849
- cacheWrite: 0,
7850
- },
7851
- contextWindow: 400000,
7852
- maxTokens: 128000,
7853
- },
7854
8037
  "gpt-5.2-pro": {
7855
8038
  id: "gpt-5.2-pro",
7856
8039
  name: "GPT-5.2 Pro",
@@ -8179,23 +8362,6 @@ export const MODELS = {
8179
8362
  contextWindow: 200000,
8180
8363
  maxTokens: 100000,
8181
8364
  },
8182
- "o3-deep-research": {
8183
- id: "o3-deep-research",
8184
- name: "o3-deep-research",
8185
- api: "openai-responses",
8186
- provider: "openai",
8187
- baseUrl: "https://api.openai.com/v1",
8188
- reasoning: true,
8189
- input: ["text", "image"],
8190
- cost: {
8191
- input: 10,
8192
- output: 40,
8193
- cacheRead: 2.5,
8194
- cacheWrite: 0,
8195
- },
8196
- contextWindow: 200000,
8197
- maxTokens: 100000,
8198
- },
8199
8365
  "o3-mini": {
8200
8366
  id: "o3-mini",
8201
8367
  name: "o3-mini",
@@ -8247,23 +8413,6 @@ export const MODELS = {
8247
8413
  contextWindow: 200000,
8248
8414
  maxTokens: 100000,
8249
8415
  },
8250
- "o4-mini-deep-research": {
8251
- id: "o4-mini-deep-research",
8252
- name: "o4-mini-deep-research",
8253
- api: "openai-responses",
8254
- provider: "openai",
8255
- baseUrl: "https://api.openai.com/v1",
8256
- reasoning: true,
8257
- input: ["text", "image"],
8258
- cost: {
8259
- input: 2,
8260
- output: 8,
8261
- cacheRead: 0.5,
8262
- cacheWrite: 0,
8263
- },
8264
- contextWindow: 200000,
8265
- maxTokens: 100000,
8266
- },
8267
8416
  },
8268
8417
  "openai-codex": {
8269
8418
  "gpt-5.2": {
@@ -8391,7 +8540,7 @@ export const MODELS = {
8391
8540
  cacheRead: 0.1,
8392
8541
  cacheWrite: 1.25,
8393
8542
  },
8394
- contextWindow: 372000,
8543
+ contextWindow: 272000,
8395
8544
  maxTokens: 128000,
8396
8545
  },
8397
8546
  "gpt-5.6-sol": {
@@ -8411,7 +8560,7 @@ export const MODELS = {
8411
8560
  cacheRead: 0.5,
8412
8561
  cacheWrite: 6.25,
8413
8562
  },
8414
- contextWindow: 372000,
8563
+ contextWindow: 272000,
8415
8564
  maxTokens: 128000,
8416
8565
  },
8417
8566
  "gpt-5.6-terra": {
@@ -8431,7 +8580,7 @@ export const MODELS = {
8431
8580
  cacheRead: 0.25,
8432
8581
  cacheWrite: 3.125,
8433
8582
  },
8434
- contextWindow: 372000,
8583
+ contextWindow: 272000,
8435
8584
  maxTokens: 128000,
8436
8585
  },
8437
8586
  },
@@ -8459,7 +8608,9 @@ export const MODELS = {
8459
8608
  api: "anthropic-messages",
8460
8609
  provider: "opencode",
8461
8610
  baseUrl: "https://opencode.ai/zen",
8611
+ compat: { "forceAdaptiveThinking": true },
8462
8612
  reasoning: true,
8613
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
8463
8614
  input: ["text", "image"],
8464
8615
  cost: {
8465
8616
  input: 10,
@@ -8529,7 +8680,7 @@ export const MODELS = {
8529
8680
  baseUrl: "https://opencode.ai/zen",
8530
8681
  compat: { "forceAdaptiveThinking": true },
8531
8682
  reasoning: true,
8532
- thinkingLevelMap: { "xhigh": "max" },
8683
+ thinkingLevelMap: { "max": "max" },
8533
8684
  input: ["text", "image"],
8534
8685
  cost: {
8535
8686
  input: 5,
@@ -8548,7 +8699,7 @@ export const MODELS = {
8548
8699
  baseUrl: "https://opencode.ai/zen",
8549
8700
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
8550
8701
  reasoning: true,
8551
- thinkingLevelMap: { "xhigh": "xhigh" },
8702
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
8552
8703
  input: ["text", "image"],
8553
8704
  cost: {
8554
8705
  input: 5,
@@ -8567,7 +8718,26 @@ export const MODELS = {
8567
8718
  baseUrl: "https://opencode.ai/zen",
8568
8719
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
8569
8720
  reasoning: true,
8570
- thinkingLevelMap: { "xhigh": "xhigh" },
8721
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
8722
+ input: ["text", "image"],
8723
+ cost: {
8724
+ input: 5,
8725
+ output: 25,
8726
+ cacheRead: 0.5,
8727
+ cacheWrite: 6.25,
8728
+ },
8729
+ contextWindow: 1000000,
8730
+ maxTokens: 128000,
8731
+ },
8732
+ "claude-opus-5": {
8733
+ id: "claude-opus-5",
8734
+ name: "Claude Opus 5",
8735
+ api: "anthropic-messages",
8736
+ provider: "opencode",
8737
+ baseUrl: "https://opencode.ai/zen",
8738
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
8739
+ reasoning: true,
8740
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
8571
8741
  input: ["text", "image"],
8572
8742
  cost: {
8573
8743
  input: 5,
@@ -8620,6 +8790,7 @@ export const MODELS = {
8620
8790
  baseUrl: "https://opencode.ai/zen",
8621
8791
  compat: { "forceAdaptiveThinking": true },
8622
8792
  reasoning: true,
8793
+ thinkingLevelMap: { "max": "max" },
8623
8794
  input: ["text", "image"],
8624
8795
  cost: {
8625
8796
  input: 3,
@@ -8636,7 +8807,9 @@ export const MODELS = {
8636
8807
  api: "anthropic-messages",
8637
8808
  provider: "opencode",
8638
8809
  baseUrl: "https://opencode.ai/zen",
8810
+ compat: { "forceAdaptiveThinking": true },
8639
8811
  reasoning: true,
8812
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
8640
8813
  input: ["text", "image"],
8641
8814
  cost: {
8642
8815
  input: 2,
@@ -8653,7 +8826,6 @@ export const MODELS = {
8653
8826
  api: "openai-completions",
8654
8827
  provider: "opencode",
8655
8828
  baseUrl: "https://opencode.ai/zen/v1",
8656
- compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
8657
8829
  reasoning: true,
8658
8830
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8659
8831
  input: ["text"],
@@ -8672,7 +8844,6 @@ export const MODELS = {
8672
8844
  api: "openai-completions",
8673
8845
  provider: "opencode",
8674
8846
  baseUrl: "https://opencode.ai/zen/v1",
8675
- compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
8676
8847
  reasoning: true,
8677
8848
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8678
8849
  input: ["text"],
@@ -8691,7 +8862,6 @@ export const MODELS = {
8691
8862
  api: "openai-completions",
8692
8863
  provider: "opencode",
8693
8864
  baseUrl: "https://opencode.ai/zen/v1",
8694
- compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
8695
8865
  reasoning: true,
8696
8866
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8697
8867
  input: ["text"],
@@ -8758,6 +8928,42 @@ export const MODELS = {
8758
8928
  contextWindow: 1048576,
8759
8929
  maxTokens: 65536,
8760
8930
  },
8931
+ "gemini-3.5-flash-lite": {
8932
+ id: "gemini-3.5-flash-lite",
8933
+ name: "Gemini 3.5 Flash Lite",
8934
+ api: "google-generative-ai",
8935
+ provider: "opencode",
8936
+ baseUrl: "https://opencode.ai/zen/v1",
8937
+ reasoning: true,
8938
+ thinkingLevelMap: { "off": null },
8939
+ input: ["text", "image"],
8940
+ cost: {
8941
+ input: 0.3,
8942
+ output: 2.5,
8943
+ cacheRead: 0.03,
8944
+ cacheWrite: 0,
8945
+ },
8946
+ contextWindow: 1048576,
8947
+ maxTokens: 65536,
8948
+ },
8949
+ "gemini-3.6-flash": {
8950
+ id: "gemini-3.6-flash",
8951
+ name: "Gemini 3.6 Flash",
8952
+ api: "google-generative-ai",
8953
+ provider: "opencode",
8954
+ baseUrl: "https://opencode.ai/zen/v1",
8955
+ reasoning: true,
8956
+ thinkingLevelMap: { "off": null },
8957
+ input: ["text", "image"],
8958
+ cost: {
8959
+ input: 1.5,
8960
+ output: 7.5,
8961
+ cacheRead: 0.15,
8962
+ cacheWrite: 0,
8963
+ },
8964
+ contextWindow: 1048576,
8965
+ maxTokens: 65536,
8966
+ },
8761
8967
  "glm-5": {
8762
8968
  id: "glm-5",
8763
8969
  name: "GLM-5",
@@ -8815,6 +9021,7 @@ export const MODELS = {
8815
9021
  api: "openai-responses",
8816
9022
  provider: "opencode",
8817
9023
  baseUrl: "https://opencode.ai/zen/v1",
9024
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8818
9025
  reasoning: true,
8819
9026
  thinkingLevelMap: { "off": null },
8820
9027
  input: ["text", "image"],
@@ -8833,6 +9040,7 @@ export const MODELS = {
8833
9040
  api: "openai-responses",
8834
9041
  provider: "opencode",
8835
9042
  baseUrl: "https://opencode.ai/zen/v1",
9043
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8836
9044
  reasoning: true,
8837
9045
  thinkingLevelMap: { "off": null },
8838
9046
  input: ["text", "image"],
@@ -8851,6 +9059,7 @@ export const MODELS = {
8851
9059
  api: "openai-responses",
8852
9060
  provider: "opencode",
8853
9061
  baseUrl: "https://opencode.ai/zen/v1",
9062
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8854
9063
  reasoning: true,
8855
9064
  thinkingLevelMap: { "off": null },
8856
9065
  input: ["text", "image"],
@@ -8869,6 +9078,7 @@ export const MODELS = {
8869
9078
  api: "openai-responses",
8870
9079
  provider: "opencode",
8871
9080
  baseUrl: "https://opencode.ai/zen/v1",
9081
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8872
9082
  reasoning: true,
8873
9083
  thinkingLevelMap: { "off": null },
8874
9084
  input: ["text", "image"],
@@ -8887,6 +9097,7 @@ export const MODELS = {
8887
9097
  api: "openai-responses",
8888
9098
  provider: "opencode",
8889
9099
  baseUrl: "https://opencode.ai/zen/v1",
9100
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8890
9101
  reasoning: true,
8891
9102
  thinkingLevelMap: { "off": null },
8892
9103
  input: ["text", "image"],
@@ -8905,6 +9116,7 @@ export const MODELS = {
8905
9116
  api: "openai-responses",
8906
9117
  provider: "opencode",
8907
9118
  baseUrl: "https://opencode.ai/zen/v1",
9119
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8908
9120
  reasoning: true,
8909
9121
  thinkingLevelMap: { "off": null },
8910
9122
  input: ["text", "image"],
@@ -8923,6 +9135,7 @@ export const MODELS = {
8923
9135
  api: "openai-responses",
8924
9136
  provider: "opencode",
8925
9137
  baseUrl: "https://opencode.ai/zen/v1",
9138
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8926
9139
  reasoning: true,
8927
9140
  thinkingLevelMap: { "off": null },
8928
9141
  input: ["text", "image"],
@@ -8941,6 +9154,7 @@ export const MODELS = {
8941
9154
  api: "openai-responses",
8942
9155
  provider: "opencode",
8943
9156
  baseUrl: "https://opencode.ai/zen/v1",
9157
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8944
9158
  reasoning: true,
8945
9159
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
8946
9160
  input: ["text", "image"],
@@ -8959,6 +9173,7 @@ export const MODELS = {
8959
9173
  api: "openai-responses",
8960
9174
  provider: "opencode",
8961
9175
  baseUrl: "https://opencode.ai/zen/v1",
9176
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8962
9177
  reasoning: true,
8963
9178
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
8964
9179
  input: ["text", "image"],
@@ -8977,6 +9192,7 @@ export const MODELS = {
8977
9192
  api: "openai-responses",
8978
9193
  provider: "opencode",
8979
9194
  baseUrl: "https://opencode.ai/zen/v1",
9195
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8980
9196
  reasoning: true,
8981
9197
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
8982
9198
  input: ["text", "image"],
@@ -8995,6 +9211,7 @@ export const MODELS = {
8995
9211
  api: "openai-responses",
8996
9212
  provider: "opencode",
8997
9213
  baseUrl: "https://opencode.ai/zen/v1",
9214
+ compat: { "sessionAffinityFormat": "openai-nosession" },
8998
9215
  reasoning: true,
8999
9216
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
9000
9217
  input: ["text", "image"],
@@ -9013,6 +9230,7 @@ export const MODELS = {
9013
9230
  api: "openai-responses",
9014
9231
  provider: "opencode",
9015
9232
  baseUrl: "https://opencode.ai/zen/v1",
9233
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9016
9234
  reasoning: true,
9017
9235
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
9018
9236
  input: ["text", "image"],
@@ -9031,6 +9249,7 @@ export const MODELS = {
9031
9249
  api: "openai-responses",
9032
9250
  provider: "opencode",
9033
9251
  baseUrl: "https://opencode.ai/zen/v1",
9252
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9034
9253
  reasoning: true,
9035
9254
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
9036
9255
  input: ["text", "image"],
@@ -9049,6 +9268,7 @@ export const MODELS = {
9049
9268
  api: "openai-responses",
9050
9269
  provider: "opencode",
9051
9270
  baseUrl: "https://opencode.ai/zen/v1",
9271
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9052
9272
  reasoning: true,
9053
9273
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
9054
9274
  input: ["text", "image"],
@@ -9067,6 +9287,7 @@ export const MODELS = {
9067
9287
  api: "openai-responses",
9068
9288
  provider: "opencode",
9069
9289
  baseUrl: "https://opencode.ai/zen/v1",
9290
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9070
9291
  reasoning: true,
9071
9292
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
9072
9293
  input: ["text", "image"],
@@ -9085,6 +9306,7 @@ export const MODELS = {
9085
9306
  api: "openai-responses",
9086
9307
  provider: "opencode",
9087
9308
  baseUrl: "https://opencode.ai/zen/v1",
9309
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9088
9310
  reasoning: true,
9089
9311
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "minimal": null, "low": null },
9090
9312
  input: ["text", "image"],
@@ -9103,6 +9325,7 @@ export const MODELS = {
9103
9325
  api: "openai-responses",
9104
9326
  provider: "opencode",
9105
9327
  baseUrl: "https://opencode.ai/zen/v1",
9328
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9106
9329
  reasoning: true,
9107
9330
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
9108
9331
  input: ["text", "image"],
@@ -9121,6 +9344,7 @@ export const MODELS = {
9121
9344
  api: "openai-responses",
9122
9345
  provider: "opencode",
9123
9346
  baseUrl: "https://opencode.ai/zen/v1",
9347
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9124
9348
  reasoning: true,
9125
9349
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
9126
9350
  input: ["text", "image"],
@@ -9139,6 +9363,7 @@ export const MODELS = {
9139
9363
  api: "openai-responses",
9140
9364
  provider: "opencode",
9141
9365
  baseUrl: "https://opencode.ai/zen/v1",
9366
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9142
9367
  reasoning: true,
9143
9368
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
9144
9369
  input: ["text", "image"],
@@ -9157,6 +9382,7 @@ export const MODELS = {
9157
9382
  api: "openai-responses",
9158
9383
  provider: "opencode",
9159
9384
  baseUrl: "https://opencode.ai/zen/v1",
9385
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9160
9386
  reasoning: true,
9161
9387
  input: ["text", "image"],
9162
9388
  cost: {
@@ -9187,23 +9413,6 @@ export const MODELS = {
9187
9413
  contextWindow: 256000,
9188
9414
  maxTokens: 256000,
9189
9415
  },
9190
- "hy3-free": {
9191
- id: "hy3-free",
9192
- name: "Hy3 Free",
9193
- api: "openai-completions",
9194
- provider: "opencode",
9195
- baseUrl: "https://opencode.ai/zen/v1",
9196
- reasoning: true,
9197
- input: ["text"],
9198
- cost: {
9199
- input: 0,
9200
- output: 0,
9201
- cacheRead: 0,
9202
- cacheWrite: 0,
9203
- },
9204
- contextWindow: 190000,
9205
- maxTokens: 64000,
9206
- },
9207
9416
  "kimi-k2.5": {
9208
9417
  id: "kimi-k2.5",
9209
9418
  name: "Kimi K2.5",
@@ -9256,6 +9465,40 @@ export const MODELS = {
9256
9465
  contextWindow: 262144,
9257
9466
  maxTokens: 262144,
9258
9467
  },
9468
+ "laguna-s-2.1-free": {
9469
+ id: "laguna-s-2.1-free",
9470
+ name: "Laguna S 2.1 Free",
9471
+ api: "openai-completions",
9472
+ provider: "opencode",
9473
+ baseUrl: "https://opencode.ai/zen/v1",
9474
+ reasoning: true,
9475
+ input: ["text"],
9476
+ cost: {
9477
+ input: 0,
9478
+ output: 0,
9479
+ cacheRead: 0,
9480
+ cacheWrite: 0,
9481
+ },
9482
+ contextWindow: 256000,
9483
+ maxTokens: 32000,
9484
+ },
9485
+ "ling-3.0-flash-free": {
9486
+ id: "ling-3.0-flash-free",
9487
+ name: "Ling-3.0-flash Free",
9488
+ api: "openai-completions",
9489
+ provider: "opencode",
9490
+ baseUrl: "https://opencode.ai/zen/v1",
9491
+ reasoning: true,
9492
+ input: ["text"],
9493
+ cost: {
9494
+ input: 0,
9495
+ output: 0,
9496
+ cacheRead: 0,
9497
+ cacheWrite: 0,
9498
+ },
9499
+ contextWindow: 262144,
9500
+ maxTokens: 32768,
9501
+ },
9259
9502
  "mimo-v2.5-free": {
9260
9503
  id: "mimo-v2.5-free",
9261
9504
  name: "MiMo V2.5 Free",
@@ -9400,7 +9643,6 @@ export const MODELS = {
9400
9643
  api: "openai-completions",
9401
9644
  provider: "opencode-go",
9402
9645
  baseUrl: "https://opencode.ai/zen/go/v1",
9403
- compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
9404
9646
  reasoning: true,
9405
9647
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9406
9648
  input: ["text"],
@@ -9419,7 +9661,6 @@ export const MODELS = {
9419
9661
  api: "openai-completions",
9420
9662
  provider: "opencode-go",
9421
9663
  baseUrl: "https://opencode.ai/zen/go/v1",
9422
- compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
9423
9664
  reasoning: true,
9424
9665
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9425
9666
  input: ["text"],
@@ -9472,6 +9713,7 @@ export const MODELS = {
9472
9713
  api: "openai-responses",
9473
9714
  provider: "opencode-go",
9474
9715
  baseUrl: "https://opencode.ai/zen/go/v1",
9716
+ compat: { "sessionAffinityFormat": "openai-nosession" },
9475
9717
  reasoning: true,
9476
9718
  input: ["text", "image"],
9477
9719
  cost: {
@@ -9483,6 +9725,23 @@ export const MODELS = {
9483
9725
  contextWindow: 500000,
9484
9726
  maxTokens: 500000,
9485
9727
  },
9728
+ "hy3": {
9729
+ id: "hy3",
9730
+ name: "Hy3",
9731
+ api: "openai-completions",
9732
+ provider: "opencode-go",
9733
+ baseUrl: "https://opencode.ai/zen/go/v1",
9734
+ reasoning: true,
9735
+ input: ["text"],
9736
+ cost: {
9737
+ input: 0.14,
9738
+ output: 0.58,
9739
+ cacheRead: 0.035,
9740
+ cacheWrite: 0,
9741
+ },
9742
+ contextWindow: 256000,
9743
+ maxTokens: 64000,
9744
+ },
9486
9745
  "kimi-k2.6": {
9487
9746
  id: "kimi-k2.6",
9488
9747
  name: "Kimi K2.6",
@@ -9817,6 +10076,7 @@ export const MODELS = {
9817
10076
  api: "openai-completions",
9818
10077
  provider: "openrouter",
9819
10078
  baseUrl: "https://openrouter.ai/api/v1",
10079
+ compat: { "cacheControlFormat": "anthropic" },
9820
10080
  reasoning: false,
9821
10081
  input: ["text", "image"],
9822
10082
  cost: {
@@ -9834,7 +10094,9 @@ export const MODELS = {
9834
10094
  api: "openai-completions",
9835
10095
  provider: "openrouter",
9836
10096
  baseUrl: "https://openrouter.ai/api/v1",
10097
+ compat: { "cacheControlFormat": "anthropic" },
9837
10098
  reasoning: true,
10099
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
9838
10100
  input: ["text", "image"],
9839
10101
  cost: {
9840
10102
  input: 10,
@@ -9851,6 +10113,7 @@ export const MODELS = {
9851
10113
  api: "openai-completions",
9852
10114
  provider: "openrouter",
9853
10115
  baseUrl: "https://openrouter.ai/api/v1",
10116
+ compat: { "cacheControlFormat": "anthropic" },
9854
10117
  reasoning: true,
9855
10118
  input: ["text", "image"],
9856
10119
  cost: {
@@ -9868,6 +10131,7 @@ export const MODELS = {
9868
10131
  api: "openai-completions",
9869
10132
  provider: "openrouter",
9870
10133
  baseUrl: "https://openrouter.ai/api/v1",
10134
+ compat: { "cacheControlFormat": "anthropic" },
9871
10135
  reasoning: true,
9872
10136
  input: ["text", "image"],
9873
10137
  cost: {
@@ -9885,6 +10149,7 @@ export const MODELS = {
9885
10149
  api: "openai-completions",
9886
10150
  provider: "openrouter",
9887
10151
  baseUrl: "https://openrouter.ai/api/v1",
10152
+ compat: { "cacheControlFormat": "anthropic" },
9888
10153
  reasoning: true,
9889
10154
  input: ["text", "image"],
9890
10155
  cost: {
@@ -9902,6 +10167,7 @@ export const MODELS = {
9902
10167
  api: "openai-completions",
9903
10168
  provider: "openrouter",
9904
10169
  baseUrl: "https://openrouter.ai/api/v1",
10170
+ compat: { "cacheControlFormat": "anthropic" },
9905
10171
  reasoning: true,
9906
10172
  input: ["text", "image"],
9907
10173
  cost: {
@@ -9919,8 +10185,9 @@ export const MODELS = {
9919
10185
  api: "openai-completions",
9920
10186
  provider: "openrouter",
9921
10187
  baseUrl: "https://openrouter.ai/api/v1",
10188
+ compat: { "cacheControlFormat": "anthropic" },
9922
10189
  reasoning: true,
9923
- thinkingLevelMap: { "xhigh": "max" },
10190
+ thinkingLevelMap: { "max": "max" },
9924
10191
  input: ["text", "image"],
9925
10192
  cost: {
9926
10193
  input: 5,
@@ -9937,8 +10204,9 @@ export const MODELS = {
9937
10204
  api: "openai-completions",
9938
10205
  provider: "openrouter",
9939
10206
  baseUrl: "https://openrouter.ai/api/v1",
10207
+ compat: { "cacheControlFormat": "anthropic" },
9940
10208
  reasoning: true,
9941
- thinkingLevelMap: { "xhigh": "xhigh" },
10209
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
9942
10210
  input: ["text", "image"],
9943
10211
  cost: {
9944
10212
  input: 5,
@@ -9955,8 +10223,9 @@ export const MODELS = {
9955
10223
  api: "openai-completions",
9956
10224
  provider: "openrouter",
9957
10225
  baseUrl: "https://openrouter.ai/api/v1",
10226
+ compat: { "cacheControlFormat": "anthropic" },
9958
10227
  reasoning: true,
9959
- thinkingLevelMap: { "xhigh": "xhigh" },
10228
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
9960
10229
  input: ["text", "image"],
9961
10230
  cost: {
9962
10231
  input: 30,
@@ -9973,8 +10242,9 @@ export const MODELS = {
9973
10242
  api: "openai-completions",
9974
10243
  provider: "openrouter",
9975
10244
  baseUrl: "https://openrouter.ai/api/v1",
10245
+ compat: { "cacheControlFormat": "anthropic" },
9976
10246
  reasoning: true,
9977
- thinkingLevelMap: { "xhigh": "xhigh" },
10247
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
9978
10248
  input: ["text", "image"],
9979
10249
  cost: {
9980
10250
  input: 5,
@@ -9991,8 +10261,47 @@ export const MODELS = {
9991
10261
  api: "openai-completions",
9992
10262
  provider: "openrouter",
9993
10263
  baseUrl: "https://openrouter.ai/api/v1",
10264
+ compat: { "cacheControlFormat": "anthropic" },
9994
10265
  reasoning: true,
9995
- thinkingLevelMap: { "xhigh": "xhigh" },
10266
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
10267
+ input: ["text", "image"],
10268
+ cost: {
10269
+ input: 10,
10270
+ output: 50,
10271
+ cacheRead: 1,
10272
+ cacheWrite: 12.5,
10273
+ },
10274
+ contextWindow: 1000000,
10275
+ maxTokens: 128000,
10276
+ },
10277
+ "anthropic/claude-opus-5": {
10278
+ id: "anthropic/claude-opus-5",
10279
+ name: "Claude Opus 5",
10280
+ api: "openai-completions",
10281
+ provider: "openrouter",
10282
+ baseUrl: "https://openrouter.ai/api/v1",
10283
+ compat: { "cacheControlFormat": "anthropic" },
10284
+ reasoning: true,
10285
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
10286
+ input: ["text", "image"],
10287
+ cost: {
10288
+ input: 5,
10289
+ output: 25,
10290
+ cacheRead: 0.5,
10291
+ cacheWrite: 6.25,
10292
+ },
10293
+ contextWindow: 1000000,
10294
+ maxTokens: 128000,
10295
+ },
10296
+ "anthropic/claude-opus-5-fast": {
10297
+ id: "anthropic/claude-opus-5-fast",
10298
+ name: "Claude Opus 5 (Fast)",
10299
+ api: "openai-completions",
10300
+ provider: "openrouter",
10301
+ baseUrl: "https://openrouter.ai/api/v1",
10302
+ compat: { "cacheControlFormat": "anthropic" },
10303
+ reasoning: true,
10304
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
9996
10305
  input: ["text", "image"],
9997
10306
  cost: {
9998
10307
  input: 10,
@@ -10009,6 +10318,7 @@ export const MODELS = {
10009
10318
  api: "openai-completions",
10010
10319
  provider: "openrouter",
10011
10320
  baseUrl: "https://openrouter.ai/api/v1",
10321
+ compat: { "cacheControlFormat": "anthropic" },
10012
10322
  reasoning: true,
10013
10323
  input: ["text", "image"],
10014
10324
  cost: {
@@ -10026,6 +10336,7 @@ export const MODELS = {
10026
10336
  api: "openai-completions",
10027
10337
  provider: "openrouter",
10028
10338
  baseUrl: "https://openrouter.ai/api/v1",
10339
+ compat: { "cacheControlFormat": "anthropic" },
10029
10340
  reasoning: true,
10030
10341
  input: ["text", "image"],
10031
10342
  cost: {
@@ -10043,7 +10354,9 @@ export const MODELS = {
10043
10354
  api: "openai-completions",
10044
10355
  provider: "openrouter",
10045
10356
  baseUrl: "https://openrouter.ai/api/v1",
10357
+ compat: { "cacheControlFormat": "anthropic" },
10046
10358
  reasoning: true,
10359
+ thinkingLevelMap: { "max": "max" },
10047
10360
  input: ["text", "image"],
10048
10361
  cost: {
10049
10362
  input: 3,
@@ -10060,7 +10373,9 @@ export const MODELS = {
10060
10373
  api: "openai-completions",
10061
10374
  provider: "openrouter",
10062
10375
  baseUrl: "https://openrouter.ai/api/v1",
10376
+ compat: { "cacheControlFormat": "anthropic" },
10063
10377
  reasoning: true,
10378
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
10064
10379
  input: ["text", "image"],
10065
10380
  cost: {
10066
10381
  input: 2,
@@ -10080,13 +10395,13 @@ export const MODELS = {
10080
10395
  reasoning: true,
10081
10396
  input: ["text"],
10082
10397
  cost: {
10083
- input: 0.25,
10084
- output: 0.7999999999999999,
10398
+ input: 0.22,
10399
+ output: 0.85,
10085
10400
  cacheRead: 0.06,
10086
10401
  cacheWrite: 0,
10087
10402
  },
10088
10403
  contextWindow: 262144,
10089
- maxTokens: 80000,
10404
+ maxTokens: 262144,
10090
10405
  },
10091
10406
  "arcee-ai/virtuoso-large": {
10092
10407
  id: "arcee-ai/virtuoso-large",
@@ -10255,7 +10570,7 @@ export const MODELS = {
10255
10570
  cacheRead: 0,
10256
10571
  cacheWrite: 0,
10257
10572
  },
10258
- contextWindow: 131072,
10573
+ contextWindow: 163840,
10259
10574
  maxTokens: 16000,
10260
10575
  },
10261
10576
  "deepseek/deepseek-chat-v3-0324": {
@@ -10340,7 +10655,7 @@ export const MODELS = {
10340
10655
  cacheRead: 0.135,
10341
10656
  cacheWrite: 0,
10342
10657
  },
10343
- contextWindow: 131072,
10658
+ contextWindow: 163840,
10344
10659
  maxTokens: 32768,
10345
10660
  },
10346
10661
  "deepseek/deepseek-v3.2": {
@@ -10388,13 +10703,13 @@ export const MODELS = {
10388
10703
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
10389
10704
  input: ["text"],
10390
10705
  cost: {
10391
- input: 0.098,
10392
- output: 0.196,
10393
- cacheRead: 0.0196,
10706
+ input: 0.14,
10707
+ output: 0.28,
10708
+ cacheRead: 0.028,
10394
10709
  cacheWrite: 0,
10395
10710
  },
10396
10711
  contextWindow: 1048576,
10397
- maxTokens: 4096,
10712
+ maxTokens: 393216,
10398
10713
  },
10399
10714
  "deepseek/deepseek-v4-pro": {
10400
10715
  id: "deepseek/deepseek-v4-pro",
@@ -10531,7 +10846,7 @@ export const MODELS = {
10531
10846
  cacheRead: 0.19999999999999998,
10532
10847
  cacheWrite: 0.375,
10533
10848
  },
10534
- contextWindow: 65536,
10849
+ contextWindow: 131072,
10535
10850
  maxTokens: 32768,
10536
10851
  },
10537
10852
  "google/gemini-3.1-flash-lite": {
@@ -10599,7 +10914,7 @@ export const MODELS = {
10599
10914
  cacheRead: 0.19999999999999998,
10600
10915
  cacheWrite: 0.375,
10601
10916
  },
10602
- contextWindow: 1048756,
10917
+ contextWindow: 1048576,
10603
10918
  maxTokens: 65536,
10604
10919
  },
10605
10920
  "google/gemini-3.5-flash": {
@@ -10619,6 +10934,40 @@ export const MODELS = {
10619
10934
  contextWindow: 1048576,
10620
10935
  maxTokens: 65536,
10621
10936
  },
10937
+ "google/gemini-3.5-flash-lite": {
10938
+ id: "google/gemini-3.5-flash-lite",
10939
+ name: "Google: Gemini 3.5 Flash Lite",
10940
+ api: "openai-completions",
10941
+ provider: "openrouter",
10942
+ baseUrl: "https://openrouter.ai/api/v1",
10943
+ reasoning: true,
10944
+ input: ["text", "image"],
10945
+ cost: {
10946
+ input: 0.3,
10947
+ output: 2.5,
10948
+ cacheRead: 0.03,
10949
+ cacheWrite: 0.08333333333333334,
10950
+ },
10951
+ contextWindow: 1048576,
10952
+ maxTokens: 65536,
10953
+ },
10954
+ "google/gemini-3.6-flash": {
10955
+ id: "google/gemini-3.6-flash",
10956
+ name: "Google: Gemini 3.6 Flash",
10957
+ api: "openai-completions",
10958
+ provider: "openrouter",
10959
+ baseUrl: "https://openrouter.ai/api/v1",
10960
+ reasoning: true,
10961
+ input: ["text", "image"],
10962
+ cost: {
10963
+ input: 1.5,
10964
+ output: 7.5,
10965
+ cacheRead: 0.15,
10966
+ cacheWrite: 0.08333333333333334,
10967
+ },
10968
+ contextWindow: 1048576,
10969
+ maxTokens: 65536,
10970
+ },
10622
10971
  "google/gemma-3-12b-it": {
10623
10972
  id: "google/gemma-3-12b-it",
10624
10973
  name: "Google: Gemma 3 12B",
@@ -10645,13 +10994,13 @@ export const MODELS = {
10645
10994
  reasoning: false,
10646
10995
  input: ["text", "image"],
10647
10996
  cost: {
10648
- input: 0.09999999999999999,
10649
- output: 0.3,
10650
- cacheRead: 0,
10997
+ input: 0.08,
10998
+ output: 0.44999999999999996,
10999
+ cacheRead: 0.04,
10651
11000
  cacheWrite: 0,
10652
11001
  },
10653
- contextWindow: 131072,
10654
- maxTokens: 4096,
11002
+ contextWindow: 262144,
11003
+ maxTokens: 131072,
10655
11004
  },
10656
11005
  "google/gemma-4-26b-a4b-it": {
10657
11006
  id: "google/gemma-4-26b-a4b-it",
@@ -10662,13 +11011,13 @@ export const MODELS = {
10662
11011
  reasoning: true,
10663
11012
  input: ["text", "image"],
10664
11013
  cost: {
10665
- input: 0.07,
10666
- output: 0.33999999999999997,
10667
- cacheRead: 0,
11014
+ input: 0.12,
11015
+ output: 0.35,
11016
+ cacheRead: 0.049999999999999996,
10668
11017
  cacheWrite: 0,
10669
11018
  },
10670
11019
  contextWindow: 262144,
10671
- maxTokens: 16384,
11020
+ maxTokens: 262144,
10672
11021
  },
10673
11022
  "google/gemma-4-26b-a4b-it:free": {
10674
11023
  id: "google/gemma-4-26b-a4b-it:free",
@@ -10696,13 +11045,13 @@ export const MODELS = {
10696
11045
  reasoning: true,
10697
11046
  input: ["text", "image"],
10698
11047
  cost: {
10699
- input: 0.12,
10700
- output: 0.37,
11048
+ input: 0.14,
11049
+ output: 0.39999999999999997,
10701
11050
  cacheRead: 0,
10702
11051
  cacheWrite: 0,
10703
11052
  },
10704
11053
  contextWindow: 262144,
10705
- maxTokens: 16384,
11054
+ maxTokens: 262144,
10706
11055
  },
10707
11056
  "google/gemma-4-31b-it:free": {
10708
11057
  id: "google/gemma-4-31b-it:free",
@@ -10790,6 +11139,23 @@ export const MODELS = {
10790
11139
  contextWindow: 262144,
10791
11140
  maxTokens: 32768,
10792
11141
  },
11142
+ "inclusionai/ling-3.0-flash:free": {
11143
+ id: "inclusionai/ling-3.0-flash:free",
11144
+ name: "Ling-3.0-flash (free)",
11145
+ api: "openai-completions",
11146
+ provider: "openrouter",
11147
+ baseUrl: "https://openrouter.ai/api/v1",
11148
+ reasoning: true,
11149
+ input: ["text"],
11150
+ cost: {
11151
+ input: 0,
11152
+ output: 0,
11153
+ cacheRead: 0,
11154
+ cacheWrite: 0,
11155
+ },
11156
+ contextWindow: 262144,
11157
+ maxTokens: 32768,
11158
+ },
10793
11159
  "inclusionai/ring-2.6-1t": {
10794
11160
  id: "inclusionai/ring-2.6-1t",
10795
11161
  name: "inclusionAI: Ring-2.6-1T",
@@ -10838,7 +11204,7 @@ export const MODELS = {
10838
11204
  cacheRead: 0.06,
10839
11205
  cacheWrite: 0,
10840
11206
  },
10841
- contextWindow: 256000,
11207
+ contextWindow: 262144,
10842
11208
  maxTokens: 80000,
10843
11209
  },
10844
11210
  "kwaipilot/kat-coder-pro-v2.5": {
@@ -10957,7 +11323,7 @@ export const MODELS = {
10957
11323
  cacheRead: 0,
10958
11324
  cacheWrite: 0,
10959
11325
  },
10960
- contextWindow: 10000000,
11326
+ contextWindow: 1310720,
10961
11327
  maxTokens: 16384,
10962
11328
  },
10963
11329
  "meta/muse-spark-1.1": {
@@ -11331,7 +11697,7 @@ export const MODELS = {
11331
11697
  cacheRead: 0.01,
11332
11698
  cacheWrite: 0,
11333
11699
  },
11334
- contextWindow: 131072,
11700
+ contextWindow: 256000,
11335
11701
  maxTokens: 4096,
11336
11702
  },
11337
11703
  "mistralai/mixtral-8x22b-instruct": {
@@ -11446,9 +11812,9 @@ export const MODELS = {
11446
11812
  reasoning: true,
11447
11813
  input: ["text", "image"],
11448
11814
  cost: {
11449
- input: 0.684,
11450
- output: 3.42,
11451
- cacheRead: 0.144,
11815
+ input: 0.646,
11816
+ output: 2.7199999999999998,
11817
+ cacheRead: 0.1088,
11452
11818
  cacheWrite: 0,
11453
11819
  },
11454
11820
  contextWindow: 262144,
@@ -11463,9 +11829,9 @@ export const MODELS = {
11463
11829
  reasoning: true,
11464
11830
  input: ["text", "image"],
11465
11831
  cost: {
11466
- input: 0.82,
11467
- output: 3.75,
11468
- cacheRead: 0.16,
11832
+ input: 0.73,
11833
+ output: 3.5,
11834
+ cacheRead: 0.15,
11469
11835
  cacheWrite: 0,
11470
11836
  },
11471
11837
  contextWindow: 262144,
@@ -11604,7 +11970,7 @@ export const MODELS = {
11604
11970
  cacheRead: 0,
11605
11971
  cacheWrite: 0,
11606
11972
  },
11607
- contextWindow: 1000000,
11973
+ contextWindow: 262144,
11608
11974
  maxTokens: 262144,
11609
11975
  },
11610
11976
  "nvidia/nemotron-3-ultra-550b-a55b": {
@@ -11621,7 +11987,7 @@ export const MODELS = {
11621
11987
  cacheRead: 0.19999999999999998,
11622
11988
  cacheWrite: 0,
11623
11989
  },
11624
- contextWindow: 1000000,
11990
+ contextWindow: 512288,
11625
11991
  maxTokens: 4096,
11626
11992
  },
11627
11993
  "nvidia/nemotron-3-ultra-550b-a55b:free": {
@@ -12060,7 +12426,7 @@ export const MODELS = {
12060
12426
  cost: {
12061
12427
  input: 1.25,
12062
12428
  output: 10,
12063
- cacheRead: 0.125,
12429
+ cacheRead: 0.13,
12064
12430
  cacheWrite: 0,
12065
12431
  },
12066
12432
  contextWindow: 400000,
@@ -12094,11 +12460,11 @@ export const MODELS = {
12094
12460
  cost: {
12095
12461
  input: 0.25,
12096
12462
  output: 2,
12097
- cacheRead: 0.024999999999999998,
12463
+ cacheRead: 0.03,
12098
12464
  cacheWrite: 0,
12099
12465
  },
12100
12466
  contextWindow: 400000,
12101
- maxTokens: 100000,
12467
+ maxTokens: 128000,
12102
12468
  },
12103
12469
  "openai/gpt-5.2": {
12104
12470
  id: "openai/gpt-5.2",
@@ -12502,8 +12868,8 @@ export const MODELS = {
12502
12868
  input: ["text"],
12503
12869
  cost: {
12504
12870
  input: 0.03,
12505
- output: 0.13,
12506
- cacheRead: 0.03,
12871
+ output: 0.14,
12872
+ cacheRead: 0,
12507
12873
  cacheWrite: 0,
12508
12874
  },
12509
12875
  contextWindow: 131072,
@@ -12798,6 +13164,40 @@ export const MODELS = {
12798
13164
  contextWindow: 262144,
12799
13165
  maxTokens: 32768,
12800
13166
  },
13167
+ "poolside/laguna-s-2.1": {
13168
+ id: "poolside/laguna-s-2.1",
13169
+ name: "Poolside: Laguna S 2.1",
13170
+ api: "openai-completions",
13171
+ provider: "openrouter",
13172
+ baseUrl: "https://openrouter.ai/api/v1",
13173
+ reasoning: true,
13174
+ input: ["text"],
13175
+ cost: {
13176
+ input: 0.09999999999999999,
13177
+ output: 0.19999999999999998,
13178
+ cacheRead: 0.01,
13179
+ cacheWrite: 0,
13180
+ },
13181
+ contextWindow: 1048576,
13182
+ maxTokens: 131072,
13183
+ },
13184
+ "poolside/laguna-s-2.1:free": {
13185
+ id: "poolside/laguna-s-2.1:free",
13186
+ name: "Poolside: Laguna S 2.1 (free)",
13187
+ api: "openai-completions",
13188
+ provider: "openrouter",
13189
+ baseUrl: "https://openrouter.ai/api/v1",
13190
+ reasoning: true,
13191
+ input: ["text"],
13192
+ cost: {
13193
+ input: 0,
13194
+ output: 0,
13195
+ cacheRead: 0,
13196
+ cacheWrite: 0,
13197
+ },
13198
+ contextWindow: 262144,
13199
+ maxTokens: 32768,
13200
+ },
12801
13201
  "poolside/laguna-xs-2.1": {
12802
13202
  id: "poolside/laguna-xs-2.1",
12803
13203
  name: "Poolside: Laguna XS 2.1",
@@ -12846,7 +13246,7 @@ export const MODELS = {
12846
13246
  cacheRead: 0,
12847
13247
  cacheWrite: 0,
12848
13248
  },
12849
- contextWindow: 131072,
13249
+ contextWindow: 32768,
12850
13250
  maxTokens: 16384,
12851
13251
  },
12852
13252
  "qwen/qwen-2.5-7b-instruct": {
@@ -12863,7 +13263,7 @@ export const MODELS = {
12863
13263
  cacheRead: 0,
12864
13264
  cacheWrite: 0,
12865
13265
  },
12866
- contextWindow: 131072,
13266
+ contextWindow: 32768,
12867
13267
  maxTokens: 32768,
12868
13268
  },
12869
13269
  "qwen/qwen-plus": {
@@ -12926,13 +13326,13 @@ export const MODELS = {
12926
13326
  reasoning: true,
12927
13327
  input: ["text"],
12928
13328
  cost: {
12929
- input: 0.12,
12930
- output: 0.24,
13329
+ input: 0.22749999999999998,
13330
+ output: 0.9099999999999999,
12931
13331
  cacheRead: 0,
12932
13332
  cacheWrite: 0,
12933
13333
  },
12934
- contextWindow: 131702,
12935
- maxTokens: 16384,
13334
+ contextWindow: 131072,
13335
+ maxTokens: 8192,
12936
13336
  },
12937
13337
  "qwen/qwen3-235b-a22b": {
12938
13338
  id: "qwen/qwen3-235b-a22b",
@@ -12994,13 +13394,13 @@ export const MODELS = {
12994
13394
  reasoning: true,
12995
13395
  input: ["text"],
12996
13396
  cost: {
12997
- input: 0.13,
12998
- output: 0.52,
13397
+ input: 0.12,
13398
+ output: 0.5,
12999
13399
  cacheRead: 0,
13000
13400
  cacheWrite: 0,
13001
13401
  },
13002
13402
  contextWindow: 131072,
13003
- maxTokens: 8192,
13403
+ maxTokens: 16384,
13004
13404
  },
13005
13405
  "qwen/qwen3-30b-a3b-instruct-2507": {
13006
13406
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -13011,13 +13411,13 @@ export const MODELS = {
13011
13411
  reasoning: false,
13012
13412
  input: ["text"],
13013
13413
  cost: {
13014
- input: 0.09999999999999999,
13015
- output: 0.3,
13414
+ input: 0.04815,
13415
+ output: 0.19305,
13016
13416
  cacheRead: 0,
13017
13417
  cacheWrite: 0,
13018
13418
  },
13019
13419
  contextWindow: 262144,
13020
- maxTokens: 4096,
13420
+ maxTokens: 32000,
13021
13421
  },
13022
13422
  "qwen/qwen3-30b-a3b-thinking-2507": {
13023
13423
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -13033,7 +13433,7 @@ export const MODELS = {
13033
13433
  cacheRead: 0,
13034
13434
  cacheWrite: 0,
13035
13435
  },
13036
- contextWindow: 131072,
13436
+ contextWindow: 81920,
13037
13437
  maxTokens: 32768,
13038
13438
  },
13039
13439
  "qwen/qwen3-32b": {
@@ -13084,7 +13484,7 @@ export const MODELS = {
13084
13484
  cacheRead: 0.09999999999999999,
13085
13485
  cacheWrite: 0,
13086
13486
  },
13087
- contextWindow: 1048576,
13487
+ contextWindow: 262144,
13088
13488
  maxTokens: 65536,
13089
13489
  },
13090
13490
  "qwen/qwen3-coder-30b-a3b-instruct": {
@@ -13101,7 +13501,7 @@ export const MODELS = {
13101
13501
  cacheRead: 0,
13102
13502
  cacheWrite: 0,
13103
13503
  },
13104
- contextWindow: 160000,
13504
+ contextWindow: 262144,
13105
13505
  maxTokens: 32768,
13106
13506
  },
13107
13507
  "qwen/qwen3-coder-flash": {
@@ -13237,7 +13637,7 @@ export const MODELS = {
13237
13637
  cacheRead: 0.09999999999999999,
13238
13638
  cacheWrite: 0,
13239
13639
  },
13240
- contextWindow: 131072,
13640
+ contextWindow: 262144,
13241
13641
  maxTokens: 32768,
13242
13642
  },
13243
13643
  "qwen/qwen3-vl-235b-a22b-thinking": {
@@ -13266,13 +13666,13 @@ export const MODELS = {
13266
13666
  reasoning: false,
13267
13667
  input: ["text", "image"],
13268
13668
  cost: {
13269
- input: 0.13,
13270
- output: 0.52,
13669
+ input: 0.15,
13670
+ output: 0.6,
13271
13671
  cacheRead: 0,
13272
13672
  cacheWrite: 0,
13273
13673
  },
13274
13674
  contextWindow: 262144,
13275
- maxTokens: 32768,
13675
+ maxTokens: 16384,
13276
13676
  },
13277
13677
  "qwen/qwen3-vl-30b-a3b-thinking": {
13278
13678
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -13288,7 +13688,7 @@ export const MODELS = {
13288
13688
  cacheRead: 0,
13289
13689
  cacheWrite: 0,
13290
13690
  },
13291
- contextWindow: 131072,
13691
+ contextWindow: 262144,
13292
13692
  maxTokens: 32768,
13293
13693
  },
13294
13694
  "qwen/qwen3-vl-32b-instruct": {
@@ -13305,7 +13705,7 @@ export const MODELS = {
13305
13705
  cacheRead: 0,
13306
13706
  cacheWrite: 0,
13307
13707
  },
13308
- contextWindow: 262144,
13708
+ contextWindow: 131072,
13309
13709
  maxTokens: 32768,
13310
13710
  },
13311
13711
  "qwen/qwen3-vl-8b-instruct": {
@@ -13322,7 +13722,7 @@ export const MODELS = {
13322
13722
  cacheRead: 0,
13323
13723
  cacheWrite: 0,
13324
13724
  },
13325
- contextWindow: 256000,
13725
+ contextWindow: 262144,
13326
13726
  maxTokens: 32768,
13327
13727
  },
13328
13728
  "qwen/qwen3-vl-8b-thinking": {
@@ -13339,7 +13739,7 @@ export const MODELS = {
13339
13739
  cacheRead: 0,
13340
13740
  cacheWrite: 0,
13341
13741
  },
13342
- contextWindow: 256000,
13742
+ contextWindow: 131072,
13343
13743
  maxTokens: 32768,
13344
13744
  },
13345
13745
  "qwen/qwen3.5-122b-a10b": {
@@ -13368,13 +13768,13 @@ export const MODELS = {
13368
13768
  reasoning: true,
13369
13769
  input: ["text", "image"],
13370
13770
  cost: {
13371
- input: 0.26,
13372
- output: 2.6,
13771
+ input: 0.195,
13772
+ output: 1.56,
13373
13773
  cacheRead: 0,
13374
13774
  cacheWrite: 0,
13375
13775
  },
13376
13776
  contextWindow: 262144,
13377
- maxTokens: 81920,
13777
+ maxTokens: 65536,
13378
13778
  },
13379
13779
  "qwen/qwen3.5-35b-a3b": {
13380
13780
  id: "qwen/qwen3.5-35b-a3b",
@@ -13487,9 +13887,9 @@ export const MODELS = {
13487
13887
  reasoning: true,
13488
13888
  input: ["text", "image"],
13489
13889
  cost: {
13490
- input: 0.44999999999999996,
13491
- output: 2.7,
13492
- cacheRead: 0,
13890
+ input: 0.3,
13891
+ output: 2,
13892
+ cacheRead: 0.15,
13493
13893
  cacheWrite: 0,
13494
13894
  },
13495
13895
  contextWindow: 262144,
@@ -13696,7 +14096,7 @@ export const MODELS = {
13696
14096
  cacheRead: 0.04,
13697
14097
  cacheWrite: 0,
13698
14098
  },
13699
- contextWindow: 256000,
14099
+ contextWindow: 262144,
13700
14100
  maxTokens: 256000,
13701
14101
  },
13702
14102
  "tencent/hy3": {
@@ -13708,13 +14108,13 @@ export const MODELS = {
13708
14108
  reasoning: true,
13709
14109
  input: ["text"],
13710
14110
  cost: {
13711
- input: 0.19999999999999998,
13712
- output: 0.7999999999999999,
13713
- cacheRead: 0.049999999999999996,
14111
+ input: 0.13199999999999998,
14112
+ output: 0.5279999999999999,
14113
+ cacheRead: 0.032999999999999995,
13714
14114
  cacheWrite: 0,
13715
14115
  },
13716
14116
  contextWindow: 262144,
13717
- maxTokens: 131072,
14117
+ maxTokens: 128000,
13718
14118
  },
13719
14119
  "tencent/hy3-preview": {
13720
14120
  id: "tencent/hy3-preview",
@@ -13733,23 +14133,6 @@ export const MODELS = {
13733
14133
  contextWindow: 262144,
13734
14134
  maxTokens: 4096,
13735
14135
  },
13736
- "tencent/hy3:free": {
13737
- id: "tencent/hy3:free",
13738
- name: "Tencent: Hy3 (free)",
13739
- api: "openai-completions",
13740
- provider: "openrouter",
13741
- baseUrl: "https://openrouter.ai/api/v1",
13742
- reasoning: true,
13743
- input: ["text"],
13744
- cost: {
13745
- input: 0,
13746
- output: 0,
13747
- cacheRead: 0,
13748
- cacheWrite: 0,
13749
- },
13750
- contextWindow: 262144,
13751
- maxTokens: 262144,
13752
- },
13753
14136
  "thedrummer/unslopnemo-12b": {
13754
14137
  id: "thedrummer/unslopnemo-12b",
13755
14138
  name: "TheDrummer: UnslopNemo 12B",
@@ -13883,7 +14266,7 @@ export const MODELS = {
13883
14266
  cacheRead: 0.0028,
13884
14267
  cacheWrite: 0,
13885
14268
  },
13886
- contextWindow: 1048576,
14269
+ contextWindow: 1050000,
13887
14270
  maxTokens: 131072,
13888
14271
  },
13889
14272
  "xiaomi/mimo-v2.5-pro": {
@@ -13900,7 +14283,7 @@ export const MODELS = {
13900
14283
  cacheRead: 0.0036,
13901
14284
  cacheWrite: 0,
13902
14285
  },
13903
- contextWindow: 1048576,
14286
+ contextWindow: 1050000,
13904
14287
  maxTokens: 131072,
13905
14288
  },
13906
14289
  "z-ai/glm-4.5": {
@@ -13968,7 +14351,7 @@ export const MODELS = {
13968
14351
  cacheRead: 0.09999999999999999,
13969
14352
  cacheWrite: 0,
13970
14353
  },
13971
- contextWindow: 202752,
14354
+ contextWindow: 204800,
13972
14355
  maxTokens: 131072,
13973
14356
  },
13974
14357
  "z-ai/glm-4.6v": {
@@ -14002,7 +14385,7 @@ export const MODELS = {
14002
14385
  cacheRead: 0.08,
14003
14386
  cacheWrite: 0,
14004
14387
  },
14005
- contextWindow: 202752,
14388
+ contextWindow: 204800,
14006
14389
  maxTokens: 131072,
14007
14390
  },
14008
14391
  "z-ai/glm-4.7-flash": {
@@ -14014,13 +14397,13 @@ export const MODELS = {
14014
14397
  reasoning: true,
14015
14398
  input: ["text"],
14016
14399
  cost: {
14017
- input: 0.060500000000000005,
14400
+ input: 0.06,
14018
14401
  output: 0.39999999999999997,
14019
- cacheRead: 0,
14402
+ cacheRead: 0.01,
14020
14403
  cacheWrite: 0,
14021
14404
  },
14022
- contextWindow: 200000,
14023
- maxTokens: 131072,
14405
+ contextWindow: 202752,
14406
+ maxTokens: 16384,
14024
14407
  },
14025
14408
  "z-ai/glm-5": {
14026
14409
  id: "z-ai/glm-5",
@@ -14070,7 +14453,7 @@ export const MODELS = {
14070
14453
  cacheRead: 0.1794,
14071
14454
  cacheWrite: 0,
14072
14455
  },
14073
- contextWindow: 202752,
14456
+ contextWindow: 204800,
14074
14457
  maxTokens: 128000,
14075
14458
  },
14076
14459
  "z-ai/glm-5.2": {
@@ -14082,9 +14465,9 @@ export const MODELS = {
14082
14465
  reasoning: true,
14083
14466
  input: ["text"],
14084
14467
  cost: {
14085
- input: 0.9786,
14086
- output: 3.0755999999999997,
14087
- cacheRead: 0.18174,
14468
+ input: 0.6692,
14469
+ output: 2.1032,
14470
+ cacheRead: 0.12428,
14088
14471
  cacheWrite: 0,
14089
14472
  },
14090
14473
  contextWindow: 1048576,
@@ -14113,6 +14496,7 @@ export const MODELS = {
14113
14496
  api: "openai-completions",
14114
14497
  provider: "openrouter",
14115
14498
  baseUrl: "https://openrouter.ai/api/v1",
14499
+ compat: { "cacheControlFormat": "anthropic" },
14116
14500
  reasoning: true,
14117
14501
  input: ["text", "image"],
14118
14502
  cost: {
@@ -14130,6 +14514,7 @@ export const MODELS = {
14130
14514
  api: "openai-completions",
14131
14515
  provider: "openrouter",
14132
14516
  baseUrl: "https://openrouter.ai/api/v1",
14517
+ compat: { "cacheControlFormat": "anthropic" },
14133
14518
  reasoning: true,
14134
14519
  input: ["text", "image"],
14135
14520
  cost: {
@@ -14147,6 +14532,7 @@ export const MODELS = {
14147
14532
  api: "openai-completions",
14148
14533
  provider: "openrouter",
14149
14534
  baseUrl: "https://openrouter.ai/api/v1",
14535
+ compat: { "cacheControlFormat": "anthropic" },
14150
14536
  reasoning: true,
14151
14537
  input: ["text", "image"],
14152
14538
  cost: {
@@ -14164,6 +14550,7 @@ export const MODELS = {
14164
14550
  api: "openai-completions",
14165
14551
  provider: "openrouter",
14166
14552
  baseUrl: "https://openrouter.ai/api/v1",
14553
+ compat: { "cacheControlFormat": "anthropic" },
14167
14554
  reasoning: true,
14168
14555
  input: ["text", "image"],
14169
14556
  cost: {
@@ -14185,7 +14572,7 @@ export const MODELS = {
14185
14572
  input: ["text", "image"],
14186
14573
  cost: {
14187
14574
  input: 1.5,
14188
- output: 9,
14575
+ output: 7.5,
14189
14576
  cacheRead: 0.15,
14190
14577
  cacheWrite: 0.08333333333333334,
14191
14578
  },
@@ -14223,59 +14610,607 @@ export const MODELS = {
14223
14610
  cacheRead: 0.3,
14224
14611
  cacheWrite: 0,
14225
14612
  },
14226
- contextWindow: 1048576,
14227
- maxTokens: 4096,
14613
+ contextWindow: 1048576,
14614
+ maxTokens: 4096,
14615
+ },
14616
+ "~openai/gpt-latest": {
14617
+ id: "~openai/gpt-latest",
14618
+ name: "OpenAI GPT Latest",
14619
+ api: "openai-completions",
14620
+ provider: "openrouter",
14621
+ baseUrl: "https://openrouter.ai/api/v1",
14622
+ reasoning: true,
14623
+ input: ["text", "image"],
14624
+ cost: {
14625
+ input: 5,
14626
+ output: 30,
14627
+ cacheRead: 0.5,
14628
+ cacheWrite: 6.25,
14629
+ },
14630
+ contextWindow: 1050000,
14631
+ maxTokens: 128000,
14632
+ },
14633
+ "~openai/gpt-mini-latest": {
14634
+ id: "~openai/gpt-mini-latest",
14635
+ name: "OpenAI GPT Mini Latest",
14636
+ api: "openai-completions",
14637
+ provider: "openrouter",
14638
+ baseUrl: "https://openrouter.ai/api/v1",
14639
+ reasoning: true,
14640
+ input: ["text", "image"],
14641
+ cost: {
14642
+ input: 0.75,
14643
+ output: 4.5,
14644
+ cacheRead: 0.075,
14645
+ cacheWrite: 0,
14646
+ },
14647
+ contextWindow: 400000,
14648
+ maxTokens: 128000,
14649
+ },
14650
+ "~x-ai/grok-latest": {
14651
+ id: "~x-ai/grok-latest",
14652
+ name: "xAI: Grok Latest",
14653
+ api: "openai-completions",
14654
+ provider: "openrouter",
14655
+ baseUrl: "https://openrouter.ai/api/v1",
14656
+ reasoning: true,
14657
+ input: ["text", "image"],
14658
+ cost: {
14659
+ input: 2,
14660
+ output: 6,
14661
+ cacheRead: 0.3,
14662
+ cacheWrite: 0,
14663
+ },
14664
+ contextWindow: 500000,
14665
+ maxTokens: 4096,
14666
+ },
14667
+ },
14668
+ "qwen-token-plan": {
14669
+ "MiniMax-M2.5": {
14670
+ id: "MiniMax-M2.5",
14671
+ name: "MiniMax-M2.5",
14672
+ api: "openai-completions",
14673
+ provider: "qwen-token-plan",
14674
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14675
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14676
+ reasoning: true,
14677
+ input: ["text"],
14678
+ cost: {
14679
+ input: 0,
14680
+ output: 0,
14681
+ cacheRead: 0,
14682
+ cacheWrite: 0,
14683
+ },
14684
+ contextWindow: 196608,
14685
+ maxTokens: 32768,
14686
+ },
14687
+ "deepseek-v3.2": {
14688
+ id: "deepseek-v3.2",
14689
+ name: "DeepSeek V3.2",
14690
+ api: "openai-completions",
14691
+ provider: "qwen-token-plan",
14692
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14693
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14694
+ reasoning: true,
14695
+ input: ["text"],
14696
+ cost: {
14697
+ input: 0,
14698
+ output: 0,
14699
+ cacheRead: 0,
14700
+ cacheWrite: 0,
14701
+ },
14702
+ contextWindow: 131072,
14703
+ maxTokens: 65536,
14704
+ },
14705
+ "deepseek-v4-flash": {
14706
+ id: "deepseek-v4-flash",
14707
+ name: "DeepSeek V4 Flash",
14708
+ api: "openai-completions",
14709
+ provider: "qwen-token-plan",
14710
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14711
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14712
+ reasoning: true,
14713
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
14714
+ input: ["text"],
14715
+ cost: {
14716
+ input: 0,
14717
+ output: 0,
14718
+ cacheRead: 0,
14719
+ cacheWrite: 0,
14720
+ },
14721
+ contextWindow: 1000000,
14722
+ maxTokens: 384000,
14723
+ },
14724
+ "deepseek-v4-pro": {
14725
+ id: "deepseek-v4-pro",
14726
+ name: "DeepSeek V4 Pro",
14727
+ api: "openai-completions",
14728
+ provider: "qwen-token-plan",
14729
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14730
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14731
+ reasoning: true,
14732
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
14733
+ input: ["text"],
14734
+ cost: {
14735
+ input: 0,
14736
+ output: 0,
14737
+ cacheRead: 0,
14738
+ cacheWrite: 0,
14739
+ },
14740
+ contextWindow: 1000000,
14741
+ maxTokens: 384000,
14742
+ },
14743
+ "glm-5": {
14744
+ id: "glm-5",
14745
+ name: "GLM-5",
14746
+ api: "openai-completions",
14747
+ provider: "qwen-token-plan",
14748
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14749
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14750
+ reasoning: true,
14751
+ input: ["text"],
14752
+ cost: {
14753
+ input: 0,
14754
+ output: 0,
14755
+ cacheRead: 0,
14756
+ cacheWrite: 0,
14757
+ },
14758
+ contextWindow: 202752,
14759
+ maxTokens: 16384,
14760
+ },
14761
+ "glm-5.1": {
14762
+ id: "glm-5.1",
14763
+ name: "GLM-5.1",
14764
+ api: "openai-completions",
14765
+ provider: "qwen-token-plan",
14766
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14767
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14768
+ reasoning: true,
14769
+ input: ["text"],
14770
+ cost: {
14771
+ input: 0,
14772
+ output: 0,
14773
+ cacheRead: 0,
14774
+ cacheWrite: 0,
14775
+ },
14776
+ contextWindow: 202752,
14777
+ maxTokens: 128000,
14778
+ },
14779
+ "glm-5.2": {
14780
+ id: "glm-5.2",
14781
+ name: "GLM-5.2",
14782
+ api: "openai-completions",
14783
+ provider: "qwen-token-plan",
14784
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14785
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14786
+ reasoning: true,
14787
+ input: ["text"],
14788
+ cost: {
14789
+ input: 0,
14790
+ output: 0,
14791
+ cacheRead: 0,
14792
+ cacheWrite: 0,
14793
+ },
14794
+ contextWindow: 1000000,
14795
+ maxTokens: 131072,
14796
+ },
14797
+ "kimi-k2.5": {
14798
+ id: "kimi-k2.5",
14799
+ name: "Kimi K2.5",
14800
+ api: "openai-completions",
14801
+ provider: "qwen-token-plan",
14802
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14803
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14804
+ reasoning: true,
14805
+ input: ["text", "image"],
14806
+ cost: {
14807
+ input: 0,
14808
+ output: 0,
14809
+ cacheRead: 0,
14810
+ cacheWrite: 0,
14811
+ },
14812
+ contextWindow: 262144,
14813
+ maxTokens: 98304,
14814
+ },
14815
+ "kimi-k2.6": {
14816
+ id: "kimi-k2.6",
14817
+ name: "Kimi K2.6",
14818
+ api: "openai-completions",
14819
+ provider: "qwen-token-plan",
14820
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14821
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14822
+ reasoning: true,
14823
+ input: ["text", "image"],
14824
+ cost: {
14825
+ input: 0,
14826
+ output: 0,
14827
+ cacheRead: 0,
14828
+ cacheWrite: 0,
14829
+ },
14830
+ contextWindow: 262144,
14831
+ maxTokens: 262144,
14832
+ },
14833
+ "kimi-k2.7-code": {
14834
+ id: "kimi-k2.7-code",
14835
+ name: "Kimi K2.7 Code",
14836
+ api: "openai-completions",
14837
+ provider: "qwen-token-plan",
14838
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14839
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14840
+ reasoning: true,
14841
+ input: ["text", "image"],
14842
+ cost: {
14843
+ input: 0,
14844
+ output: 0,
14845
+ cacheRead: 0,
14846
+ cacheWrite: 0,
14847
+ },
14848
+ contextWindow: 262144,
14849
+ maxTokens: 262144,
14850
+ },
14851
+ "qwen3.6-flash": {
14852
+ id: "qwen3.6-flash",
14853
+ name: "Qwen3.6 Flash",
14854
+ api: "openai-completions",
14855
+ provider: "qwen-token-plan",
14856
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14857
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14858
+ reasoning: true,
14859
+ input: ["text", "image"],
14860
+ cost: {
14861
+ input: 0,
14862
+ output: 0,
14863
+ cacheRead: 0,
14864
+ cacheWrite: 0,
14865
+ },
14866
+ contextWindow: 1000000,
14867
+ maxTokens: 65536,
14868
+ },
14869
+ "qwen3.6-plus": {
14870
+ id: "qwen3.6-plus",
14871
+ name: "Qwen3.6 Plus",
14872
+ api: "openai-completions",
14873
+ provider: "qwen-token-plan",
14874
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14875
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14876
+ reasoning: true,
14877
+ input: ["text", "image"],
14878
+ cost: {
14879
+ input: 0,
14880
+ output: 0,
14881
+ cacheRead: 0,
14882
+ cacheWrite: 0,
14883
+ },
14884
+ contextWindow: 1000000,
14885
+ maxTokens: 65536,
14886
+ },
14887
+ "qwen3.7-max": {
14888
+ id: "qwen3.7-max",
14889
+ name: "Qwen3.7 Max",
14890
+ api: "openai-completions",
14891
+ provider: "qwen-token-plan",
14892
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14893
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14894
+ reasoning: true,
14895
+ input: ["text"],
14896
+ cost: {
14897
+ input: 0,
14898
+ output: 0,
14899
+ cacheRead: 0,
14900
+ cacheWrite: 0,
14901
+ },
14902
+ contextWindow: 1000000,
14903
+ maxTokens: 131072,
14904
+ },
14905
+ "qwen3.7-plus": {
14906
+ id: "qwen3.7-plus",
14907
+ name: "Qwen3.7 Plus",
14908
+ api: "openai-completions",
14909
+ provider: "qwen-token-plan",
14910
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14911
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14912
+ reasoning: true,
14913
+ input: ["text", "image"],
14914
+ cost: {
14915
+ input: 0,
14916
+ output: 0,
14917
+ cacheRead: 0,
14918
+ cacheWrite: 0,
14919
+ },
14920
+ contextWindow: 1000000,
14921
+ maxTokens: 65536,
14922
+ },
14923
+ "qwen3.8-max-preview": {
14924
+ id: "qwen3.8-max-preview",
14925
+ name: "Qwen3.8 Max Preview",
14926
+ api: "openai-completions",
14927
+ provider: "qwen-token-plan",
14928
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14929
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14930
+ reasoning: true,
14931
+ input: ["text", "image"],
14932
+ cost: {
14933
+ input: 0,
14934
+ output: 0,
14935
+ cacheRead: 0,
14936
+ cacheWrite: 0,
14937
+ },
14938
+ contextWindow: 1000000,
14939
+ maxTokens: 131072,
14940
+ },
14941
+ },
14942
+ "qwen-token-plan-cn": {
14943
+ "MiniMax-M2.5": {
14944
+ id: "MiniMax-M2.5",
14945
+ name: "MiniMax-M2.5",
14946
+ api: "openai-completions",
14947
+ provider: "qwen-token-plan-cn",
14948
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
14949
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14950
+ reasoning: true,
14951
+ input: ["text"],
14952
+ cost: {
14953
+ input: 0,
14954
+ output: 0,
14955
+ cacheRead: 0,
14956
+ cacheWrite: 0,
14957
+ },
14958
+ contextWindow: 196608,
14959
+ maxTokens: 32768,
14960
+ },
14961
+ "deepseek-v3.2": {
14962
+ id: "deepseek-v3.2",
14963
+ name: "DeepSeek V3.2",
14964
+ api: "openai-completions",
14965
+ provider: "qwen-token-plan-cn",
14966
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
14967
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14968
+ reasoning: true,
14969
+ input: ["text"],
14970
+ cost: {
14971
+ input: 0,
14972
+ output: 0,
14973
+ cacheRead: 0,
14974
+ cacheWrite: 0,
14975
+ },
14976
+ contextWindow: 131072,
14977
+ maxTokens: 65536,
14978
+ },
14979
+ "deepseek-v4-flash": {
14980
+ id: "deepseek-v4-flash",
14981
+ name: "DeepSeek V4 Flash",
14982
+ api: "openai-completions",
14983
+ provider: "qwen-token-plan-cn",
14984
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
14985
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14986
+ reasoning: true,
14987
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
14988
+ input: ["text"],
14989
+ cost: {
14990
+ input: 0,
14991
+ output: 0,
14992
+ cacheRead: 0,
14993
+ cacheWrite: 0,
14994
+ },
14995
+ contextWindow: 1000000,
14996
+ maxTokens: 384000,
14997
+ },
14998
+ "deepseek-v4-pro": {
14999
+ id: "deepseek-v4-pro",
15000
+ name: "DeepSeek V4 Pro",
15001
+ api: "openai-completions",
15002
+ provider: "qwen-token-plan-cn",
15003
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15004
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15005
+ reasoning: true,
15006
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
15007
+ input: ["text"],
15008
+ cost: {
15009
+ input: 0,
15010
+ output: 0,
15011
+ cacheRead: 0,
15012
+ cacheWrite: 0,
15013
+ },
15014
+ contextWindow: 1000000,
15015
+ maxTokens: 384000,
15016
+ },
15017
+ "glm-5": {
15018
+ id: "glm-5",
15019
+ name: "GLM-5",
15020
+ api: "openai-completions",
15021
+ provider: "qwen-token-plan-cn",
15022
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15023
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15024
+ reasoning: true,
15025
+ input: ["text"],
15026
+ cost: {
15027
+ input: 0,
15028
+ output: 0,
15029
+ cacheRead: 0,
15030
+ cacheWrite: 0,
15031
+ },
15032
+ contextWindow: 202752,
15033
+ maxTokens: 16384,
15034
+ },
15035
+ "glm-5.1": {
15036
+ id: "glm-5.1",
15037
+ name: "GLM-5.1",
15038
+ api: "openai-completions",
15039
+ provider: "qwen-token-plan-cn",
15040
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15041
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15042
+ reasoning: true,
15043
+ input: ["text"],
15044
+ cost: {
15045
+ input: 0,
15046
+ output: 0,
15047
+ cacheRead: 0,
15048
+ cacheWrite: 0,
15049
+ },
15050
+ contextWindow: 202752,
15051
+ maxTokens: 128000,
15052
+ },
15053
+ "glm-5.2": {
15054
+ id: "glm-5.2",
15055
+ name: "GLM-5.2",
15056
+ api: "openai-completions",
15057
+ provider: "qwen-token-plan-cn",
15058
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15059
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15060
+ reasoning: true,
15061
+ input: ["text"],
15062
+ cost: {
15063
+ input: 0,
15064
+ output: 0,
15065
+ cacheRead: 0,
15066
+ cacheWrite: 0,
15067
+ },
15068
+ contextWindow: 1000000,
15069
+ maxTokens: 131072,
15070
+ },
15071
+ "kimi-k2.5": {
15072
+ id: "kimi-k2.5",
15073
+ name: "Kimi K2.5",
15074
+ api: "openai-completions",
15075
+ provider: "qwen-token-plan-cn",
15076
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15077
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15078
+ reasoning: true,
15079
+ input: ["text", "image"],
15080
+ cost: {
15081
+ input: 0,
15082
+ output: 0,
15083
+ cacheRead: 0,
15084
+ cacheWrite: 0,
15085
+ },
15086
+ contextWindow: 262144,
15087
+ maxTokens: 98304,
15088
+ },
15089
+ "kimi-k2.6": {
15090
+ id: "kimi-k2.6",
15091
+ name: "Kimi K2.6",
15092
+ api: "openai-completions",
15093
+ provider: "qwen-token-plan-cn",
15094
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15095
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15096
+ reasoning: true,
15097
+ input: ["text", "image"],
15098
+ cost: {
15099
+ input: 0,
15100
+ output: 0,
15101
+ cacheRead: 0,
15102
+ cacheWrite: 0,
15103
+ },
15104
+ contextWindow: 262144,
15105
+ maxTokens: 262144,
15106
+ },
15107
+ "kimi-k2.7-code": {
15108
+ id: "kimi-k2.7-code",
15109
+ name: "Kimi K2.7 Code",
15110
+ api: "openai-completions",
15111
+ provider: "qwen-token-plan-cn",
15112
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15113
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15114
+ reasoning: true,
15115
+ input: ["text", "image"],
15116
+ cost: {
15117
+ input: 0,
15118
+ output: 0,
15119
+ cacheRead: 0,
15120
+ cacheWrite: 0,
15121
+ },
15122
+ contextWindow: 262144,
15123
+ maxTokens: 262144,
15124
+ },
15125
+ "qwen3.6-flash": {
15126
+ id: "qwen3.6-flash",
15127
+ name: "Qwen3.6 Flash",
15128
+ api: "openai-completions",
15129
+ provider: "qwen-token-plan-cn",
15130
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15131
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15132
+ reasoning: true,
15133
+ input: ["text", "image"],
15134
+ cost: {
15135
+ input: 0,
15136
+ output: 0,
15137
+ cacheRead: 0,
15138
+ cacheWrite: 0,
15139
+ },
15140
+ contextWindow: 1000000,
15141
+ maxTokens: 65536,
15142
+ },
15143
+ "qwen3.6-plus": {
15144
+ id: "qwen3.6-plus",
15145
+ name: "Qwen3.6 Plus",
15146
+ api: "openai-completions",
15147
+ provider: "qwen-token-plan-cn",
15148
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15149
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15150
+ reasoning: true,
15151
+ input: ["text", "image"],
15152
+ cost: {
15153
+ input: 0,
15154
+ output: 0,
15155
+ cacheRead: 0,
15156
+ cacheWrite: 0,
15157
+ },
15158
+ contextWindow: 1000000,
15159
+ maxTokens: 65536,
14228
15160
  },
14229
- "~openai/gpt-latest": {
14230
- id: "~openai/gpt-latest",
14231
- name: "OpenAI GPT Latest",
15161
+ "qwen3.7-max": {
15162
+ id: "qwen3.7-max",
15163
+ name: "Qwen3.7 Max",
14232
15164
  api: "openai-completions",
14233
- provider: "openrouter",
14234
- baseUrl: "https://openrouter.ai/api/v1",
15165
+ provider: "qwen-token-plan-cn",
15166
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15167
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14235
15168
  reasoning: true,
14236
- input: ["text", "image"],
15169
+ input: ["text"],
14237
15170
  cost: {
14238
- input: 5,
14239
- output: 30,
14240
- cacheRead: 0.5,
14241
- cacheWrite: 6.25,
15171
+ input: 0,
15172
+ output: 0,
15173
+ cacheRead: 0,
15174
+ cacheWrite: 0,
14242
15175
  },
14243
- contextWindow: 1050000,
14244
- maxTokens: 128000,
15176
+ contextWindow: 1000000,
15177
+ maxTokens: 131072,
14245
15178
  },
14246
- "~openai/gpt-mini-latest": {
14247
- id: "~openai/gpt-mini-latest",
14248
- name: "OpenAI GPT Mini Latest",
15179
+ "qwen3.7-plus": {
15180
+ id: "qwen3.7-plus",
15181
+ name: "Qwen3.7 Plus",
14249
15182
  api: "openai-completions",
14250
- provider: "openrouter",
14251
- baseUrl: "https://openrouter.ai/api/v1",
15183
+ provider: "qwen-token-plan-cn",
15184
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15185
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14252
15186
  reasoning: true,
14253
15187
  input: ["text", "image"],
14254
15188
  cost: {
14255
- input: 0.75,
14256
- output: 4.5,
14257
- cacheRead: 0.075,
15189
+ input: 0,
15190
+ output: 0,
15191
+ cacheRead: 0,
14258
15192
  cacheWrite: 0,
14259
15193
  },
14260
- contextWindow: 400000,
14261
- maxTokens: 128000,
15194
+ contextWindow: 1000000,
15195
+ maxTokens: 65536,
14262
15196
  },
14263
- "~x-ai/grok-latest": {
14264
- id: "~x-ai/grok-latest",
14265
- name: "xAI: Grok Latest",
15197
+ "qwen3.8-max-preview": {
15198
+ id: "qwen3.8-max-preview",
15199
+ name: "Qwen3.8 Max Preview",
14266
15200
  api: "openai-completions",
14267
- provider: "openrouter",
14268
- baseUrl: "https://openrouter.ai/api/v1",
15201
+ provider: "qwen-token-plan-cn",
15202
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15203
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14269
15204
  reasoning: true,
14270
15205
  input: ["text", "image"],
14271
15206
  cost: {
14272
- input: 2,
14273
- output: 6,
14274
- cacheRead: 0.3,
15207
+ input: 0,
15208
+ output: 0,
15209
+ cacheRead: 0,
14275
15210
  cacheWrite: 0,
14276
15211
  },
14277
- contextWindow: 500000,
14278
- maxTokens: 4096,
15212
+ contextWindow: 1000000,
15213
+ maxTokens: 131072,
14279
15214
  },
14280
15215
  },
14281
15216
  "together": {
@@ -14965,10 +15900,10 @@ export const MODELS = {
14965
15900
  reasoning: true,
14966
15901
  input: ["text"],
14967
15902
  cost: {
14968
- input: 1.25,
14969
- output: 3.75,
14970
- cacheRead: 0.25,
14971
- cacheWrite: 1.5625,
15903
+ input: 2.5,
15904
+ output: 7.5,
15905
+ cacheRead: 0.5,
15906
+ cacheWrite: 3.125,
14972
15907
  },
14973
15908
  contextWindow: 991000,
14974
15909
  maxTokens: 64000,
@@ -15081,7 +16016,9 @@ export const MODELS = {
15081
16016
  api: "anthropic-messages",
15082
16017
  provider: "vercel-ai-gateway",
15083
16018
  baseUrl: "https://ai-gateway.vercel.sh",
16019
+ compat: { "forceAdaptiveThinking": true },
15084
16020
  reasoning: true,
16021
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
15085
16022
  input: ["text", "image"],
15086
16023
  cost: {
15087
16024
  input: 10,
@@ -15168,7 +16105,7 @@ export const MODELS = {
15168
16105
  baseUrl: "https://ai-gateway.vercel.sh",
15169
16106
  compat: { "forceAdaptiveThinking": true },
15170
16107
  reasoning: true,
15171
- thinkingLevelMap: { "xhigh": "max" },
16108
+ thinkingLevelMap: { "max": "max" },
15172
16109
  input: ["text", "image"],
15173
16110
  cost: {
15174
16111
  input: 5,
@@ -15187,7 +16124,7 @@ export const MODELS = {
15187
16124
  baseUrl: "https://ai-gateway.vercel.sh",
15188
16125
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
15189
16126
  reasoning: true,
15190
- thinkingLevelMap: { "xhigh": "xhigh" },
16127
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
15191
16128
  input: ["text", "image"],
15192
16129
  cost: {
15193
16130
  input: 5,
@@ -15198,34 +16135,53 @@ export const MODELS = {
15198
16135
  contextWindow: 1000000,
15199
16136
  maxTokens: 128000,
15200
16137
  },
15201
- "anthropic/claude-opus-4.7-fast": {
15202
- id: "anthropic/claude-opus-4.7-fast",
15203
- name: "Claude Opus 4.7 (Fast)",
16138
+ "anthropic/claude-opus-4.8": {
16139
+ id: "anthropic/claude-opus-4.8",
16140
+ name: "Claude Opus 4.8",
15204
16141
  api: "anthropic-messages",
15205
16142
  provider: "vercel-ai-gateway",
15206
16143
  baseUrl: "https://ai-gateway.vercel.sh",
15207
16144
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
15208
16145
  reasoning: true,
15209
- thinkingLevelMap: { "xhigh": "xhigh" },
16146
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
15210
16147
  input: ["text", "image"],
15211
16148
  cost: {
15212
- input: 30,
15213
- output: 150,
15214
- cacheRead: 3,
15215
- cacheWrite: 37.5,
16149
+ input: 5,
16150
+ output: 25,
16151
+ cacheRead: 0.5,
16152
+ cacheWrite: 6.25,
15216
16153
  },
15217
16154
  contextWindow: 1000000,
15218
16155
  maxTokens: 128000,
15219
16156
  },
15220
- "anthropic/claude-opus-4.8": {
15221
- id: "anthropic/claude-opus-4.8",
15222
- name: "Claude Opus 4.8",
16157
+ "anthropic/claude-opus-4.8-fast": {
16158
+ id: "anthropic/claude-opus-4.8-fast",
16159
+ name: "Claude Opus 4.8 (Fast)",
15223
16160
  api: "anthropic-messages",
15224
16161
  provider: "vercel-ai-gateway",
15225
16162
  baseUrl: "https://ai-gateway.vercel.sh",
15226
16163
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
15227
16164
  reasoning: true,
15228
- thinkingLevelMap: { "xhigh": "xhigh" },
16165
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
16166
+ input: ["text", "image"],
16167
+ cost: {
16168
+ input: 10,
16169
+ output: 50,
16170
+ cacheRead: 1,
16171
+ cacheWrite: 12.5,
16172
+ },
16173
+ contextWindow: 1000000,
16174
+ maxTokens: 128000,
16175
+ },
16176
+ "anthropic/claude-opus-5": {
16177
+ id: "anthropic/claude-opus-5",
16178
+ name: "Claude Opus 5",
16179
+ api: "anthropic-messages",
16180
+ provider: "vercel-ai-gateway",
16181
+ baseUrl: "https://ai-gateway.vercel.sh",
16182
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
16183
+ reasoning: true,
16184
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
15229
16185
  input: ["text", "image"],
15230
16186
  cost: {
15231
16187
  input: 5,
@@ -15236,15 +16192,15 @@ export const MODELS = {
15236
16192
  contextWindow: 1000000,
15237
16193
  maxTokens: 128000,
15238
16194
  },
15239
- "anthropic/claude-opus-4.8-fast": {
15240
- id: "anthropic/claude-opus-4.8-fast",
15241
- name: "Claude Opus 4.8 (Fast)",
16195
+ "anthropic/claude-opus-5-fast": {
16196
+ id: "anthropic/claude-opus-5-fast",
16197
+ name: "Claude Opus 5 (Fast)",
15242
16198
  api: "anthropic-messages",
15243
16199
  provider: "vercel-ai-gateway",
15244
16200
  baseUrl: "https://ai-gateway.vercel.sh",
15245
16201
  compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
15246
16202
  reasoning: true,
15247
- thinkingLevelMap: { "xhigh": "xhigh" },
16203
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
15248
16204
  input: ["text", "image"],
15249
16205
  cost: {
15250
16206
  input: 10,
@@ -15297,6 +16253,7 @@ export const MODELS = {
15297
16253
  baseUrl: "https://ai-gateway.vercel.sh",
15298
16254
  compat: { "forceAdaptiveThinking": true },
15299
16255
  reasoning: true,
16256
+ thinkingLevelMap: { "max": "max" },
15300
16257
  input: ["text", "image"],
15301
16258
  cost: {
15302
16259
  input: 3,
@@ -15313,7 +16270,9 @@ export const MODELS = {
15313
16270
  api: "anthropic-messages",
15314
16271
  provider: "vercel-ai-gateway",
15315
16272
  baseUrl: "https://ai-gateway.vercel.sh",
16273
+ compat: { "forceAdaptiveThinking": true },
15316
16274
  reasoning: true,
16275
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
15317
16276
  input: ["text", "image"],
15318
16277
  cost: {
15319
16278
  input: 2,
@@ -15647,23 +16606,6 @@ export const MODELS = {
15647
16606
  contextWindow: 1000000,
15648
16607
  maxTokens: 65000,
15649
16608
  },
15650
- "google/gemini-3.1-flash-lite-preview": {
15651
- id: "google/gemini-3.1-flash-lite-preview",
15652
- name: "Gemini 3.1 Flash Lite Preview",
15653
- api: "anthropic-messages",
15654
- provider: "vercel-ai-gateway",
15655
- baseUrl: "https://ai-gateway.vercel.sh",
15656
- reasoning: true,
15657
- input: ["text", "image"],
15658
- cost: {
15659
- input: 0.25,
15660
- output: 1.5,
15661
- cacheRead: 0.03,
15662
- cacheWrite: 0,
15663
- },
15664
- contextWindow: 1000000,
15665
- maxTokens: 65000,
15666
- },
15667
16609
  "google/gemini-3.1-pro-preview": {
15668
16610
  id: "google/gemini-3.1-pro-preview",
15669
16611
  name: "Gemini 3.1 Pro Preview",
@@ -15698,6 +16640,40 @@ export const MODELS = {
15698
16640
  contextWindow: 1000000,
15699
16641
  maxTokens: 64000,
15700
16642
  },
16643
+ "google/gemini-3.5-flash-lite": {
16644
+ id: "google/gemini-3.5-flash-lite",
16645
+ name: "Gemini 3.5 Flash Lite",
16646
+ api: "anthropic-messages",
16647
+ provider: "vercel-ai-gateway",
16648
+ baseUrl: "https://ai-gateway.vercel.sh",
16649
+ reasoning: true,
16650
+ input: ["text", "image"],
16651
+ cost: {
16652
+ input: 0.3,
16653
+ output: 2.5,
16654
+ cacheRead: 0.03,
16655
+ cacheWrite: 0,
16656
+ },
16657
+ contextWindow: 1000000,
16658
+ maxTokens: 65000,
16659
+ },
16660
+ "google/gemini-3.6-flash": {
16661
+ id: "google/gemini-3.6-flash",
16662
+ name: "Gemini 3.6 Flash",
16663
+ api: "anthropic-messages",
16664
+ provider: "vercel-ai-gateway",
16665
+ baseUrl: "https://ai-gateway.vercel.sh",
16666
+ reasoning: true,
16667
+ input: ["text", "image"],
16668
+ cost: {
16669
+ input: 1.5,
16670
+ output: 7.5,
16671
+ cacheRead: 0.15,
16672
+ cacheWrite: 0,
16673
+ },
16674
+ contextWindow: 1000000,
16675
+ maxTokens: 64000,
16676
+ },
15701
16677
  "google/gemma-4-26b-a4b-it": {
15702
16678
  id: "google/gemma-4-26b-a4b-it",
15703
16679
  name: "Gemma 4 26B A4B IT",
@@ -15766,6 +16742,23 @@ export const MODELS = {
15766
16742
  contextWindow: 32000,
15767
16743
  maxTokens: 16384,
15768
16744
  },
16745
+ "inclusionai/ling-3.0-flash-free": {
16746
+ id: "inclusionai/ling-3.0-flash-free",
16747
+ name: "Ling 3.0 Flash",
16748
+ api: "anthropic-messages",
16749
+ provider: "vercel-ai-gateway",
16750
+ baseUrl: "https://ai-gateway.vercel.sh",
16751
+ reasoning: true,
16752
+ input: ["text"],
16753
+ cost: {
16754
+ input: 0,
16755
+ output: 0,
16756
+ cacheRead: 0,
16757
+ cacheWrite: 0,
16758
+ },
16759
+ contextWindow: 256000,
16760
+ maxTokens: 256000,
16761
+ },
15769
16762
  "interfaze/interfaze-beta": {
15770
16763
  id: "interfaze/interfaze-beta",
15771
16764
  name: "Interfaze Beta",
@@ -16667,23 +17660,6 @@ export const MODELS = {
16667
17660
  contextWindow: 400000,
16668
17661
  maxTokens: 128000,
16669
17662
  },
16670
- "openai/gpt-5-chat": {
16671
- id: "openai/gpt-5-chat",
16672
- name: "GPT 5 Chat",
16673
- api: "anthropic-messages",
16674
- provider: "vercel-ai-gateway",
16675
- baseUrl: "https://ai-gateway.vercel.sh",
16676
- reasoning: false,
16677
- input: ["text", "image"],
16678
- cost: {
16679
- input: 1.25,
16680
- output: 10,
16681
- cacheRead: 0.125,
16682
- cacheWrite: 0,
16683
- },
16684
- contextWindow: 128000,
16685
- maxTokens: 16384,
16686
- },
16687
17663
  "openai/gpt-5-codex": {
16688
17664
  id: "openai/gpt-5-codex",
16689
17665
  name: "GPT-5-Codex",
@@ -16855,24 +17831,6 @@ export const MODELS = {
16855
17831
  contextWindow: 400000,
16856
17832
  maxTokens: 128000,
16857
17833
  },
16858
- "openai/gpt-5.2-chat": {
16859
- id: "openai/gpt-5.2-chat",
16860
- name: "GPT 5.2 Chat",
16861
- api: "anthropic-messages",
16862
- provider: "vercel-ai-gateway",
16863
- baseUrl: "https://ai-gateway.vercel.sh",
16864
- reasoning: false,
16865
- thinkingLevelMap: { "xhigh": "xhigh" },
16866
- input: ["text", "image"],
16867
- cost: {
16868
- input: 1.75,
16869
- output: 14,
16870
- cacheRead: 0.175,
16871
- cacheWrite: 0,
16872
- },
16873
- contextWindow: 128000,
16874
- maxTokens: 16384,
16875
- },
16876
17834
  "openai/gpt-5.2-codex": {
16877
17835
  id: "openai/gpt-5.2-codex",
16878
17836
  name: "GPT 5.2 Codex",
@@ -17260,6 +18218,40 @@ export const MODELS = {
17260
18218
  contextWindow: 200000,
17261
18219
  maxTokens: 100000,
17262
18220
  },
18221
+ "poolside/laguna-s-2.1": {
18222
+ id: "poolside/laguna-s-2.1",
18223
+ name: "Laguna S 2.1",
18224
+ api: "anthropic-messages",
18225
+ provider: "vercel-ai-gateway",
18226
+ baseUrl: "https://ai-gateway.vercel.sh",
18227
+ reasoning: true,
18228
+ input: ["text"],
18229
+ cost: {
18230
+ input: 0.09999999999999999,
18231
+ output: 0.19999999999999998,
18232
+ cacheRead: 0.01,
18233
+ cacheWrite: 0,
18234
+ },
18235
+ contextWindow: 1000000,
18236
+ maxTokens: 131072,
18237
+ },
18238
+ "poolside/laguna-s-2.1-free": {
18239
+ id: "poolside/laguna-s-2.1-free",
18240
+ name: "Laguna S 2.1 Free",
18241
+ api: "anthropic-messages",
18242
+ provider: "vercel-ai-gateway",
18243
+ baseUrl: "https://ai-gateway.vercel.sh",
18244
+ reasoning: true,
18245
+ input: ["text"],
18246
+ cost: {
18247
+ input: 0,
18248
+ output: 0,
18249
+ cacheRead: 0,
18250
+ cacheWrite: 0,
18251
+ },
18252
+ contextWindow: 256000,
18253
+ maxTokens: 32768,
18254
+ },
17263
18255
  "sakana/fugu-ultra": {
17264
18256
  id: "sakana/fugu-ultra",
17265
18257
  name: "Fugu Ultra",
@@ -17311,6 +18303,23 @@ export const MODELS = {
17311
18303
  contextWindow: 256000,
17312
18304
  maxTokens: 256000,
17313
18305
  },
18306
+ "tencent/hy3": {
18307
+ id: "tencent/hy3",
18308
+ name: "Hy3",
18309
+ api: "anthropic-messages",
18310
+ provider: "vercel-ai-gateway",
18311
+ baseUrl: "https://ai-gateway.vercel.sh",
18312
+ reasoning: true,
18313
+ input: ["text"],
18314
+ cost: {
18315
+ input: 0.14,
18316
+ output: 0.58,
18317
+ cacheRead: 0.035,
18318
+ cacheWrite: 0,
18319
+ },
18320
+ contextWindow: 262144,
18321
+ maxTokens: 262144,
18322
+ },
17314
18323
  "thinkingmachines/inkling": {
17315
18324
  id: "thinkingmachines/inkling",
17316
18325
  name: "Inkling",
@@ -17894,10 +18903,12 @@ export const MODELS = {
17894
18903
  "grok-4.5": {
17895
18904
  id: "grok-4.5",
17896
18905
  name: "Grok 4.5",
17897
- api: "openai-completions",
18906
+ api: "openai-responses",
17898
18907
  provider: "xai",
17899
18908
  baseUrl: "https://api.x.ai/v1",
18909
+ compat: { "supportsLongCacheRetention": false },
17900
18910
  reasoning: true,
18911
+ thinkingLevelMap: { "off": null, "minimal": null },
17901
18912
  input: ["text", "image"],
17902
18913
  cost: {
17903
18914
  input: 2,