@earendil-works/pi-ai 0.80.1 → 0.80.3

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 (148) hide show
  1. package/README.md +14 -1
  2. package/dist/api/anthropic-messages.d.ts.map +1 -1
  3. package/dist/api/anthropic-messages.js +17 -10
  4. package/dist/api/anthropic-messages.js.map +1 -1
  5. package/dist/api/azure-openai-responses.d.ts.map +1 -1
  6. package/dist/api/azure-openai-responses.js +11 -17
  7. package/dist/api/azure-openai-responses.js.map +1 -1
  8. package/dist/api/bedrock-converse-stream.d.ts.map +1 -1
  9. package/dist/api/bedrock-converse-stream.js +17 -8
  10. package/dist/api/bedrock-converse-stream.js.map +1 -1
  11. package/dist/api/google-generative-ai.d.ts.map +1 -1
  12. package/dist/api/google-generative-ai.js +4 -2
  13. package/dist/api/google-generative-ai.js.map +1 -1
  14. package/dist/api/google-vertex.d.ts.map +1 -1
  15. package/dist/api/google-vertex.js +4 -2
  16. package/dist/api/google-vertex.js.map +1 -1
  17. package/dist/api/mistral-conversations.d.ts.map +1 -1
  18. package/dist/api/mistral-conversations.js +1 -1
  19. package/dist/api/mistral-conversations.js.map +1 -1
  20. package/dist/api/openai-codex-responses.d.ts.map +1 -1
  21. package/dist/api/openai-codex-responses.js +11 -25
  22. package/dist/api/openai-codex-responses.js.map +1 -1
  23. package/dist/api/openai-completions.d.ts.map +1 -1
  24. package/dist/api/openai-completions.js +105 -44
  25. package/dist/api/openai-completions.js.map +1 -1
  26. package/dist/api/openai-responses-shared.d.ts.map +1 -1
  27. package/dist/api/openai-responses-shared.js +152 -175
  28. package/dist/api/openai-responses-shared.js.map +1 -1
  29. package/dist/api/openai-responses.d.ts.map +1 -1
  30. package/dist/api/openai-responses.js +3 -15
  31. package/dist/api/openai-responses.js.map +1 -1
  32. package/dist/api/openrouter-images.d.ts.map +1 -1
  33. package/dist/api/openrouter-images.js +2 -1
  34. package/dist/api/openrouter-images.js.map +1 -1
  35. package/dist/api/simple-options.d.ts +3 -2
  36. package/dist/api/simple-options.d.ts.map +1 -1
  37. package/dist/api/simple-options.js +11 -2
  38. package/dist/api/simple-options.js.map +1 -1
  39. package/dist/auth/helpers.d.ts +1 -1
  40. package/dist/auth/helpers.d.ts.map +1 -1
  41. package/dist/auth/helpers.js +2 -2
  42. package/dist/auth/helpers.js.map +1 -1
  43. package/dist/auth/resolve.d.ts +6 -2
  44. package/dist/auth/resolve.d.ts.map +1 -1
  45. package/dist/auth/resolve.js +19 -4
  46. package/dist/auth/resolve.js.map +1 -1
  47. package/dist/auth/types.d.ts +7 -7
  48. package/dist/auth/types.d.ts.map +1 -1
  49. package/dist/auth/types.js.map +1 -1
  50. package/dist/compat.d.ts +1 -0
  51. package/dist/compat.d.ts.map +1 -1
  52. package/dist/compat.js +1 -0
  53. package/dist/compat.js.map +1 -1
  54. package/dist/image-models.generated.d.ts +34 -19
  55. package/dist/image-models.generated.d.ts.map +1 -1
  56. package/dist/image-models.generated.js +60 -45
  57. package/dist/image-models.generated.js.map +1 -1
  58. package/dist/images-models.d.ts.map +1 -1
  59. package/dist/images-models.js +4 -1
  60. package/dist/images-models.js.map +1 -1
  61. package/dist/index.d.ts +1 -0
  62. package/dist/index.d.ts.map +1 -1
  63. package/dist/index.js +1 -0
  64. package/dist/index.js.map +1 -1
  65. package/dist/legacy-api-aliases.d.ts +42 -0
  66. package/dist/legacy-api-aliases.d.ts.map +1 -0
  67. package/dist/legacy-api-aliases.js +49 -0
  68. package/dist/legacy-api-aliases.js.map +1 -0
  69. package/dist/models.d.ts.map +1 -1
  70. package/dist/models.generated.d.ts +754 -198
  71. package/dist/models.generated.d.ts.map +1 -1
  72. package/dist/models.js +4 -1
  73. package/dist/models.js.map +1 -1
  74. package/dist/providers/amazon-bedrock.models.d.ts +119 -0
  75. package/dist/providers/amazon-bedrock.models.d.ts.map +1 -1
  76. package/dist/providers/amazon-bedrock.models.js +127 -8
  77. package/dist/providers/amazon-bedrock.models.js.map +1 -1
  78. package/dist/providers/anthropic.models.d.ts +20 -34
  79. package/dist/providers/anthropic.models.d.ts.map +1 -1
  80. package/dist/providers/anthropic.models.js +18 -34
  81. package/dist/providers/anthropic.models.js.map +1 -1
  82. package/dist/providers/cloudflare-auth.d.ts.map +1 -1
  83. package/dist/providers/cloudflare-auth.js +18 -20
  84. package/dist/providers/cloudflare-auth.js.map +1 -1
  85. package/dist/providers/fireworks.models.d.ts +23 -0
  86. package/dist/providers/fireworks.models.d.ts.map +1 -1
  87. package/dist/providers/fireworks.models.js +19 -1
  88. package/dist/providers/fireworks.models.js.map +1 -1
  89. package/dist/providers/groq.models.d.ts.map +1 -1
  90. package/dist/providers/groq.models.js +1 -1
  91. package/dist/providers/groq.models.js.map +1 -1
  92. package/dist/providers/huggingface.models.d.ts +80 -0
  93. package/dist/providers/huggingface.models.d.ts.map +1 -1
  94. package/dist/providers/huggingface.models.js +72 -0
  95. package/dist/providers/huggingface.models.js.map +1 -1
  96. package/dist/providers/minimax-cn.models.d.ts.map +1 -1
  97. package/dist/providers/minimax-cn.models.js +4 -4
  98. package/dist/providers/minimax-cn.models.js.map +1 -1
  99. package/dist/providers/minimax.models.d.ts.map +1 -1
  100. package/dist/providers/minimax.models.js +4 -4
  101. package/dist/providers/minimax.models.js.map +1 -1
  102. package/dist/providers/nvidia.models.d.ts +28 -0
  103. package/dist/providers/nvidia.models.d.ts.map +1 -1
  104. package/dist/providers/nvidia.models.js +19 -0
  105. package/dist/providers/nvidia.models.js.map +1 -1
  106. package/dist/providers/opencode-go.models.d.ts.map +1 -1
  107. package/dist/providers/opencode-go.models.js +4 -4
  108. package/dist/providers/opencode-go.models.js.map +1 -1
  109. package/dist/providers/openrouter.models.d.ts +29 -32
  110. package/dist/providers/openrouter.models.d.ts.map +1 -1
  111. package/dist/providers/openrouter.models.js +100 -101
  112. package/dist/providers/openrouter.models.js.map +1 -1
  113. package/dist/providers/together.models.d.ts +31 -0
  114. package/dist/providers/together.models.d.ts.map +1 -1
  115. package/dist/providers/together.models.js +19 -0
  116. package/dist/providers/together.models.js.map +1 -1
  117. package/dist/providers/vercel-ai-gateway.models.d.ts +343 -51
  118. package/dist/providers/vercel-ai-gateway.models.d.ts.map +1 -1
  119. package/dist/providers/vercel-ai-gateway.models.js +367 -77
  120. package/dist/providers/vercel-ai-gateway.models.js.map +1 -1
  121. package/dist/providers/xiaomi-token-plan-ams.models.d.ts.map +1 -1
  122. package/dist/providers/xiaomi-token-plan-ams.models.js +12 -12
  123. package/dist/providers/xiaomi-token-plan-ams.models.js.map +1 -1
  124. package/dist/providers/xiaomi-token-plan-cn.models.d.ts.map +1 -1
  125. package/dist/providers/xiaomi-token-plan-cn.models.js +12 -12
  126. package/dist/providers/xiaomi-token-plan-cn.models.js.map +1 -1
  127. package/dist/providers/xiaomi-token-plan-sgp.models.d.ts.map +1 -1
  128. package/dist/providers/xiaomi-token-plan-sgp.models.js +12 -12
  129. package/dist/providers/xiaomi-token-plan-sgp.models.js.map +1 -1
  130. package/dist/providers/xiaomi.models.d.ts.map +1 -1
  131. package/dist/providers/xiaomi.models.js +15 -15
  132. package/dist/providers/xiaomi.models.js.map +1 -1
  133. package/dist/types.d.ts +6 -0
  134. package/dist/types.d.ts.map +1 -1
  135. package/dist/types.js.map +1 -1
  136. package/dist/utils/error-body.d.ts +25 -0
  137. package/dist/utils/error-body.d.ts.map +1 -0
  138. package/dist/utils/error-body.js +109 -0
  139. package/dist/utils/error-body.js.map +1 -0
  140. package/dist/utils/estimate.d.ts +17 -0
  141. package/dist/utils/estimate.d.ts.map +1 -0
  142. package/dist/utils/estimate.js +95 -0
  143. package/dist/utils/estimate.js.map +1 -0
  144. package/dist/utils/retry.d.ts +12 -0
  145. package/dist/utils/retry.d.ts.map +1 -0
  146. package/dist/utils/retry.js +86 -0
  147. package/dist/utils/retry.js.map +1 -0
  148. package/package.json +1 -1
@@ -109,7 +109,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
109
109
  api: "anthropic-messages",
110
110
  provider: "vercel-ai-gateway",
111
111
  baseUrl: "https://ai-gateway.vercel.sh",
112
- reasoning: true,
112
+ reasoning: false,
113
113
  input: ["text"],
114
114
  cost: {
115
115
  input: 1.5,
@@ -126,7 +126,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
126
126
  api: "anthropic-messages",
127
127
  provider: "vercel-ai-gateway",
128
128
  baseUrl: "https://ai-gateway.vercel.sh",
129
- reasoning: true,
129
+ reasoning: false,
130
130
  input: ["text"],
131
131
  cost: {
132
132
  input: 0.15,
@@ -143,7 +143,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
143
143
  api: "anthropic-messages",
144
144
  provider: "vercel-ai-gateway",
145
145
  baseUrl: "https://ai-gateway.vercel.sh",
146
- reasoning: true,
146
+ reasoning: false,
147
147
  input: ["text"],
148
148
  cost: {
149
149
  input: 0.5,
@@ -256,6 +256,40 @@ export const VERCEL_AI_GATEWAY_MODELS = {
256
256
  contextWindow: 131072,
257
257
  maxTokens: 32768,
258
258
  },
259
+ "alibaba/qwen3-vl-235b-a22b-instruct": {
260
+ id: "alibaba/qwen3-vl-235b-a22b-instruct",
261
+ name: "Qwen3 VL 235B A22B Instruct",
262
+ api: "anthropic-messages",
263
+ provider: "vercel-ai-gateway",
264
+ baseUrl: "https://ai-gateway.vercel.sh",
265
+ reasoning: false,
266
+ input: ["text", "image"],
267
+ cost: {
268
+ input: 0.4,
269
+ output: 1.6,
270
+ cacheRead: 0,
271
+ cacheWrite: 0,
272
+ },
273
+ contextWindow: 131072,
274
+ maxTokens: 129024,
275
+ },
276
+ "alibaba/qwen3-vl-instruct": {
277
+ id: "alibaba/qwen3-vl-instruct",
278
+ name: "Qwen3 VL 235B A22B Instruct",
279
+ api: "anthropic-messages",
280
+ provider: "vercel-ai-gateway",
281
+ baseUrl: "https://ai-gateway.vercel.sh",
282
+ reasoning: false,
283
+ input: ["text", "image"],
284
+ cost: {
285
+ input: 0.4,
286
+ output: 1.6,
287
+ cacheRead: 0,
288
+ cacheWrite: 0,
289
+ },
290
+ contextWindow: 131072,
291
+ maxTokens: 129024,
292
+ },
259
293
  "alibaba/qwen3-vl-thinking": {
260
294
  id: "alibaba/qwen3-vl-thinking",
261
295
  name: "Qwen3 VL 235B A22B Thinking",
@@ -375,6 +409,74 @@ export const VERCEL_AI_GATEWAY_MODELS = {
375
409
  contextWindow: 1000000,
376
410
  maxTokens: 64000,
377
411
  },
412
+ "amazon/nova-2-lite": {
413
+ id: "amazon/nova-2-lite",
414
+ name: "Nova 2 Lite",
415
+ api: "anthropic-messages",
416
+ provider: "vercel-ai-gateway",
417
+ baseUrl: "https://ai-gateway.vercel.sh",
418
+ reasoning: true,
419
+ input: ["text", "image"],
420
+ cost: {
421
+ input: 0.3,
422
+ output: 2.5,
423
+ cacheRead: 0.075,
424
+ cacheWrite: 0,
425
+ },
426
+ contextWindow: 1000000,
427
+ maxTokens: 1000000,
428
+ },
429
+ "amazon/nova-lite": {
430
+ id: "amazon/nova-lite",
431
+ name: "Nova Lite",
432
+ api: "anthropic-messages",
433
+ provider: "vercel-ai-gateway",
434
+ baseUrl: "https://ai-gateway.vercel.sh",
435
+ reasoning: false,
436
+ input: ["text", "image"],
437
+ cost: {
438
+ input: 0.06,
439
+ output: 0.24,
440
+ cacheRead: 0,
441
+ cacheWrite: 0,
442
+ },
443
+ contextWindow: 300000,
444
+ maxTokens: 8192,
445
+ },
446
+ "amazon/nova-micro": {
447
+ id: "amazon/nova-micro",
448
+ name: "Nova Micro",
449
+ api: "anthropic-messages",
450
+ provider: "vercel-ai-gateway",
451
+ baseUrl: "https://ai-gateway.vercel.sh",
452
+ reasoning: false,
453
+ input: ["text"],
454
+ cost: {
455
+ input: 0.035,
456
+ output: 0.14,
457
+ cacheRead: 0,
458
+ cacheWrite: 0,
459
+ },
460
+ contextWindow: 128000,
461
+ maxTokens: 8192,
462
+ },
463
+ "amazon/nova-pro": {
464
+ id: "amazon/nova-pro",
465
+ name: "Nova Pro",
466
+ api: "anthropic-messages",
467
+ provider: "vercel-ai-gateway",
468
+ baseUrl: "https://ai-gateway.vercel.sh",
469
+ reasoning: false,
470
+ input: ["text", "image"],
471
+ cost: {
472
+ input: 0.8,
473
+ output: 3.2,
474
+ cacheRead: 0,
475
+ cacheWrite: 0,
476
+ },
477
+ contextWindow: 300000,
478
+ maxTokens: 8192,
479
+ },
378
480
  "anthropic/claude-3-haiku": {
379
481
  id: "anthropic/claude-3-haiku",
380
482
  name: "Claude 3 Haiku",
@@ -586,6 +688,24 @@ export const VERCEL_AI_GATEWAY_MODELS = {
586
688
  contextWindow: 1000000,
587
689
  maxTokens: 128000,
588
690
  },
691
+ "anthropic/claude-sonnet-5": {
692
+ id: "anthropic/claude-sonnet-5",
693
+ name: "Claude Sonnet 5",
694
+ api: "anthropic-messages",
695
+ provider: "vercel-ai-gateway",
696
+ baseUrl: "https://ai-gateway.vercel.sh",
697
+ compat: { "forceAdaptiveThinking": true },
698
+ reasoning: true,
699
+ input: ["text", "image"],
700
+ cost: {
701
+ input: 2,
702
+ output: 10,
703
+ cacheRead: 0.2,
704
+ cacheWrite: 2.5,
705
+ },
706
+ contextWindow: 1000000,
707
+ maxTokens: 128000,
708
+ },
589
709
  "arcee-ai/trinity-large-preview": {
590
710
  id: "arcee-ai/trinity-large-preview",
591
711
  name: "Trinity Large Preview",
@@ -620,6 +740,23 @@ export const VERCEL_AI_GATEWAY_MODELS = {
620
740
  contextWindow: 262100,
621
741
  maxTokens: 80000,
622
742
  },
743
+ "arcee-ai/trinity-mini": {
744
+ id: "arcee-ai/trinity-mini",
745
+ name: "Trinity Mini",
746
+ api: "anthropic-messages",
747
+ provider: "vercel-ai-gateway",
748
+ baseUrl: "https://ai-gateway.vercel.sh",
749
+ reasoning: false,
750
+ input: ["text"],
751
+ cost: {
752
+ input: 0.045,
753
+ output: 0.15,
754
+ cacheRead: 0,
755
+ cacheWrite: 0,
756
+ },
757
+ contextWindow: 131072,
758
+ maxTokens: 131072,
759
+ },
623
760
  "bytedance/seed-1.6": {
624
761
  id: "bytedance/seed-1.6",
625
762
  name: "Seed 1.6",
@@ -627,7 +764,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
627
764
  provider: "vercel-ai-gateway",
628
765
  baseUrl: "https://ai-gateway.vercel.sh",
629
766
  reasoning: true,
630
- input: ["text"],
767
+ input: ["text", "image"],
631
768
  cost: {
632
769
  input: 0.25,
633
770
  output: 2,
@@ -637,6 +774,23 @@ export const VERCEL_AI_GATEWAY_MODELS = {
637
774
  contextWindow: 256000,
638
775
  maxTokens: 32000,
639
776
  },
777
+ "bytedance/seed-1.8": {
778
+ id: "bytedance/seed-1.8",
779
+ name: "Bytedance Seed 1.8",
780
+ api: "anthropic-messages",
781
+ provider: "vercel-ai-gateway",
782
+ baseUrl: "https://ai-gateway.vercel.sh",
783
+ reasoning: true,
784
+ input: ["text", "image"],
785
+ cost: {
786
+ input: 0.25,
787
+ output: 2,
788
+ cacheRead: 0.05,
789
+ cacheWrite: 0,
790
+ },
791
+ contextWindow: 256000,
792
+ maxTokens: 64000,
793
+ },
640
794
  "cohere/command-a": {
641
795
  id: "cohere/command-a",
642
796
  name: "Command A",
@@ -697,13 +851,13 @@ export const VERCEL_AI_GATEWAY_MODELS = {
697
851
  reasoning: true,
698
852
  input: ["text"],
699
853
  cost: {
700
- input: 0.56,
701
- output: 1.68,
702
- cacheRead: 0.28,
854
+ input: 0.6,
855
+ output: 1.7,
856
+ cacheRead: 0,
703
857
  cacheWrite: 0,
704
858
  },
705
- contextWindow: 163840,
706
- maxTokens: 8192,
859
+ contextWindow: 128000,
860
+ maxTokens: 128000,
707
861
  },
708
862
  "deepseek/deepseek-v3.1-terminus": {
709
863
  id: "deepseek/deepseek-v3.1-terminus",
@@ -728,8 +882,8 @@ export const VERCEL_AI_GATEWAY_MODELS = {
728
882
  api: "anthropic-messages",
729
883
  provider: "vercel-ai-gateway",
730
884
  baseUrl: "https://ai-gateway.vercel.sh",
731
- reasoning: true,
732
- input: ["text", "image"],
885
+ reasoning: false,
886
+ input: ["text"],
733
887
  cost: {
734
888
  input: 0.28,
735
889
  output: 0.42,
@@ -746,7 +900,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
746
900
  provider: "vercel-ai-gateway",
747
901
  baseUrl: "https://ai-gateway.vercel.sh",
748
902
  reasoning: true,
749
- input: ["text", "image"],
903
+ input: ["text"],
750
904
  cost: {
751
905
  input: 0.62,
752
906
  output: 1.85,
@@ -763,7 +917,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
763
917
  provider: "vercel-ai-gateway",
764
918
  baseUrl: "https://ai-gateway.vercel.sh",
765
919
  reasoning: true,
766
- input: ["text", "image"],
920
+ input: ["text"],
767
921
  cost: {
768
922
  input: 0.14,
769
923
  output: 0.28,
@@ -966,7 +1120,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
966
1120
  api: "anthropic-messages",
967
1121
  provider: "vercel-ai-gateway",
968
1122
  baseUrl: "https://ai-gateway.vercel.sh",
969
- reasoning: false,
1123
+ reasoning: true,
970
1124
  input: ["text", "image"],
971
1125
  cost: {
972
1126
  input: 0.14,
@@ -1011,6 +1165,40 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1011
1165
  contextWindow: 32000,
1012
1166
  maxTokens: 16384,
1013
1167
  },
1168
+ "interfaze/interfaze-beta": {
1169
+ id: "interfaze/interfaze-beta",
1170
+ name: "Interfaze Beta",
1171
+ api: "anthropic-messages",
1172
+ provider: "vercel-ai-gateway",
1173
+ baseUrl: "https://ai-gateway.vercel.sh",
1174
+ reasoning: true,
1175
+ input: ["text", "image"],
1176
+ cost: {
1177
+ input: 1.5,
1178
+ output: 3.5,
1179
+ cacheRead: 0,
1180
+ cacheWrite: 0,
1181
+ },
1182
+ contextWindow: 1000000,
1183
+ maxTokens: 32000,
1184
+ },
1185
+ "kwaipilot/kat-coder-pro-v1": {
1186
+ id: "kwaipilot/kat-coder-pro-v1",
1187
+ name: "KAT-Coder-Pro V1",
1188
+ api: "anthropic-messages",
1189
+ provider: "vercel-ai-gateway",
1190
+ baseUrl: "https://ai-gateway.vercel.sh",
1191
+ reasoning: false,
1192
+ input: ["text"],
1193
+ cost: {
1194
+ input: 0.3,
1195
+ output: 1.2,
1196
+ cacheRead: 0.06,
1197
+ cacheWrite: 0,
1198
+ },
1199
+ contextWindow: 256000,
1200
+ maxTokens: 32000,
1201
+ },
1014
1202
  "kwaipilot/kat-coder-pro-v2": {
1015
1203
  id: "kwaipilot/kat-coder-pro-v2",
1016
1204
  name: "Kat Coder Pro V2",
@@ -1045,6 +1233,23 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1045
1233
  contextWindow: 128000,
1046
1234
  maxTokens: 100000,
1047
1235
  },
1236
+ "meituan/longcat-flash-thinking-2601": {
1237
+ id: "meituan/longcat-flash-thinking-2601",
1238
+ name: "LongCat Flash Thinking 2601",
1239
+ api: "anthropic-messages",
1240
+ provider: "vercel-ai-gateway",
1241
+ baseUrl: "https://ai-gateway.vercel.sh",
1242
+ reasoning: true,
1243
+ input: ["text"],
1244
+ cost: {
1245
+ input: 0,
1246
+ output: 0,
1247
+ cacheRead: 0,
1248
+ cacheWrite: 0,
1249
+ },
1250
+ contextWindow: 32768,
1251
+ maxTokens: 32768,
1252
+ },
1048
1253
  "meta/llama-3.1-70b": {
1049
1254
  id: "meta/llama-3.1-70b",
1050
1255
  name: "Llama 3.1 70B Instruct",
@@ -1358,7 +1563,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1358
1563
  provider: "vercel-ai-gateway",
1359
1564
  baseUrl: "https://ai-gateway.vercel.sh",
1360
1565
  reasoning: false,
1361
- input: ["text"],
1566
+ input: ["text", "image"],
1362
1567
  cost: {
1363
1568
  input: 0.1,
1364
1569
  output: 0.3,
@@ -1368,6 +1573,57 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1368
1573
  contextWindow: 256000,
1369
1574
  maxTokens: 256000,
1370
1575
  },
1576
+ "mistral/magistral-medium": {
1577
+ id: "mistral/magistral-medium",
1578
+ name: "Magistral Medium 2509",
1579
+ api: "anthropic-messages",
1580
+ provider: "vercel-ai-gateway",
1581
+ baseUrl: "https://ai-gateway.vercel.sh",
1582
+ reasoning: true,
1583
+ input: ["text", "image"],
1584
+ cost: {
1585
+ input: 2,
1586
+ output: 5,
1587
+ cacheRead: 0,
1588
+ cacheWrite: 0,
1589
+ },
1590
+ contextWindow: 128000,
1591
+ maxTokens: 64000,
1592
+ },
1593
+ "mistral/magistral-small": {
1594
+ id: "mistral/magistral-small",
1595
+ name: "Magistral Small 2509",
1596
+ api: "anthropic-messages",
1597
+ provider: "vercel-ai-gateway",
1598
+ baseUrl: "https://ai-gateway.vercel.sh",
1599
+ reasoning: true,
1600
+ input: ["text", "image"],
1601
+ cost: {
1602
+ input: 0.5,
1603
+ output: 1.5,
1604
+ cacheRead: 0,
1605
+ cacheWrite: 0,
1606
+ },
1607
+ contextWindow: 128000,
1608
+ maxTokens: 64000,
1609
+ },
1610
+ "mistral/ministral-14b": {
1611
+ id: "mistral/ministral-14b",
1612
+ name: "Ministral 14B",
1613
+ api: "anthropic-messages",
1614
+ provider: "vercel-ai-gateway",
1615
+ baseUrl: "https://ai-gateway.vercel.sh",
1616
+ reasoning: false,
1617
+ input: ["text", "image"],
1618
+ cost: {
1619
+ input: 0.2,
1620
+ output: 0.2,
1621
+ cacheRead: 0,
1622
+ cacheWrite: 0,
1623
+ },
1624
+ contextWindow: 256000,
1625
+ maxTokens: 256000,
1626
+ },
1371
1627
  "mistral/ministral-3b": {
1372
1628
  id: "mistral/ministral-3b",
1373
1629
  name: "Ministral 3B",
@@ -1402,6 +1658,23 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1402
1658
  contextWindow: 128000,
1403
1659
  maxTokens: 4000,
1404
1660
  },
1661
+ "mistral/mistral-large-3": {
1662
+ id: "mistral/mistral-large-3",
1663
+ name: "Mistral Large 3",
1664
+ api: "anthropic-messages",
1665
+ provider: "vercel-ai-gateway",
1666
+ baseUrl: "https://ai-gateway.vercel.sh",
1667
+ reasoning: false,
1668
+ input: ["text", "image"],
1669
+ cost: {
1670
+ input: 0.5,
1671
+ output: 1.5,
1672
+ cacheRead: 0,
1673
+ cacheWrite: 0,
1674
+ },
1675
+ contextWindow: 256000,
1676
+ maxTokens: 256000,
1677
+ },
1405
1678
  "mistral/mistral-medium": {
1406
1679
  id: "mistral/mistral-medium",
1407
1680
  name: "Mistral Medium 3.1",
@@ -1426,7 +1699,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1426
1699
  provider: "vercel-ai-gateway",
1427
1700
  baseUrl: "https://ai-gateway.vercel.sh",
1428
1701
  reasoning: true,
1429
- input: ["text"],
1702
+ input: ["text", "image"],
1430
1703
  cost: {
1431
1704
  input: 1.5,
1432
1705
  output: 7.5,
@@ -1530,13 +1803,13 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1530
1803
  reasoning: true,
1531
1804
  input: ["text"],
1532
1805
  cost: {
1533
- input: 0.6,
1534
- output: 2.5,
1535
- cacheRead: 0.15,
1806
+ input: 0.47,
1807
+ output: 2,
1808
+ cacheRead: 0.141,
1536
1809
  cacheWrite: 0,
1537
1810
  },
1538
- contextWindow: 262114,
1539
- maxTokens: 262114,
1811
+ contextWindow: 216144,
1812
+ maxTokens: 216144,
1540
1813
  },
1541
1814
  "moonshotai/kimi-k2.5": {
1542
1815
  id: "moonshotai/kimi-k2.5",
@@ -1606,6 +1879,23 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1606
1879
  contextWindow: 262144,
1607
1880
  maxTokens: 32768,
1608
1881
  },
1882
+ "nvidia/nemotron-3-nano-30b-a3b": {
1883
+ id: "nvidia/nemotron-3-nano-30b-a3b",
1884
+ name: "Nemotron 3 Nano 30B A3B",
1885
+ api: "anthropic-messages",
1886
+ provider: "vercel-ai-gateway",
1887
+ baseUrl: "https://ai-gateway.vercel.sh",
1888
+ reasoning: true,
1889
+ input: ["text"],
1890
+ cost: {
1891
+ input: 0.05,
1892
+ output: 0.24,
1893
+ cacheRead: 0,
1894
+ cacheWrite: 0,
1895
+ },
1896
+ contextWindow: 262144,
1897
+ maxTokens: 262144,
1898
+ },
1609
1899
  "nvidia/nemotron-3-super-120b-a12b": {
1610
1900
  id: "nvidia/nemotron-3-super-120b-a12b",
1611
1901
  name: "NVIDIA Nemotron 3 Super 120B A12B",
@@ -1674,6 +1964,23 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1674
1964
  contextWindow: 131072,
1675
1965
  maxTokens: 131072,
1676
1966
  },
1967
+ "openai/gpt-3.5-turbo": {
1968
+ id: "openai/gpt-3.5-turbo",
1969
+ name: "GPT-3.5 Turbo",
1970
+ api: "anthropic-messages",
1971
+ provider: "vercel-ai-gateway",
1972
+ baseUrl: "https://ai-gateway.vercel.sh",
1973
+ reasoning: false,
1974
+ input: ["text"],
1975
+ cost: {
1976
+ input: 0.5,
1977
+ output: 1.5,
1978
+ cacheRead: 0,
1979
+ cacheWrite: 0,
1980
+ },
1981
+ contextWindow: 16385,
1982
+ maxTokens: 4096,
1983
+ },
1677
1984
  "openai/gpt-4-turbo": {
1678
1985
  id: "openai/gpt-4-turbo",
1679
1986
  name: "GPT-4 Turbo",
@@ -1799,7 +2106,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1799
2106
  api: "anthropic-messages",
1800
2107
  provider: "vercel-ai-gateway",
1801
2108
  baseUrl: "https://ai-gateway.vercel.sh",
1802
- reasoning: true,
2109
+ reasoning: false,
1803
2110
  input: ["text", "image"],
1804
2111
  cost: {
1805
2112
  input: 1.25,
@@ -1935,7 +2242,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1935
2242
  api: "anthropic-messages",
1936
2243
  provider: "vercel-ai-gateway",
1937
2244
  baseUrl: "https://ai-gateway.vercel.sh",
1938
- reasoning: true,
2245
+ reasoning: false,
1939
2246
  input: ["text", "image"],
1940
2247
  cost: {
1941
2248
  input: 1.25,
@@ -1987,7 +2294,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
1987
2294
  api: "anthropic-messages",
1988
2295
  provider: "vercel-ai-gateway",
1989
2296
  baseUrl: "https://ai-gateway.vercel.sh",
1990
- reasoning: true,
2297
+ reasoning: false,
1991
2298
  thinkingLevelMap: { "xhigh": "xhigh" },
1992
2299
  input: ["text", "image"],
1993
2300
  cost: {
@@ -2041,7 +2348,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2041
2348
  api: "anthropic-messages",
2042
2349
  provider: "vercel-ai-gateway",
2043
2350
  baseUrl: "https://ai-gateway.vercel.sh",
2044
- reasoning: true,
2351
+ reasoning: false,
2045
2352
  thinkingLevelMap: { "xhigh": "xhigh" },
2046
2353
  input: ["text", "image"],
2047
2354
  cost: {
@@ -2188,13 +2495,13 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2188
2495
  reasoning: true,
2189
2496
  input: ["text"],
2190
2497
  cost: {
2191
- input: 0.35,
2192
- output: 0.75,
2193
- cacheRead: 0.25,
2498
+ input: 0.1,
2499
+ output: 0.5,
2500
+ cacheRead: 0,
2194
2501
  cacheWrite: 0,
2195
2502
  },
2196
2503
  contextWindow: 131072,
2197
- maxTokens: 131000,
2504
+ maxTokens: 131072,
2198
2505
  },
2199
2506
  "openai/gpt-oss-20b": {
2200
2507
  id: "openai/gpt-oss-20b",
@@ -2332,40 +2639,6 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2332
2639
  contextWindow: 200000,
2333
2640
  maxTokens: 100000,
2334
2641
  },
2335
- "perplexity/sonar": {
2336
- id: "perplexity/sonar",
2337
- name: "Sonar",
2338
- api: "anthropic-messages",
2339
- provider: "vercel-ai-gateway",
2340
- baseUrl: "https://ai-gateway.vercel.sh",
2341
- reasoning: false,
2342
- input: ["text", "image"],
2343
- cost: {
2344
- input: 0,
2345
- output: 0,
2346
- cacheRead: 0,
2347
- cacheWrite: 0,
2348
- },
2349
- contextWindow: 127000,
2350
- maxTokens: 8000,
2351
- },
2352
- "perplexity/sonar-pro": {
2353
- id: "perplexity/sonar-pro",
2354
- name: "Sonar Pro",
2355
- api: "anthropic-messages",
2356
- provider: "vercel-ai-gateway",
2357
- baseUrl: "https://ai-gateway.vercel.sh",
2358
- reasoning: false,
2359
- input: ["text", "image"],
2360
- cost: {
2361
- input: 0,
2362
- output: 0,
2363
- cacheRead: 0,
2364
- cacheWrite: 0,
2365
- },
2366
- contextWindow: 200000,
2367
- maxTokens: 8000,
2368
- },
2369
2642
  "sakana/fugu-ultra": {
2370
2643
  id: "sakana/fugu-ultra",
2371
2644
  name: "Fugu Ultra",
@@ -2645,7 +2918,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2645
2918
  provider: "vercel-ai-gateway",
2646
2919
  baseUrl: "https://ai-gateway.vercel.sh",
2647
2920
  reasoning: true,
2648
- input: ["text", "image"],
2921
+ input: ["text"],
2649
2922
  cost: {
2650
2923
  input: 0.435,
2651
2924
  output: 0.87,
@@ -2657,7 +2930,7 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2657
2930
  },
2658
2931
  "zai/glm-4.5": {
2659
2932
  id: "zai/glm-4.5",
2660
- name: "GLM-4.5",
2933
+ name: "GLM 4.5",
2661
2934
  api: "anthropic-messages",
2662
2935
  provider: "vercel-ai-gateway",
2663
2936
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -2766,13 +3039,13 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2766
3039
  reasoning: true,
2767
3040
  input: ["text"],
2768
3041
  cost: {
2769
- input: 2.25,
2770
- output: 2.75,
2771
- cacheRead: 2.25,
3042
+ input: 0.6,
3043
+ output: 2.2,
3044
+ cacheRead: 0.12,
2772
3045
  cacheWrite: 0,
2773
3046
  },
2774
- contextWindow: 131000,
2775
- maxTokens: 40000,
3047
+ contextWindow: 200000,
3048
+ maxTokens: 120000,
2776
3049
  },
2777
3050
  "zai/glm-4.7-flash": {
2778
3051
  id: "zai/glm-4.7-flash",
@@ -2817,8 +3090,8 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2817
3090
  reasoning: true,
2818
3091
  input: ["text"],
2819
3092
  cost: {
2820
- input: 1,
2821
- output: 3.2,
3093
+ input: 0.95,
3094
+ output: 3.15,
2822
3095
  cacheRead: 0.2,
2823
3096
  cacheWrite: 0,
2824
3097
  },
@@ -2849,15 +3122,15 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2849
3122
  provider: "vercel-ai-gateway",
2850
3123
  baseUrl: "https://ai-gateway.vercel.sh",
2851
3124
  reasoning: true,
2852
- input: ["text", "image"],
3125
+ input: ["text"],
2853
3126
  cost: {
2854
- input: 1.4,
2855
- output: 4.4,
3127
+ input: 1.3,
3128
+ output: 4.3,
2856
3129
  cacheRead: 0.26,
2857
3130
  cacheWrite: 0,
2858
3131
  },
2859
- contextWindow: 202800,
2860
- maxTokens: 64000,
3132
+ contextWindow: 202000,
3133
+ maxTokens: 202000,
2861
3134
  },
2862
3135
  "zai/glm-5.2": {
2863
3136
  id: "zai/glm-5.2",
@@ -2876,6 +3149,23 @@ export const VERCEL_AI_GATEWAY_MODELS = {
2876
3149
  contextWindow: 1000000,
2877
3150
  maxTokens: 128000,
2878
3151
  },
3152
+ "zai/glm-5.2-fast": {
3153
+ id: "zai/glm-5.2-fast",
3154
+ name: "GLM 5.2 Fast",
3155
+ api: "anthropic-messages",
3156
+ provider: "vercel-ai-gateway",
3157
+ baseUrl: "https://ai-gateway.vercel.sh",
3158
+ reasoning: true,
3159
+ input: ["text"],
3160
+ cost: {
3161
+ input: 3,
3162
+ output: 10.25,
3163
+ cacheRead: 0.5,
3164
+ cacheWrite: 0,
3165
+ },
3166
+ contextWindow: 1000000,
3167
+ maxTokens: 128000,
3168
+ },
2879
3169
  "zai/glm-5v-turbo": {
2880
3170
  id: "zai/glm-5v-turbo",
2881
3171
  name: "GLM 5V Turbo",