llama-stack-client 0.4.4 → 0.4.5

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 (291) hide show
  1. package/.github/workflows/ci.yml +4 -4
  2. package/.github/workflows/release-doctor.yml +1 -1
  3. package/.release-please-manifest.json +1 -1
  4. package/.stats.yml +4 -4
  5. package/CHANGELOG.md +41 -0
  6. package/LICENSE +1 -1
  7. package/README.md +41 -27
  8. package/api.md +32 -25
  9. package/dist/CHANGELOG.md +41 -0
  10. package/dist/LICENSE +1 -1
  11. package/dist/README.md +41 -27
  12. package/dist/core.d.ts.map +1 -1
  13. package/dist/core.js +5 -0
  14. package/dist/core.js.map +1 -1
  15. package/dist/core.mjs +5 -0
  16. package/dist/core.mjs.map +1 -1
  17. package/dist/index.d.mts +12 -6
  18. package/dist/index.d.ts +12 -6
  19. package/dist/index.d.ts.map +1 -1
  20. package/dist/index.js.map +1 -1
  21. package/dist/index.mjs.map +1 -1
  22. package/dist/package.json +1 -1
  23. package/dist/resources/alpha/admin.d.ts +41 -0
  24. package/dist/resources/alpha/admin.d.ts.map +1 -0
  25. package/dist/resources/alpha/admin.js +46 -0
  26. package/dist/resources/alpha/admin.js.map +1 -0
  27. package/dist/resources/alpha/admin.mjs +42 -0
  28. package/dist/resources/alpha/admin.mjs.map +1 -0
  29. package/dist/resources/alpha/alpha.d.ts +8 -4
  30. package/dist/resources/alpha/alpha.d.ts.map +1 -1
  31. package/dist/resources/alpha/alpha.js +6 -2
  32. package/dist/resources/alpha/alpha.js.map +1 -1
  33. package/dist/resources/alpha/alpha.mjs +6 -2
  34. package/dist/resources/alpha/alpha.mjs.map +1 -1
  35. package/dist/resources/alpha/benchmarks.d.ts +37 -1
  36. package/dist/resources/alpha/benchmarks.d.ts.map +1 -1
  37. package/dist/resources/alpha/eval/eval.d.ts +28 -7
  38. package/dist/resources/alpha/eval/eval.d.ts.map +1 -1
  39. package/dist/resources/alpha/eval/eval.js.map +1 -1
  40. package/dist/resources/alpha/eval/eval.mjs.map +1 -1
  41. package/dist/resources/alpha/index.d.ts +2 -1
  42. package/dist/resources/alpha/index.d.ts.map +1 -1
  43. package/dist/resources/alpha/index.js +3 -1
  44. package/dist/resources/alpha/index.js.map +1 -1
  45. package/dist/resources/alpha/index.mjs +1 -0
  46. package/dist/resources/alpha/index.mjs.map +1 -1
  47. package/dist/resources/alpha/inference.d.ts +58 -7
  48. package/dist/resources/alpha/inference.d.ts.map +1 -1
  49. package/dist/resources/alpha/post-training/index.d.ts +2 -2
  50. package/dist/resources/alpha/post-training/index.d.ts.map +1 -1
  51. package/dist/resources/alpha/post-training/index.js.map +1 -1
  52. package/dist/resources/alpha/post-training/index.mjs +1 -1
  53. package/dist/resources/alpha/post-training/index.mjs.map +1 -1
  54. package/dist/resources/alpha/post-training/job.d.ts +4 -16
  55. package/dist/resources/alpha/post-training/job.d.ts.map +1 -1
  56. package/dist/resources/alpha/post-training/job.js +5 -6
  57. package/dist/resources/alpha/post-training/job.js.map +1 -1
  58. package/dist/resources/alpha/post-training/job.mjs +5 -6
  59. package/dist/resources/alpha/post-training/job.mjs.map +1 -1
  60. package/dist/resources/alpha/post-training/post-training.d.ts +38 -11
  61. package/dist/resources/alpha/post-training/post-training.d.ts.map +1 -1
  62. package/dist/resources/alpha/post-training/post-training.js.map +1 -1
  63. package/dist/resources/alpha/post-training/post-training.mjs +1 -1
  64. package/dist/resources/alpha/post-training/post-training.mjs.map +1 -1
  65. package/dist/resources/beta/datasets.d.ts +102 -8
  66. package/dist/resources/beta/datasets.d.ts.map +1 -1
  67. package/dist/resources/chat/chat.d.ts +119 -17
  68. package/dist/resources/chat/chat.d.ts.map +1 -1
  69. package/dist/resources/chat/chat.js.map +1 -1
  70. package/dist/resources/chat/chat.mjs.map +1 -1
  71. package/dist/resources/chat/completions.d.ts +1130 -73
  72. package/dist/resources/chat/completions.d.ts.map +1 -1
  73. package/dist/resources/chat/completions.js +0 -2
  74. package/dist/resources/chat/completions.js.map +1 -1
  75. package/dist/resources/chat/completions.mjs +0 -2
  76. package/dist/resources/chat/completions.mjs.map +1 -1
  77. package/dist/resources/completions.d.ts +135 -26
  78. package/dist/resources/completions.d.ts.map +1 -1
  79. package/dist/resources/completions.js.map +1 -1
  80. package/dist/resources/completions.mjs.map +1 -1
  81. package/dist/resources/conversations/conversations.d.ts +56 -12
  82. package/dist/resources/conversations/conversations.d.ts.map +1 -1
  83. package/dist/resources/conversations/conversations.js +0 -8
  84. package/dist/resources/conversations/conversations.js.map +1 -1
  85. package/dist/resources/conversations/conversations.mjs +0 -8
  86. package/dist/resources/conversations/conversations.mjs.map +1 -1
  87. package/dist/resources/conversations/items.d.ts +364 -81
  88. package/dist/resources/conversations/items.d.ts.map +1 -1
  89. package/dist/resources/conversations/items.js +0 -6
  90. package/dist/resources/conversations/items.js.map +1 -1
  91. package/dist/resources/conversations/items.mjs +0 -6
  92. package/dist/resources/conversations/items.mjs.map +1 -1
  93. package/dist/resources/embeddings.d.ts +46 -8
  94. package/dist/resources/embeddings.d.ts.map +1 -1
  95. package/dist/resources/embeddings.js +0 -2
  96. package/dist/resources/embeddings.js.map +1 -1
  97. package/dist/resources/embeddings.mjs +0 -2
  98. package/dist/resources/embeddings.mjs.map +1 -1
  99. package/dist/resources/files.d.ts +58 -31
  100. package/dist/resources/files.d.ts.map +1 -1
  101. package/dist/resources/files.js +4 -16
  102. package/dist/resources/files.js.map +1 -1
  103. package/dist/resources/files.mjs +4 -16
  104. package/dist/resources/files.mjs.map +1 -1
  105. package/dist/resources/index.d.ts +3 -3
  106. package/dist/resources/index.d.ts.map +1 -1
  107. package/dist/resources/index.js.map +1 -1
  108. package/dist/resources/index.mjs.map +1 -1
  109. package/dist/resources/inspect.d.ts +3 -45
  110. package/dist/resources/inspect.d.ts.map +1 -1
  111. package/dist/resources/inspect.js +0 -4
  112. package/dist/resources/inspect.js.map +1 -1
  113. package/dist/resources/inspect.mjs +0 -4
  114. package/dist/resources/inspect.mjs.map +1 -1
  115. package/dist/resources/models/models.d.ts +21 -6
  116. package/dist/resources/models/models.d.ts.map +1 -1
  117. package/dist/resources/models/models.js +0 -6
  118. package/dist/resources/models/models.js.map +1 -1
  119. package/dist/resources/models/models.mjs +0 -6
  120. package/dist/resources/models/models.mjs.map +1 -1
  121. package/dist/resources/moderations.d.ts +38 -5
  122. package/dist/resources/moderations.d.ts.map +1 -1
  123. package/dist/resources/moderations.js +1 -3
  124. package/dist/resources/moderations.js.map +1 -1
  125. package/dist/resources/moderations.mjs +1 -3
  126. package/dist/resources/moderations.mjs.map +1 -1
  127. package/dist/resources/prompts/prompts.d.ts +24 -10
  128. package/dist/resources/prompts/prompts.d.ts.map +1 -1
  129. package/dist/resources/prompts/prompts.js +2 -10
  130. package/dist/resources/prompts/prompts.js.map +1 -1
  131. package/dist/resources/prompts/prompts.mjs +2 -10
  132. package/dist/resources/prompts/prompts.mjs.map +1 -1
  133. package/dist/resources/prompts/versions.d.ts +0 -2
  134. package/dist/resources/prompts/versions.d.ts.map +1 -1
  135. package/dist/resources/prompts/versions.js +0 -2
  136. package/dist/resources/prompts/versions.js.map +1 -1
  137. package/dist/resources/prompts/versions.mjs +0 -2
  138. package/dist/resources/prompts/versions.mjs.map +1 -1
  139. package/dist/resources/providers.d.ts +5 -12
  140. package/dist/resources/providers.d.ts.map +1 -1
  141. package/dist/resources/providers.js +0 -4
  142. package/dist/resources/providers.js.map +1 -1
  143. package/dist/resources/providers.mjs +0 -4
  144. package/dist/resources/providers.mjs.map +1 -1
  145. package/dist/resources/responses/input-items.d.ts +108 -9
  146. package/dist/resources/responses/input-items.d.ts.map +1 -1
  147. package/dist/resources/responses/responses.d.ts +881 -124
  148. package/dist/resources/responses/responses.d.ts.map +1 -1
  149. package/dist/resources/responses/responses.js.map +1 -1
  150. package/dist/resources/responses/responses.mjs.map +1 -1
  151. package/dist/resources/routes.d.ts +10 -9
  152. package/dist/resources/routes.d.ts.map +1 -1
  153. package/dist/resources/routes.js.map +1 -1
  154. package/dist/resources/routes.mjs.map +1 -1
  155. package/dist/resources/safety.d.ts +131 -9
  156. package/dist/resources/safety.d.ts.map +1 -1
  157. package/dist/resources/safety.js +1 -3
  158. package/dist/resources/safety.js.map +1 -1
  159. package/dist/resources/safety.mjs +1 -3
  160. package/dist/resources/safety.mjs.map +1 -1
  161. package/dist/resources/scoring-functions.d.ts +23 -1
  162. package/dist/resources/scoring-functions.d.ts.map +1 -1
  163. package/dist/resources/scoring.d.ts +24 -0
  164. package/dist/resources/scoring.d.ts.map +1 -1
  165. package/dist/resources/shared.d.ts +139 -7
  166. package/dist/resources/shared.d.ts.map +1 -1
  167. package/dist/resources/shields.d.ts +21 -0
  168. package/dist/resources/shields.d.ts.map +1 -1
  169. package/dist/resources/vector-io.d.ts +75 -58
  170. package/dist/resources/vector-io.d.ts.map +1 -1
  171. package/dist/resources/vector-io.js +1 -1
  172. package/dist/resources/vector-io.mjs +1 -1
  173. package/dist/resources/vector-stores/file-batches.d.ts +13 -7
  174. package/dist/resources/vector-stores/file-batches.d.ts.map +1 -1
  175. package/dist/resources/vector-stores/file-batches.js +3 -6
  176. package/dist/resources/vector-stores/file-batches.js.map +1 -1
  177. package/dist/resources/vector-stores/file-batches.mjs +3 -6
  178. package/dist/resources/vector-stores/file-batches.mjs.map +1 -1
  179. package/dist/resources/vector-stores/files.d.ts +31 -9
  180. package/dist/resources/vector-stores/files.d.ts.map +1 -1
  181. package/dist/resources/vector-stores/files.js +4 -4
  182. package/dist/resources/vector-stores/files.js.map +1 -1
  183. package/dist/resources/vector-stores/files.mjs +4 -4
  184. package/dist/resources/vector-stores/files.mjs.map +1 -1
  185. package/dist/resources/vector-stores/vector-stores.d.ts +87 -13
  186. package/dist/resources/vector-stores/vector-stores.d.ts.map +1 -1
  187. package/dist/resources/vector-stores/vector-stores.js +5 -10
  188. package/dist/resources/vector-stores/vector-stores.js.map +1 -1
  189. package/dist/resources/vector-stores/vector-stores.mjs +5 -10
  190. package/dist/resources/vector-stores/vector-stores.mjs.map +1 -1
  191. package/dist/src/core.ts +6 -0
  192. package/dist/src/index.ts +11 -16
  193. package/dist/src/resources/alpha/admin.ts +84 -0
  194. package/dist/src/resources/alpha/alpha.ts +16 -8
  195. package/dist/src/resources/alpha/benchmarks.ts +37 -1
  196. package/dist/src/resources/alpha/eval/eval.ts +28 -7
  197. package/dist/src/resources/alpha/index.ts +2 -0
  198. package/dist/src/resources/alpha/inference.ts +58 -7
  199. package/dist/src/resources/alpha/post-training/index.ts +2 -10
  200. package/dist/src/resources/alpha/post-training/job.ts +5 -26
  201. package/dist/src/resources/alpha/post-training/post-training.ts +38 -22
  202. package/dist/src/resources/beta/datasets.ts +103 -8
  203. package/dist/src/resources/chat/chat.ts +122 -20
  204. package/dist/src/resources/chat/completions.ts +1138 -79
  205. package/dist/src/resources/completions.ts +137 -28
  206. package/dist/src/resources/conversations/conversations.ts +67 -15
  207. package/dist/src/resources/conversations/items.ts +479 -110
  208. package/dist/src/resources/embeddings.ts +46 -8
  209. package/dist/src/resources/files.ts +58 -31
  210. package/dist/src/resources/index.ts +3 -3
  211. package/dist/src/resources/inspect.ts +3 -57
  212. package/dist/src/resources/models/models.ts +21 -6
  213. package/dist/src/resources/moderations.ts +38 -5
  214. package/dist/src/resources/prompts/prompts.ts +26 -12
  215. package/dist/src/resources/prompts/versions.ts +0 -2
  216. package/dist/src/resources/providers.ts +5 -16
  217. package/dist/src/resources/responses/input-items.ts +138 -15
  218. package/dist/src/resources/responses/responses.ts +1076 -196
  219. package/dist/src/resources/routes.ts +10 -14
  220. package/dist/src/resources/safety.ts +131 -8
  221. package/dist/src/resources/scoring-functions.ts +23 -1
  222. package/dist/src/resources/scoring.ts +24 -0
  223. package/dist/src/resources/shared.ts +148 -7
  224. package/dist/src/resources/shields.ts +21 -0
  225. package/dist/src/resources/vector-io.ts +95 -78
  226. package/dist/src/resources/vector-stores/file-batches.ts +13 -7
  227. package/dist/src/resources/vector-stores/files.ts +31 -11
  228. package/dist/src/resources/vector-stores/vector-stores.ts +89 -15
  229. package/dist/src/version.ts +1 -1
  230. package/dist/version.d.ts +1 -1
  231. package/dist/version.js +1 -1
  232. package/dist/version.mjs +1 -1
  233. package/package.json +1 -1
  234. package/src/core.ts +6 -0
  235. package/src/index.ts +11 -16
  236. package/src/resources/alpha/admin.ts +84 -0
  237. package/src/resources/alpha/alpha.ts +16 -8
  238. package/src/resources/alpha/benchmarks.ts +37 -1
  239. package/src/resources/alpha/eval/eval.ts +28 -7
  240. package/src/resources/alpha/index.ts +2 -0
  241. package/src/resources/alpha/inference.ts +58 -7
  242. package/src/resources/alpha/post-training/index.ts +2 -10
  243. package/src/resources/alpha/post-training/job.ts +5 -26
  244. package/src/resources/alpha/post-training/post-training.ts +38 -22
  245. package/src/resources/beta/datasets.ts +103 -8
  246. package/src/resources/chat/chat.ts +122 -20
  247. package/src/resources/chat/completions.ts +1138 -79
  248. package/src/resources/completions.ts +137 -28
  249. package/src/resources/conversations/conversations.ts +67 -15
  250. package/src/resources/conversations/items.ts +479 -110
  251. package/src/resources/embeddings.ts +46 -8
  252. package/src/resources/files.ts +58 -31
  253. package/src/resources/index.ts +3 -3
  254. package/src/resources/inspect.ts +3 -57
  255. package/src/resources/models/models.ts +21 -6
  256. package/src/resources/moderations.ts +38 -5
  257. package/src/resources/prompts/prompts.ts +26 -12
  258. package/src/resources/prompts/versions.ts +0 -2
  259. package/src/resources/providers.ts +5 -16
  260. package/src/resources/responses/input-items.ts +138 -15
  261. package/src/resources/responses/responses.ts +1076 -196
  262. package/src/resources/routes.ts +10 -14
  263. package/src/resources/safety.ts +131 -8
  264. package/src/resources/scoring-functions.ts +23 -1
  265. package/src/resources/scoring.ts +24 -0
  266. package/src/resources/shared.ts +148 -7
  267. package/src/resources/shields.ts +21 -0
  268. package/src/resources/vector-io.ts +95 -78
  269. package/src/resources/vector-stores/file-batches.ts +13 -7
  270. package/src/resources/vector-stores/files.ts +31 -11
  271. package/src/resources/vector-stores/vector-stores.ts +89 -15
  272. package/src/version.ts +1 -1
  273. package/tests/api-resources/alpha/admin.test.ts +111 -0
  274. package/tests/api-resources/alpha/eval/eval.test.ts +40 -20
  275. package/tests/api-resources/alpha/inference.test.ts +1 -1
  276. package/tests/api-resources/alpha/post-training/job.test.ts +21 -12
  277. package/tests/api-resources/alpha/post-training/post-training.test.ts +12 -2
  278. package/tests/api-resources/beta/datasets.test.ts +9 -3
  279. package/tests/api-resources/chat/completions.test.ts +20 -7
  280. package/tests/api-resources/completions.test.ts +5 -5
  281. package/tests/api-resources/conversations/items.test.ts +22 -3
  282. package/tests/api-resources/embeddings.test.ts +2 -2
  283. package/tests/api-resources/files.test.ts +6 -1
  284. package/tests/api-resources/responses/input-items.test.ts +7 -1
  285. package/tests/api-resources/responses/responses.test.ts +19 -5
  286. package/tests/api-resources/safety.test.ts +9 -5
  287. package/tests/api-resources/tool-runtime.test.ts +9 -2
  288. package/tests/api-resources/vector-io.test.ts +13 -4
  289. package/tests/api-resources/vector-stores/file-batches.test.ts +7 -1
  290. package/tests/api-resources/vector-stores/files.test.ts +7 -1
  291. package/tests/api-resources/vector-stores/vector-stores.test.ts +14 -2
@@ -102,6 +102,10 @@ export interface ResponseObject {
102
102
 
103
103
  status: string;
104
104
 
105
+ store: boolean;
106
+
107
+ completed_at?: number | null;
108
+
105
109
  /**
106
110
  * Error details for failed OpenAI response requests.
107
111
  */
@@ -109,6 +113,8 @@ export interface ResponseObject {
109
113
 
110
114
  instructions?: string | null;
111
115
 
116
+ max_output_tokens?: number | null;
117
+
112
118
  max_tool_calls?: number | null;
113
119
 
114
120
  metadata?: { [key: string]: string } | null;
@@ -124,6 +130,15 @@ export interface ResponseObject {
124
130
  */
125
131
  prompt?: ResponseObject.Prompt | null;
126
132
 
133
+ /**
134
+ * Configuration for reasoning effort in OpenAI responses.
135
+ *
136
+ * Controls how much reasoning the model performs before generating a response.
137
+ */
138
+ reasoning?: ResponseObject.Reasoning | null;
139
+
140
+ safety_identifier?: string | null;
141
+
127
142
  temperature?: number | null;
128
143
 
129
144
  /**
@@ -131,6 +146,21 @@ export interface ResponseObject {
131
146
  */
132
147
  text?: ResponseObject.Text;
133
148
 
149
+ /**
150
+ * Constrains the tools available to the model to a pre-defined set.
151
+ */
152
+ tool_choice?:
153
+ | 'auto'
154
+ | 'required'
155
+ | 'none'
156
+ | ResponseObject.OpenAIResponseInputToolChoiceAllowedTools
157
+ | ResponseObject.OpenAIResponseInputToolChoiceFileSearch
158
+ | ResponseObject.OpenAIResponseInputToolChoiceWebSearch
159
+ | ResponseObject.OpenAIResponseInputToolChoiceFunctionTool
160
+ | ResponseObject.OpenAIResponseInputToolChoiceMcpTool
161
+ | ResponseObject.OpenAIResponseInputToolChoiceCustomTool
162
+ | null;
163
+
134
164
  tools?: Array<
135
165
  | ResponseObject.OpenAIResponseInputToolWebSearch
136
166
  | ResponseObject.OpenAIResponseInputToolFileSearch
@@ -163,7 +193,7 @@ export namespace ResponseObject {
163
193
  | OpenAIResponseMessageOutput.OpenAIResponseInputMessageContentFile
164
194
  >
165
195
  | Array<
166
- | OpenAIResponseMessageOutput.OpenAIResponseOutputMessageContentOutputText
196
+ | OpenAIResponseMessageOutput.OpenAIResponseOutputMessageContentOutputTextOutput
167
197
  | OpenAIResponseMessageOutput.OpenAIResponseContentPartRefusal
168
198
  >;
169
199
 
@@ -214,20 +244,22 @@ export namespace ResponseObject {
214
244
  type?: 'input_file';
215
245
  }
216
246
 
217
- export interface OpenAIResponseOutputMessageContentOutputText {
247
+ export interface OpenAIResponseOutputMessageContentOutputTextOutput {
218
248
  text: string;
219
249
 
220
250
  annotations?: Array<
221
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFileCitation
222
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationCitation
223
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationContainerFileCitation
224
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFilePath
251
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationFileCitation
252
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationCitation
253
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationContainerFileCitation
254
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationFilePath
225
255
  >;
226
256
 
257
+ logprobs?: Array<OpenAIResponseOutputMessageContentOutputTextOutput.Logprob> | null;
258
+
227
259
  type?: 'output_text';
228
260
  }
229
261
 
230
- export namespace OpenAIResponseOutputMessageContentOutputText {
262
+ export namespace OpenAIResponseOutputMessageContentOutputTextOutput {
231
263
  /**
232
264
  * File citation annotation for referencing specific files in response content.
233
265
  */
@@ -277,6 +309,55 @@ export namespace ResponseObject {
277
309
 
278
310
  type?: 'file_path';
279
311
  }
312
+
313
+ /**
314
+ * The log probability for a token from an OpenAI-compatible chat completion
315
+ * response.
316
+ */
317
+ export interface Logprob {
318
+ /**
319
+ * The token.
320
+ */
321
+ token: string;
322
+
323
+ /**
324
+ * The log probability of the token.
325
+ */
326
+ logprob: number;
327
+
328
+ /**
329
+ * The bytes for the token.
330
+ */
331
+ bytes?: Array<number> | null;
332
+
333
+ /**
334
+ * The top log probabilities for the token.
335
+ */
336
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
337
+ }
338
+
339
+ export namespace Logprob {
340
+ /**
341
+ * The top log probability for a token from an OpenAI-compatible chat completion
342
+ * response.
343
+ */
344
+ export interface TopLogprob {
345
+ /**
346
+ * The token.
347
+ */
348
+ token: string;
349
+
350
+ /**
351
+ * The log probability of the token.
352
+ */
353
+ logprob: number;
354
+
355
+ /**
356
+ * The bytes for the token.
357
+ */
358
+ bytes?: Array<number> | null;
359
+ }
360
+ }
280
361
  }
281
362
 
282
363
  /**
@@ -473,6 +554,15 @@ export namespace ResponseObject {
473
554
  }
474
555
  }
475
556
 
557
+ /**
558
+ * Configuration for reasoning effort in OpenAI responses.
559
+ *
560
+ * Controls how much reasoning the model performs before generating a response.
561
+ */
562
+ export interface Reasoning {
563
+ effort?: 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' | null;
564
+ }
565
+
476
566
  /**
477
567
  * Text response configuration for OpenAI responses.
478
568
  */
@@ -500,6 +590,60 @@ export namespace ResponseObject {
500
590
  }
501
591
  }
502
592
 
593
+ /**
594
+ * Constrains the tools available to the model to a pre-defined set.
595
+ */
596
+ export interface OpenAIResponseInputToolChoiceAllowedTools {
597
+ tools: Array<{ [key: string]: string }>;
598
+
599
+ mode?: 'auto' | 'required';
600
+
601
+ type?: 'allowed_tools';
602
+ }
603
+
604
+ /**
605
+ * Indicates that the model should use file search to generate a response.
606
+ */
607
+ export interface OpenAIResponseInputToolChoiceFileSearch {
608
+ type?: 'file_search';
609
+ }
610
+
611
+ /**
612
+ * Indicates that the model should use web search to generate a response
613
+ */
614
+ export interface OpenAIResponseInputToolChoiceWebSearch {
615
+ type?: 'web_search' | 'web_search_preview' | 'web_search_preview_2025_03_11' | 'web_search_2025_08_26';
616
+ }
617
+
618
+ /**
619
+ * Forces the model to call a specific function.
620
+ */
621
+ export interface OpenAIResponseInputToolChoiceFunctionTool {
622
+ name: string;
623
+
624
+ type?: 'function';
625
+ }
626
+
627
+ /**
628
+ * Forces the model to call a specific tool on a remote MCP server
629
+ */
630
+ export interface OpenAIResponseInputToolChoiceMcpTool {
631
+ server_label: string;
632
+
633
+ name?: string | null;
634
+
635
+ type?: 'mcp';
636
+ }
637
+
638
+ /**
639
+ * Forces the model to call a custom tool.
640
+ */
641
+ export interface OpenAIResponseInputToolChoiceCustomTool {
642
+ name: string;
643
+
644
+ type?: 'custom';
645
+ }
646
+
503
647
  /**
504
648
  * Web search tool configuration for OpenAI response inputs.
505
649
  */
@@ -521,6 +665,22 @@ export namespace ResponseObject {
521
665
 
522
666
  /**
523
667
  * Options for ranking and filtering search results.
668
+ *
669
+ * This class configures how search results are ranked and filtered. You can use
670
+ * algorithm-based rerankers (weighted, RRF) or neural rerankers. Defaults from
671
+ * VectorStoresConfig are used when parameters are not provided.
672
+ *
673
+ * Examples: # Weighted ranker with custom alpha
674
+ * SearchRankingOptions(ranker="weighted", alpha=0.7)
675
+ *
676
+ * # RRF ranker with custom impact factor
677
+ * SearchRankingOptions(ranker="rrf", impact_factor=50.0)
678
+ *
679
+ * # Use config defaults (just specify ranker type)
680
+ * SearchRankingOptions(ranker="weighted") # Uses alpha from VectorStoresConfig
681
+ *
682
+ * # Score threshold filtering
683
+ * SearchRankingOptions(ranker="weighted", score_threshold=0.5)
524
684
  */
525
685
  ranking_options?: OpenAIResponseInputToolFileSearch.RankingOptions | null;
526
686
 
@@ -530,11 +690,48 @@ export namespace ResponseObject {
530
690
  export namespace OpenAIResponseInputToolFileSearch {
531
691
  /**
532
692
  * Options for ranking and filtering search results.
693
+ *
694
+ * This class configures how search results are ranked and filtered. You can use
695
+ * algorithm-based rerankers (weighted, RRF) or neural rerankers. Defaults from
696
+ * VectorStoresConfig are used when parameters are not provided.
697
+ *
698
+ * Examples: # Weighted ranker with custom alpha
699
+ * SearchRankingOptions(ranker="weighted", alpha=0.7)
700
+ *
701
+ * # RRF ranker with custom impact factor
702
+ * SearchRankingOptions(ranker="rrf", impact_factor=50.0)
703
+ *
704
+ * # Use config defaults (just specify ranker type)
705
+ * SearchRankingOptions(ranker="weighted") # Uses alpha from VectorStoresConfig
706
+ *
707
+ * # Score threshold filtering
708
+ * SearchRankingOptions(ranker="weighted", score_threshold=0.5)
533
709
  */
534
710
  export interface RankingOptions {
711
+ /**
712
+ * Weight factor for weighted ranker
713
+ */
714
+ alpha?: number | null;
715
+
716
+ /**
717
+ * Impact factor for RRF algorithm
718
+ */
719
+ impact_factor?: number | null;
720
+
721
+ /**
722
+ * Model identifier for neural reranker
723
+ */
724
+ model?: string | null;
725
+
535
726
  ranker?: string | null;
536
727
 
537
728
  score_threshold?: number | null;
729
+
730
+ /**
731
+ * Weights for combining vector, keyword, and neural scores. Keys: 'vector',
732
+ * 'keyword', 'neural'
733
+ */
734
+ weights?: { [key: string]: number } | null;
538
735
  }
539
736
  }
540
737
 
@@ -582,19 +779,19 @@ export namespace ResponseObject {
582
779
  export interface Usage {
583
780
  input_tokens: number;
584
781
 
585
- output_tokens: number;
586
-
587
- total_tokens: number;
588
-
589
782
  /**
590
783
  * Token details for input tokens in OpenAI response usage.
591
784
  */
592
- input_tokens_details?: Usage.InputTokensDetails | null;
785
+ input_tokens_details: Usage.InputTokensDetails;
786
+
787
+ output_tokens: number;
593
788
 
594
789
  /**
595
790
  * Token details for output tokens in OpenAI response usage.
596
791
  */
597
- output_tokens_details?: Usage.OutputTokensDetails | null;
792
+ output_tokens_details: Usage.OutputTokensDetails;
793
+
794
+ total_tokens: number;
598
795
  }
599
796
 
600
797
  export namespace Usage {
@@ -602,14 +799,14 @@ export namespace ResponseObject {
602
799
  * Token details for input tokens in OpenAI response usage.
603
800
  */
604
801
  export interface InputTokensDetails {
605
- cached_tokens?: number | null;
802
+ cached_tokens: number;
606
803
  }
607
804
 
608
805
  /**
609
806
  * Token details for output tokens in OpenAI response usage.
610
807
  */
611
808
  export interface OutputTokensDetails {
612
- reasoning_tokens?: number | null;
809
+ reasoning_tokens: number;
613
810
  }
614
811
  }
615
812
  }
@@ -785,6 +982,8 @@ export namespace ResponseObjectStream {
785
982
  | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFilePath
786
983
  >;
787
984
 
985
+ logprobs?: Array<OpenAIResponseOutputMessageContentOutputText.Logprob> | null;
986
+
788
987
  type?: 'output_text';
789
988
  }
790
989
 
@@ -838,6 +1037,55 @@ export namespace ResponseObjectStream {
838
1037
 
839
1038
  type?: 'file_path';
840
1039
  }
1040
+
1041
+ /**
1042
+ * The log probability for a token from an OpenAI-compatible chat completion
1043
+ * response.
1044
+ */
1045
+ export interface Logprob {
1046
+ /**
1047
+ * The token.
1048
+ */
1049
+ token: string;
1050
+
1051
+ /**
1052
+ * The log probability of the token.
1053
+ */
1054
+ logprob: number;
1055
+
1056
+ /**
1057
+ * The bytes for the token.
1058
+ */
1059
+ bytes?: Array<number> | null;
1060
+
1061
+ /**
1062
+ * The top log probabilities for the token.
1063
+ */
1064
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
1065
+ }
1066
+
1067
+ export namespace Logprob {
1068
+ /**
1069
+ * The top log probability for a token from an OpenAI-compatible chat completion
1070
+ * response.
1071
+ */
1072
+ export interface TopLogprob {
1073
+ /**
1074
+ * The token.
1075
+ */
1076
+ token: string;
1077
+
1078
+ /**
1079
+ * The log probability of the token.
1080
+ */
1081
+ logprob: number;
1082
+
1083
+ /**
1084
+ * The bytes for the token.
1085
+ */
1086
+ bytes?: Array<number> | null;
1087
+ }
1088
+ }
841
1089
  }
842
1090
 
843
1091
  /**
@@ -1074,6 +1322,8 @@ export namespace ResponseObjectStream {
1074
1322
  | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFilePath
1075
1323
  >;
1076
1324
 
1325
+ logprobs?: Array<OpenAIResponseOutputMessageContentOutputText.Logprob> | null;
1326
+
1077
1327
  type?: 'output_text';
1078
1328
  }
1079
1329
 
@@ -1127,6 +1377,55 @@ export namespace ResponseObjectStream {
1127
1377
 
1128
1378
  type?: 'file_path';
1129
1379
  }
1380
+
1381
+ /**
1382
+ * The log probability for a token from an OpenAI-compatible chat completion
1383
+ * response.
1384
+ */
1385
+ export interface Logprob {
1386
+ /**
1387
+ * The token.
1388
+ */
1389
+ token: string;
1390
+
1391
+ /**
1392
+ * The log probability of the token.
1393
+ */
1394
+ logprob: number;
1395
+
1396
+ /**
1397
+ * The bytes for the token.
1398
+ */
1399
+ bytes?: Array<number> | null;
1400
+
1401
+ /**
1402
+ * The top log probabilities for the token.
1403
+ */
1404
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
1405
+ }
1406
+
1407
+ export namespace Logprob {
1408
+ /**
1409
+ * The top log probability for a token from an OpenAI-compatible chat completion
1410
+ * response.
1411
+ */
1412
+ export interface TopLogprob {
1413
+ /**
1414
+ * The token.
1415
+ */
1416
+ token: string;
1417
+
1418
+ /**
1419
+ * The log probability of the token.
1420
+ */
1421
+ logprob: number;
1422
+
1423
+ /**
1424
+ * The bytes for the token.
1425
+ */
1426
+ bytes?: Array<number> | null;
1427
+ }
1428
+ }
1130
1429
  }
1131
1430
 
1132
1431
  /**
@@ -1274,9 +1573,62 @@ export namespace ResponseObjectStream {
1274
1573
 
1275
1574
  sequence_number: number;
1276
1575
 
1576
+ logprobs?: Array<OpenAIResponseObjectStreamResponseOutputTextDelta.Logprob> | null;
1577
+
1277
1578
  type?: 'response.output_text.delta';
1278
1579
  }
1279
1580
 
1581
+ export namespace OpenAIResponseObjectStreamResponseOutputTextDelta {
1582
+ /**
1583
+ * The log probability for a token from an OpenAI-compatible chat completion
1584
+ * response.
1585
+ */
1586
+ export interface Logprob {
1587
+ /**
1588
+ * The token.
1589
+ */
1590
+ token: string;
1591
+
1592
+ /**
1593
+ * The log probability of the token.
1594
+ */
1595
+ logprob: number;
1596
+
1597
+ /**
1598
+ * The bytes for the token.
1599
+ */
1600
+ bytes?: Array<number> | null;
1601
+
1602
+ /**
1603
+ * The top log probabilities for the token.
1604
+ */
1605
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
1606
+ }
1607
+
1608
+ export namespace Logprob {
1609
+ /**
1610
+ * The top log probability for a token from an OpenAI-compatible chat completion
1611
+ * response.
1612
+ */
1613
+ export interface TopLogprob {
1614
+ /**
1615
+ * The token.
1616
+ */
1617
+ token: string;
1618
+
1619
+ /**
1620
+ * The log probability of the token.
1621
+ */
1622
+ logprob: number;
1623
+
1624
+ /**
1625
+ * The bytes for the token.
1626
+ */
1627
+ bytes?: Array<number> | null;
1628
+ }
1629
+ }
1630
+ }
1631
+
1280
1632
  /**
1281
1633
  * Streaming event for when text output is completed.
1282
1634
  */
@@ -1472,7 +1824,7 @@ export namespace ResponseObjectStream {
1472
1824
  | OpenAIResponseContentPartOutputText.OpenAIResponseAnnotationFilePath
1473
1825
  >;
1474
1826
 
1475
- logprobs?: Array<{ [key: string]: unknown }> | null;
1827
+ logprobs?: Array<OpenAIResponseContentPartOutputText.Logprob> | null;
1476
1828
 
1477
1829
  type?: 'output_text';
1478
1830
  }
@@ -1527,6 +1879,55 @@ export namespace ResponseObjectStream {
1527
1879
 
1528
1880
  type?: 'file_path';
1529
1881
  }
1882
+
1883
+ /**
1884
+ * The log probability for a token from an OpenAI-compatible chat completion
1885
+ * response.
1886
+ */
1887
+ export interface Logprob {
1888
+ /**
1889
+ * The token.
1890
+ */
1891
+ token: string;
1892
+
1893
+ /**
1894
+ * The log probability of the token.
1895
+ */
1896
+ logprob: number;
1897
+
1898
+ /**
1899
+ * The bytes for the token.
1900
+ */
1901
+ bytes?: Array<number> | null;
1902
+
1903
+ /**
1904
+ * The top log probabilities for the token.
1905
+ */
1906
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
1907
+ }
1908
+
1909
+ export namespace Logprob {
1910
+ /**
1911
+ * The top log probability for a token from an OpenAI-compatible chat completion
1912
+ * response.
1913
+ */
1914
+ export interface TopLogprob {
1915
+ /**
1916
+ * The token.
1917
+ */
1918
+ token: string;
1919
+
1920
+ /**
1921
+ * The log probability of the token.
1922
+ */
1923
+ logprob: number;
1924
+
1925
+ /**
1926
+ * The bytes for the token.
1927
+ */
1928
+ bytes?: Array<number> | null;
1929
+ }
1930
+ }
1530
1931
  }
1531
1932
 
1532
1933
  /**
@@ -1587,7 +1988,7 @@ export namespace ResponseObjectStream {
1587
1988
  | OpenAIResponseContentPartOutputText.OpenAIResponseAnnotationFilePath
1588
1989
  >;
1589
1990
 
1590
- logprobs?: Array<{ [key: string]: unknown }> | null;
1991
+ logprobs?: Array<OpenAIResponseContentPartOutputText.Logprob> | null;
1591
1992
 
1592
1993
  type?: 'output_text';
1593
1994
  }
@@ -1642,6 +2043,55 @@ export namespace ResponseObjectStream {
1642
2043
 
1643
2044
  type?: 'file_path';
1644
2045
  }
2046
+
2047
+ /**
2048
+ * The log probability for a token from an OpenAI-compatible chat completion
2049
+ * response.
2050
+ */
2051
+ export interface Logprob {
2052
+ /**
2053
+ * The token.
2054
+ */
2055
+ token: string;
2056
+
2057
+ /**
2058
+ * The log probability of the token.
2059
+ */
2060
+ logprob: number;
2061
+
2062
+ /**
2063
+ * The bytes for the token.
2064
+ */
2065
+ bytes?: Array<number> | null;
2066
+
2067
+ /**
2068
+ * The top log probabilities for the token.
2069
+ */
2070
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
2071
+ }
2072
+
2073
+ export namespace Logprob {
2074
+ /**
2075
+ * The top log probability for a token from an OpenAI-compatible chat completion
2076
+ * response.
2077
+ */
2078
+ export interface TopLogprob {
2079
+ /**
2080
+ * The token.
2081
+ */
2082
+ token: string;
2083
+
2084
+ /**
2085
+ * The log probability of the token.
2086
+ */
2087
+ logprob: number;
2088
+
2089
+ /**
2090
+ * The bytes for the token.
2091
+ */
2092
+ bytes?: Array<number> | null;
2093
+ }
2094
+ }
1645
2095
  }
1646
2096
 
1647
2097
  /**
@@ -2020,6 +2470,10 @@ export interface ResponseListResponse {
2020
2470
 
2021
2471
  status: string;
2022
2472
 
2473
+ store: boolean;
2474
+
2475
+ completed_at?: number | null;
2476
+
2023
2477
  /**
2024
2478
  * Error details for failed OpenAI response requests.
2025
2479
  */
@@ -2027,6 +2481,8 @@ export interface ResponseListResponse {
2027
2481
 
2028
2482
  instructions?: string | null;
2029
2483
 
2484
+ max_output_tokens?: number | null;
2485
+
2030
2486
  max_tool_calls?: number | null;
2031
2487
 
2032
2488
  metadata?: { [key: string]: string } | null;
@@ -2042,6 +2498,15 @@ export interface ResponseListResponse {
2042
2498
  */
2043
2499
  prompt?: ResponseListResponse.Prompt | null;
2044
2500
 
2501
+ /**
2502
+ * Configuration for reasoning effort in OpenAI responses.
2503
+ *
2504
+ * Controls how much reasoning the model performs before generating a response.
2505
+ */
2506
+ reasoning?: ResponseListResponse.Reasoning | null;
2507
+
2508
+ safety_identifier?: string | null;
2509
+
2045
2510
  temperature?: number | null;
2046
2511
 
2047
2512
  /**
@@ -2049,6 +2514,21 @@ export interface ResponseListResponse {
2049
2514
  */
2050
2515
  text?: ResponseListResponse.Text;
2051
2516
 
2517
+ /**
2518
+ * Constrains the tools available to the model to a pre-defined set.
2519
+ */
2520
+ tool_choice?:
2521
+ | 'auto'
2522
+ | 'required'
2523
+ | 'none'
2524
+ | ResponseListResponse.OpenAIResponseInputToolChoiceAllowedTools
2525
+ | ResponseListResponse.OpenAIResponseInputToolChoiceFileSearch
2526
+ | ResponseListResponse.OpenAIResponseInputToolChoiceWebSearch
2527
+ | ResponseListResponse.OpenAIResponseInputToolChoiceFunctionTool
2528
+ | ResponseListResponse.OpenAIResponseInputToolChoiceMcpTool
2529
+ | ResponseListResponse.OpenAIResponseInputToolChoiceCustomTool
2530
+ | null;
2531
+
2052
2532
  tools?: Array<
2053
2533
  | ResponseListResponse.OpenAIResponseInputToolWebSearch
2054
2534
  | ResponseListResponse.OpenAIResponseInputToolFileSearch
@@ -2081,7 +2561,7 @@ export namespace ResponseListResponse {
2081
2561
  | OpenAIResponseMessageOutput.OpenAIResponseInputMessageContentFile
2082
2562
  >
2083
2563
  | Array<
2084
- | OpenAIResponseMessageOutput.OpenAIResponseOutputMessageContentOutputText
2564
+ | OpenAIResponseMessageOutput.OpenAIResponseOutputMessageContentOutputTextOutput
2085
2565
  | OpenAIResponseMessageOutput.OpenAIResponseContentPartRefusal
2086
2566
  >;
2087
2567
 
@@ -2132,20 +2612,22 @@ export namespace ResponseListResponse {
2132
2612
  type?: 'input_file';
2133
2613
  }
2134
2614
 
2135
- export interface OpenAIResponseOutputMessageContentOutputText {
2615
+ export interface OpenAIResponseOutputMessageContentOutputTextOutput {
2136
2616
  text: string;
2137
2617
 
2138
2618
  annotations?: Array<
2139
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFileCitation
2140
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationCitation
2141
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationContainerFileCitation
2142
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFilePath
2619
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationFileCitation
2620
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationCitation
2621
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationContainerFileCitation
2622
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationFilePath
2143
2623
  >;
2144
2624
 
2625
+ logprobs?: Array<OpenAIResponseOutputMessageContentOutputTextOutput.Logprob> | null;
2626
+
2145
2627
  type?: 'output_text';
2146
2628
  }
2147
2629
 
2148
- export namespace OpenAIResponseOutputMessageContentOutputText {
2630
+ export namespace OpenAIResponseOutputMessageContentOutputTextOutput {
2149
2631
  /**
2150
2632
  * File citation annotation for referencing specific files in response content.
2151
2633
  */
@@ -2195,6 +2677,55 @@ export namespace ResponseListResponse {
2195
2677
 
2196
2678
  type?: 'file_path';
2197
2679
  }
2680
+
2681
+ /**
2682
+ * The log probability for a token from an OpenAI-compatible chat completion
2683
+ * response.
2684
+ */
2685
+ export interface Logprob {
2686
+ /**
2687
+ * The token.
2688
+ */
2689
+ token: string;
2690
+
2691
+ /**
2692
+ * The log probability of the token.
2693
+ */
2694
+ logprob: number;
2695
+
2696
+ /**
2697
+ * The bytes for the token.
2698
+ */
2699
+ bytes?: Array<number> | null;
2700
+
2701
+ /**
2702
+ * The top log probabilities for the token.
2703
+ */
2704
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
2705
+ }
2706
+
2707
+ export namespace Logprob {
2708
+ /**
2709
+ * The top log probability for a token from an OpenAI-compatible chat completion
2710
+ * response.
2711
+ */
2712
+ export interface TopLogprob {
2713
+ /**
2714
+ * The token.
2715
+ */
2716
+ token: string;
2717
+
2718
+ /**
2719
+ * The log probability of the token.
2720
+ */
2721
+ logprob: number;
2722
+
2723
+ /**
2724
+ * The bytes for the token.
2725
+ */
2726
+ bytes?: Array<number> | null;
2727
+ }
2728
+ }
2198
2729
  }
2199
2730
 
2200
2731
  /**
@@ -2372,7 +2903,7 @@ export namespace ResponseListResponse {
2372
2903
  | OpenAIResponseMessageOutput.OpenAIResponseInputMessageContentFile
2373
2904
  >
2374
2905
  | Array<
2375
- | OpenAIResponseMessageOutput.OpenAIResponseOutputMessageContentOutputText
2906
+ | OpenAIResponseMessageOutput.OpenAIResponseOutputMessageContentOutputTextOutput
2376
2907
  | OpenAIResponseMessageOutput.OpenAIResponseContentPartRefusal
2377
2908
  >;
2378
2909
 
@@ -2423,20 +2954,22 @@ export namespace ResponseListResponse {
2423
2954
  type?: 'input_file';
2424
2955
  }
2425
2956
 
2426
- export interface OpenAIResponseOutputMessageContentOutputText {
2957
+ export interface OpenAIResponseOutputMessageContentOutputTextOutput {
2427
2958
  text: string;
2428
2959
 
2429
2960
  annotations?: Array<
2430
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFileCitation
2431
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationCitation
2432
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationContainerFileCitation
2433
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFilePath
2961
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationFileCitation
2962
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationCitation
2963
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationContainerFileCitation
2964
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationFilePath
2434
2965
  >;
2435
2966
 
2967
+ logprobs?: Array<OpenAIResponseOutputMessageContentOutputTextOutput.Logprob> | null;
2968
+
2436
2969
  type?: 'output_text';
2437
2970
  }
2438
2971
 
2439
- export namespace OpenAIResponseOutputMessageContentOutputText {
2972
+ export namespace OpenAIResponseOutputMessageContentOutputTextOutput {
2440
2973
  /**
2441
2974
  * File citation annotation for referencing specific files in response content.
2442
2975
  */
@@ -2486,6 +3019,55 @@ export namespace ResponseListResponse {
2486
3019
 
2487
3020
  type?: 'file_path';
2488
3021
  }
3022
+
3023
+ /**
3024
+ * The log probability for a token from an OpenAI-compatible chat completion
3025
+ * response.
3026
+ */
3027
+ export interface Logprob {
3028
+ /**
3029
+ * The token.
3030
+ */
3031
+ token: string;
3032
+
3033
+ /**
3034
+ * The log probability of the token.
3035
+ */
3036
+ logprob: number;
3037
+
3038
+ /**
3039
+ * The bytes for the token.
3040
+ */
3041
+ bytes?: Array<number> | null;
3042
+
3043
+ /**
3044
+ * The top log probabilities for the token.
3045
+ */
3046
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
3047
+ }
3048
+
3049
+ export namespace Logprob {
3050
+ /**
3051
+ * The top log probability for a token from an OpenAI-compatible chat completion
3052
+ * response.
3053
+ */
3054
+ export interface TopLogprob {
3055
+ /**
3056
+ * The token.
3057
+ */
3058
+ token: string;
3059
+
3060
+ /**
3061
+ * The log probability of the token.
3062
+ */
3063
+ logprob: number;
3064
+
3065
+ /**
3066
+ * The bytes for the token.
3067
+ */
3068
+ bytes?: Array<number> | null;
3069
+ }
3070
+ }
2489
3071
  }
2490
3072
 
2491
3073
  /**
@@ -2512,7 +3094,7 @@ export namespace ResponseListResponse {
2512
3094
  | OpenAIResponseMessageOutput.OpenAIResponseInputMessageContentFile
2513
3095
  >
2514
3096
  | Array<
2515
- | OpenAIResponseMessageOutput.OpenAIResponseOutputMessageContentOutputText
3097
+ | OpenAIResponseMessageOutput.OpenAIResponseOutputMessageContentOutputTextOutput
2516
3098
  | OpenAIResponseMessageOutput.OpenAIResponseContentPartRefusal
2517
3099
  >;
2518
3100
 
@@ -2563,20 +3145,22 @@ export namespace ResponseListResponse {
2563
3145
  type?: 'input_file';
2564
3146
  }
2565
3147
 
2566
- export interface OpenAIResponseOutputMessageContentOutputText {
3148
+ export interface OpenAIResponseOutputMessageContentOutputTextOutput {
2567
3149
  text: string;
2568
3150
 
2569
3151
  annotations?: Array<
2570
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFileCitation
2571
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationCitation
2572
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationContainerFileCitation
2573
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFilePath
3152
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationFileCitation
3153
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationCitation
3154
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationContainerFileCitation
3155
+ | OpenAIResponseOutputMessageContentOutputTextOutput.OpenAIResponseAnnotationFilePath
2574
3156
  >;
2575
3157
 
3158
+ logprobs?: Array<OpenAIResponseOutputMessageContentOutputTextOutput.Logprob> | null;
3159
+
2576
3160
  type?: 'output_text';
2577
3161
  }
2578
3162
 
2579
- export namespace OpenAIResponseOutputMessageContentOutputText {
3163
+ export namespace OpenAIResponseOutputMessageContentOutputTextOutput {
2580
3164
  /**
2581
3165
  * File citation annotation for referencing specific files in response content.
2582
3166
  */
@@ -2626,6 +3210,55 @@ export namespace ResponseListResponse {
2626
3210
 
2627
3211
  type?: 'file_path';
2628
3212
  }
3213
+
3214
+ /**
3215
+ * The log probability for a token from an OpenAI-compatible chat completion
3216
+ * response.
3217
+ */
3218
+ export interface Logprob {
3219
+ /**
3220
+ * The token.
3221
+ */
3222
+ token: string;
3223
+
3224
+ /**
3225
+ * The log probability of the token.
3226
+ */
3227
+ logprob: number;
3228
+
3229
+ /**
3230
+ * The bytes for the token.
3231
+ */
3232
+ bytes?: Array<number> | null;
3233
+
3234
+ /**
3235
+ * The top log probabilities for the token.
3236
+ */
3237
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
3238
+ }
3239
+
3240
+ export namespace Logprob {
3241
+ /**
3242
+ * The top log probability for a token from an OpenAI-compatible chat completion
3243
+ * response.
3244
+ */
3245
+ export interface TopLogprob {
3246
+ /**
3247
+ * The token.
3248
+ */
3249
+ token: string;
3250
+
3251
+ /**
3252
+ * The log probability of the token.
3253
+ */
3254
+ logprob: number;
3255
+
3256
+ /**
3257
+ * The bytes for the token.
3258
+ */
3259
+ bytes?: Array<number> | null;
3260
+ }
3261
+ }
2629
3262
  }
2630
3263
 
2631
3264
  /**
@@ -2822,6 +3455,15 @@ export namespace ResponseListResponse {
2822
3455
  }
2823
3456
  }
2824
3457
 
3458
+ /**
3459
+ * Configuration for reasoning effort in OpenAI responses.
3460
+ *
3461
+ * Controls how much reasoning the model performs before generating a response.
3462
+ */
3463
+ export interface Reasoning {
3464
+ effort?: 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' | null;
3465
+ }
3466
+
2825
3467
  /**
2826
3468
  * Text response configuration for OpenAI responses.
2827
3469
  */
@@ -2850,10 +3492,64 @@ export namespace ResponseListResponse {
2850
3492
  }
2851
3493
 
2852
3494
  /**
2853
- * Web search tool configuration for OpenAI response inputs.
3495
+ * Constrains the tools available to the model to a pre-defined set.
2854
3496
  */
2855
- export interface OpenAIResponseInputToolWebSearch {
2856
- search_context_size?: string | null;
3497
+ export interface OpenAIResponseInputToolChoiceAllowedTools {
3498
+ tools: Array<{ [key: string]: string }>;
3499
+
3500
+ mode?: 'auto' | 'required';
3501
+
3502
+ type?: 'allowed_tools';
3503
+ }
3504
+
3505
+ /**
3506
+ * Indicates that the model should use file search to generate a response.
3507
+ */
3508
+ export interface OpenAIResponseInputToolChoiceFileSearch {
3509
+ type?: 'file_search';
3510
+ }
3511
+
3512
+ /**
3513
+ * Indicates that the model should use web search to generate a response
3514
+ */
3515
+ export interface OpenAIResponseInputToolChoiceWebSearch {
3516
+ type?: 'web_search' | 'web_search_preview' | 'web_search_preview_2025_03_11' | 'web_search_2025_08_26';
3517
+ }
3518
+
3519
+ /**
3520
+ * Forces the model to call a specific function.
3521
+ */
3522
+ export interface OpenAIResponseInputToolChoiceFunctionTool {
3523
+ name: string;
3524
+
3525
+ type?: 'function';
3526
+ }
3527
+
3528
+ /**
3529
+ * Forces the model to call a specific tool on a remote MCP server
3530
+ */
3531
+ export interface OpenAIResponseInputToolChoiceMcpTool {
3532
+ server_label: string;
3533
+
3534
+ name?: string | null;
3535
+
3536
+ type?: 'mcp';
3537
+ }
3538
+
3539
+ /**
3540
+ * Forces the model to call a custom tool.
3541
+ */
3542
+ export interface OpenAIResponseInputToolChoiceCustomTool {
3543
+ name: string;
3544
+
3545
+ type?: 'custom';
3546
+ }
3547
+
3548
+ /**
3549
+ * Web search tool configuration for OpenAI response inputs.
3550
+ */
3551
+ export interface OpenAIResponseInputToolWebSearch {
3552
+ search_context_size?: string | null;
2857
3553
 
2858
3554
  type?: 'web_search' | 'web_search_preview' | 'web_search_preview_2025_03_11' | 'web_search_2025_08_26';
2859
3555
  }
@@ -2870,6 +3566,22 @@ export namespace ResponseListResponse {
2870
3566
 
2871
3567
  /**
2872
3568
  * Options for ranking and filtering search results.
3569
+ *
3570
+ * This class configures how search results are ranked and filtered. You can use
3571
+ * algorithm-based rerankers (weighted, RRF) or neural rerankers. Defaults from
3572
+ * VectorStoresConfig are used when parameters are not provided.
3573
+ *
3574
+ * Examples: # Weighted ranker with custom alpha
3575
+ * SearchRankingOptions(ranker="weighted", alpha=0.7)
3576
+ *
3577
+ * # RRF ranker with custom impact factor
3578
+ * SearchRankingOptions(ranker="rrf", impact_factor=50.0)
3579
+ *
3580
+ * # Use config defaults (just specify ranker type)
3581
+ * SearchRankingOptions(ranker="weighted") # Uses alpha from VectorStoresConfig
3582
+ *
3583
+ * # Score threshold filtering
3584
+ * SearchRankingOptions(ranker="weighted", score_threshold=0.5)
2873
3585
  */
2874
3586
  ranking_options?: OpenAIResponseInputToolFileSearch.RankingOptions | null;
2875
3587
 
@@ -2879,11 +3591,48 @@ export namespace ResponseListResponse {
2879
3591
  export namespace OpenAIResponseInputToolFileSearch {
2880
3592
  /**
2881
3593
  * Options for ranking and filtering search results.
3594
+ *
3595
+ * This class configures how search results are ranked and filtered. You can use
3596
+ * algorithm-based rerankers (weighted, RRF) or neural rerankers. Defaults from
3597
+ * VectorStoresConfig are used when parameters are not provided.
3598
+ *
3599
+ * Examples: # Weighted ranker with custom alpha
3600
+ * SearchRankingOptions(ranker="weighted", alpha=0.7)
3601
+ *
3602
+ * # RRF ranker with custom impact factor
3603
+ * SearchRankingOptions(ranker="rrf", impact_factor=50.0)
3604
+ *
3605
+ * # Use config defaults (just specify ranker type)
3606
+ * SearchRankingOptions(ranker="weighted") # Uses alpha from VectorStoresConfig
3607
+ *
3608
+ * # Score threshold filtering
3609
+ * SearchRankingOptions(ranker="weighted", score_threshold=0.5)
2882
3610
  */
2883
3611
  export interface RankingOptions {
3612
+ /**
3613
+ * Weight factor for weighted ranker
3614
+ */
3615
+ alpha?: number | null;
3616
+
3617
+ /**
3618
+ * Impact factor for RRF algorithm
3619
+ */
3620
+ impact_factor?: number | null;
3621
+
3622
+ /**
3623
+ * Model identifier for neural reranker
3624
+ */
3625
+ model?: string | null;
3626
+
2884
3627
  ranker?: string | null;
2885
3628
 
2886
3629
  score_threshold?: number | null;
3630
+
3631
+ /**
3632
+ * Weights for combining vector, keyword, and neural scores. Keys: 'vector',
3633
+ * 'keyword', 'neural'
3634
+ */
3635
+ weights?: { [key: string]: number } | null;
2887
3636
  }
2888
3637
  }
2889
3638
 
@@ -2931,19 +3680,19 @@ export namespace ResponseListResponse {
2931
3680
  export interface Usage {
2932
3681
  input_tokens: number;
2933
3682
 
2934
- output_tokens: number;
2935
-
2936
- total_tokens: number;
2937
-
2938
3683
  /**
2939
3684
  * Token details for input tokens in OpenAI response usage.
2940
3685
  */
2941
- input_tokens_details?: Usage.InputTokensDetails | null;
3686
+ input_tokens_details: Usage.InputTokensDetails;
3687
+
3688
+ output_tokens: number;
2942
3689
 
2943
3690
  /**
2944
3691
  * Token details for output tokens in OpenAI response usage.
2945
3692
  */
2946
- output_tokens_details?: Usage.OutputTokensDetails | null;
3693
+ output_tokens_details: Usage.OutputTokensDetails;
3694
+
3695
+ total_tokens: number;
2947
3696
  }
2948
3697
 
2949
3698
  export namespace Usage {
@@ -2951,14 +3700,14 @@ export namespace ResponseListResponse {
2951
3700
  * Token details for input tokens in OpenAI response usage.
2952
3701
  */
2953
3702
  export interface InputTokensDetails {
2954
- cached_tokens?: number | null;
3703
+ cached_tokens: number;
2955
3704
  }
2956
3705
 
2957
3706
  /**
2958
3707
  * Token details for output tokens in OpenAI response usage.
2959
3708
  */
2960
3709
  export interface OutputTokensDetails {
2961
- reasoning_tokens?: number | null;
3710
+ reasoning_tokens: number;
2962
3711
  }
2963
3712
  }
2964
3713
  }
@@ -2977,6 +3726,9 @@ export interface ResponseDeleteResponse {
2977
3726
  export type ResponseCreateParams = ResponseCreateParamsNonStreaming | ResponseCreateParamsStreaming;
2978
3727
 
2979
3728
  export interface ResponseCreateParamsBase {
3729
+ /**
3730
+ * Input message(s) to create the response.
3731
+ */
2980
3732
  input:
2981
3733
  | string
2982
3734
  | Array<
@@ -2989,25 +3741,69 @@ export interface ResponseCreateParamsBase {
2989
3741
  | ResponseCreateParams.OpenAIResponseMcpApprovalRequest
2990
3742
  | ResponseCreateParams.OpenAIResponseInputFunctionToolCallOutput
2991
3743
  | ResponseCreateParams.OpenAIResponseMcpApprovalResponse
2992
- | ResponseCreateParams.OpenAIResponseMessageInput
2993
3744
  >;
2994
3745
 
3746
+ /**
3747
+ * The underlying LLM used for completions.
3748
+ */
2995
3749
  model: string;
2996
3750
 
3751
+ /**
3752
+ * Optional ID of a conversation to add the response to.
3753
+ */
2997
3754
  conversation?: string | null;
2998
3755
 
2999
- include?: Array<string> | null;
3756
+ /**
3757
+ * List of guardrails to apply during response generation.
3758
+ */
3759
+ guardrails?: Array<string | ResponseCreateParams.ResponseGuardrailSpec> | null;
3760
+
3761
+ /**
3762
+ * Additional fields to include in the response.
3763
+ */
3764
+ include?: Array<
3765
+ | 'web_search_call.action.sources'
3766
+ | 'code_interpreter_call.outputs'
3767
+ | 'computer_call_output.output.image_url'
3768
+ | 'file_search_call.results'
3769
+ | 'message.input_image.image_url'
3770
+ | 'message.output_text.logprobs'
3771
+ | 'reasoning.encrypted_content'
3772
+ > | null;
3000
3773
 
3774
+ /**
3775
+ * Instructions to guide the model's behavior.
3776
+ */
3001
3777
  instructions?: string | null;
3002
3778
 
3779
+ /**
3780
+ * Maximum number of inference iterations.
3781
+ */
3003
3782
  max_infer_iters?: number | null;
3004
3783
 
3784
+ /**
3785
+ * Upper bound for the number of tokens that can be generated for a response.
3786
+ */
3787
+ max_output_tokens?: number | null;
3788
+
3789
+ /**
3790
+ * Max number of total calls to built-in tools that can be processed in a response.
3791
+ */
3005
3792
  max_tool_calls?: number | null;
3006
3793
 
3794
+ /**
3795
+ * Dictionary of metadata key-value pairs to attach to the response.
3796
+ */
3007
3797
  metadata?: { [key: string]: string } | null;
3008
3798
 
3799
+ /**
3800
+ * Whether to enable parallel tool calls.
3801
+ */
3009
3802
  parallel_tool_calls?: boolean | null;
3010
3803
 
3804
+ /**
3805
+ * Optional ID of a previous response to continue from.
3806
+ */
3011
3807
  previous_response_id?: string | null;
3012
3808
 
3013
3809
  /**
@@ -3015,10 +3811,31 @@ export interface ResponseCreateParamsBase {
3015
3811
  */
3016
3812
  prompt?: ResponseCreateParams.Prompt | null;
3017
3813
 
3814
+ /**
3815
+ * Configuration for reasoning effort in OpenAI responses.
3816
+ *
3817
+ * Controls how much reasoning the model performs before generating a response.
3818
+ */
3819
+ reasoning?: ResponseCreateParams.Reasoning | null;
3820
+
3821
+ /**
3822
+ * A stable identifier used for safety monitoring and abuse detection.
3823
+ */
3824
+ safety_identifier?: string | null;
3825
+
3826
+ /**
3827
+ * Whether to store the response in the database.
3828
+ */
3018
3829
  store?: boolean | null;
3019
3830
 
3831
+ /**
3832
+ * Whether to stream the response.
3833
+ */
3020
3834
  stream?: boolean | null;
3021
3835
 
3836
+ /**
3837
+ * Sampling temperature.
3838
+ */
3022
3839
  temperature?: number | null;
3023
3840
 
3024
3841
  /**
@@ -3026,6 +3843,24 @@ export interface ResponseCreateParamsBase {
3026
3843
  */
3027
3844
  text?: ResponseCreateParams.Text | null;
3028
3845
 
3846
+ /**
3847
+ * How the model should select which tool to call (if any).
3848
+ */
3849
+ tool_choice?:
3850
+ | 'auto'
3851
+ | 'required'
3852
+ | 'none'
3853
+ | ResponseCreateParams.OpenAIResponseInputToolChoiceAllowedTools
3854
+ | ResponseCreateParams.OpenAIResponseInputToolChoiceFileSearch
3855
+ | ResponseCreateParams.OpenAIResponseInputToolChoiceWebSearch
3856
+ | ResponseCreateParams.OpenAIResponseInputToolChoiceFunctionTool
3857
+ | ResponseCreateParams.OpenAIResponseInputToolChoiceMcpTool
3858
+ | ResponseCreateParams.OpenAIResponseInputToolChoiceCustomTool
3859
+ | null;
3860
+
3861
+ /**
3862
+ * List of tools available to the model.
3863
+ */
3029
3864
  tools?: Array<
3030
3865
  | ResponseCreateParams.OpenAIResponseInputToolWebSearch
3031
3866
  | ResponseCreateParams.OpenAIResponseInputToolFileSearch
@@ -3049,7 +3884,7 @@ export namespace ResponseCreateParams {
3049
3884
  | OpenAIResponseMessageInput.OpenAIResponseInputMessageContentFile
3050
3885
  >
3051
3886
  | Array<
3052
- | OpenAIResponseMessageInput.OpenAIResponseOutputMessageContentOutputText
3887
+ | OpenAIResponseMessageInput.OpenAIResponseOutputMessageContentOutputTextInput
3053
3888
  | OpenAIResponseMessageInput.OpenAIResponseContentPartRefusal
3054
3889
  >;
3055
3890
 
@@ -3100,20 +3935,22 @@ export namespace ResponseCreateParams {
3100
3935
  type?: 'input_file';
3101
3936
  }
3102
3937
 
3103
- export interface OpenAIResponseOutputMessageContentOutputText {
3938
+ export interface OpenAIResponseOutputMessageContentOutputTextInput {
3104
3939
  text: string;
3105
3940
 
3106
3941
  annotations?: Array<
3107
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFileCitation
3108
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationCitation
3109
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationContainerFileCitation
3110
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFilePath
3942
+ | OpenAIResponseOutputMessageContentOutputTextInput.OpenAIResponseAnnotationFileCitation
3943
+ | OpenAIResponseOutputMessageContentOutputTextInput.OpenAIResponseAnnotationCitation
3944
+ | OpenAIResponseOutputMessageContentOutputTextInput.OpenAIResponseAnnotationContainerFileCitation
3945
+ | OpenAIResponseOutputMessageContentOutputTextInput.OpenAIResponseAnnotationFilePath
3111
3946
  >;
3112
3947
 
3948
+ logprobs?: Array<OpenAIResponseOutputMessageContentOutputTextInput.Logprob> | null;
3949
+
3113
3950
  type?: 'output_text';
3114
3951
  }
3115
3952
 
3116
- export namespace OpenAIResponseOutputMessageContentOutputText {
3953
+ export namespace OpenAIResponseOutputMessageContentOutputTextInput {
3117
3954
  /**
3118
3955
  * File citation annotation for referencing specific files in response content.
3119
3956
  */
@@ -3163,6 +4000,55 @@ export namespace ResponseCreateParams {
3163
4000
 
3164
4001
  type?: 'file_path';
3165
4002
  }
4003
+
4004
+ /**
4005
+ * The log probability for a token from an OpenAI-compatible chat completion
4006
+ * response.
4007
+ */
4008
+ export interface Logprob {
4009
+ /**
4010
+ * The token.
4011
+ */
4012
+ token: string;
4013
+
4014
+ /**
4015
+ * The log probability of the token.
4016
+ */
4017
+ logprob: number;
4018
+
4019
+ /**
4020
+ * The bytes for the token.
4021
+ */
4022
+ bytes?: Array<number> | null;
4023
+
4024
+ /**
4025
+ * The top log probabilities for the token.
4026
+ */
4027
+ top_logprobs?: Array<Logprob.TopLogprob> | null;
4028
+ }
4029
+
4030
+ export namespace Logprob {
4031
+ /**
4032
+ * The top log probability for a token from an OpenAI-compatible chat completion
4033
+ * response.
4034
+ */
4035
+ export interface TopLogprob {
4036
+ /**
4037
+ * The token.
4038
+ */
4039
+ token: string;
4040
+
4041
+ /**
4042
+ * The log probability of the token.
4043
+ */
4044
+ logprob: number;
4045
+
4046
+ /**
4047
+ * The bytes for the token.
4048
+ */
4049
+ bytes?: Array<number> | null;
4050
+ }
4051
+ }
3166
4052
  }
3167
4053
 
3168
4054
  /**
@@ -3327,143 +4213,10 @@ export namespace ResponseCreateParams {
3327
4213
  }
3328
4214
 
3329
4215
  /**
3330
- * Corresponds to the various Message types in the Responses API. They are all
3331
- * under one type because the Responses API gives them all the same "type" value,
3332
- * and there is no way to tell them apart in certain scenarios.
4216
+ * Specification for a guardrail to apply during response generation.
3333
4217
  */
3334
- export interface OpenAIResponseMessageInput {
3335
- content:
3336
- | string
3337
- | Array<
3338
- | OpenAIResponseMessageInput.OpenAIResponseInputMessageContentText
3339
- | OpenAIResponseMessageInput.OpenAIResponseInputMessageContentImage
3340
- | OpenAIResponseMessageInput.OpenAIResponseInputMessageContentFile
3341
- >
3342
- | Array<
3343
- | OpenAIResponseMessageInput.OpenAIResponseOutputMessageContentOutputText
3344
- | OpenAIResponseMessageInput.OpenAIResponseContentPartRefusal
3345
- >;
3346
-
3347
- role: 'system' | 'developer' | 'user' | 'assistant';
3348
-
3349
- id?: string | null;
3350
-
3351
- status?: string | null;
3352
-
3353
- type?: 'message';
3354
- }
3355
-
3356
- export namespace OpenAIResponseMessageInput {
3357
- /**
3358
- * Text content for input messages in OpenAI response format.
3359
- */
3360
- export interface OpenAIResponseInputMessageContentText {
3361
- text: string;
3362
-
3363
- type?: 'input_text';
3364
- }
3365
-
3366
- /**
3367
- * Image content for input messages in OpenAI response format.
3368
- */
3369
- export interface OpenAIResponseInputMessageContentImage {
3370
- detail?: 'low' | 'high' | 'auto';
3371
-
3372
- file_id?: string | null;
3373
-
3374
- image_url?: string | null;
3375
-
3376
- type?: 'input_image';
3377
- }
3378
-
3379
- /**
3380
- * File content for input messages in OpenAI response format.
3381
- */
3382
- export interface OpenAIResponseInputMessageContentFile {
3383
- file_data?: string | null;
3384
-
3385
- file_id?: string | null;
3386
-
3387
- file_url?: string | null;
3388
-
3389
- filename?: string | null;
3390
-
3391
- type?: 'input_file';
3392
- }
3393
-
3394
- export interface OpenAIResponseOutputMessageContentOutputText {
3395
- text: string;
3396
-
3397
- annotations?: Array<
3398
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFileCitation
3399
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationCitation
3400
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationContainerFileCitation
3401
- | OpenAIResponseOutputMessageContentOutputText.OpenAIResponseAnnotationFilePath
3402
- >;
3403
-
3404
- type?: 'output_text';
3405
- }
3406
-
3407
- export namespace OpenAIResponseOutputMessageContentOutputText {
3408
- /**
3409
- * File citation annotation for referencing specific files in response content.
3410
- */
3411
- export interface OpenAIResponseAnnotationFileCitation {
3412
- file_id: string;
3413
-
3414
- filename: string;
3415
-
3416
- index: number;
3417
-
3418
- type?: 'file_citation';
3419
- }
3420
-
3421
- /**
3422
- * URL citation annotation for referencing external web resources.
3423
- */
3424
- export interface OpenAIResponseAnnotationCitation {
3425
- end_index: number;
3426
-
3427
- start_index: number;
3428
-
3429
- title: string;
3430
-
3431
- url: string;
3432
-
3433
- type?: 'url_citation';
3434
- }
3435
-
3436
- export interface OpenAIResponseAnnotationContainerFileCitation {
3437
- container_id: string;
3438
-
3439
- end_index: number;
3440
-
3441
- file_id: string;
3442
-
3443
- filename: string;
3444
-
3445
- start_index: number;
3446
-
3447
- type?: 'container_file_citation';
3448
- }
3449
-
3450
- export interface OpenAIResponseAnnotationFilePath {
3451
- file_id: string;
3452
-
3453
- index: number;
3454
-
3455
- type?: 'file_path';
3456
- }
3457
- }
3458
-
3459
- /**
3460
- * Refusal content within a streamed response part.
3461
- */
3462
- export interface OpenAIResponseContentPartRefusal {
3463
- refusal: string;
3464
-
3465
- type?: 'refusal';
3466
- }
4218
+ export interface ResponseGuardrailSpec {
4219
+ type: string;
3467
4220
  }
3468
4221
 
3469
4222
  /**
@@ -3521,6 +4274,15 @@ export namespace ResponseCreateParams {
3521
4274
  }
3522
4275
  }
3523
4276
 
4277
+ /**
4278
+ * Configuration for reasoning effort in OpenAI responses.
4279
+ *
4280
+ * Controls how much reasoning the model performs before generating a response.
4281
+ */
4282
+ export interface Reasoning {
4283
+ effort?: 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' | null;
4284
+ }
4285
+
3524
4286
  /**
3525
4287
  * Text response configuration for OpenAI responses.
3526
4288
  */
@@ -3548,6 +4310,60 @@ export namespace ResponseCreateParams {
3548
4310
  }
3549
4311
  }
3550
4312
 
4313
+ /**
4314
+ * Constrains the tools available to the model to a pre-defined set.
4315
+ */
4316
+ export interface OpenAIResponseInputToolChoiceAllowedTools {
4317
+ tools: Array<{ [key: string]: string }>;
4318
+
4319
+ mode?: 'auto' | 'required';
4320
+
4321
+ type?: 'allowed_tools';
4322
+ }
4323
+
4324
+ /**
4325
+ * Indicates that the model should use file search to generate a response.
4326
+ */
4327
+ export interface OpenAIResponseInputToolChoiceFileSearch {
4328
+ type?: 'file_search';
4329
+ }
4330
+
4331
+ /**
4332
+ * Indicates that the model should use web search to generate a response
4333
+ */
4334
+ export interface OpenAIResponseInputToolChoiceWebSearch {
4335
+ type?: 'web_search' | 'web_search_preview' | 'web_search_preview_2025_03_11' | 'web_search_2025_08_26';
4336
+ }
4337
+
4338
+ /**
4339
+ * Forces the model to call a specific function.
4340
+ */
4341
+ export interface OpenAIResponseInputToolChoiceFunctionTool {
4342
+ name: string;
4343
+
4344
+ type?: 'function';
4345
+ }
4346
+
4347
+ /**
4348
+ * Forces the model to call a specific tool on a remote MCP server
4349
+ */
4350
+ export interface OpenAIResponseInputToolChoiceMcpTool {
4351
+ server_label: string;
4352
+
4353
+ name?: string | null;
4354
+
4355
+ type?: 'mcp';
4356
+ }
4357
+
4358
+ /**
4359
+ * Forces the model to call a custom tool.
4360
+ */
4361
+ export interface OpenAIResponseInputToolChoiceCustomTool {
4362
+ name: string;
4363
+
4364
+ type?: 'custom';
4365
+ }
4366
+
3551
4367
  /**
3552
4368
  * Web search tool configuration for OpenAI response inputs.
3553
4369
  */
@@ -3569,6 +4385,22 @@ export namespace ResponseCreateParams {
3569
4385
 
3570
4386
  /**
3571
4387
  * Options for ranking and filtering search results.
4388
+ *
4389
+ * This class configures how search results are ranked and filtered. You can use
4390
+ * algorithm-based rerankers (weighted, RRF) or neural rerankers. Defaults from
4391
+ * VectorStoresConfig are used when parameters are not provided.
4392
+ *
4393
+ * Examples: # Weighted ranker with custom alpha
4394
+ * SearchRankingOptions(ranker="weighted", alpha=0.7)
4395
+ *
4396
+ * # RRF ranker with custom impact factor
4397
+ * SearchRankingOptions(ranker="rrf", impact_factor=50.0)
4398
+ *
4399
+ * # Use config defaults (just specify ranker type)
4400
+ * SearchRankingOptions(ranker="weighted") # Uses alpha from VectorStoresConfig
4401
+ *
4402
+ * # Score threshold filtering
4403
+ * SearchRankingOptions(ranker="weighted", score_threshold=0.5)
3572
4404
  */
3573
4405
  ranking_options?: OpenAIResponseInputToolFileSearch.RankingOptions | null;
3574
4406
 
@@ -3578,11 +4410,48 @@ export namespace ResponseCreateParams {
3578
4410
  export namespace OpenAIResponseInputToolFileSearch {
3579
4411
  /**
3580
4412
  * Options for ranking and filtering search results.
4413
+ *
4414
+ * This class configures how search results are ranked and filtered. You can use
4415
+ * algorithm-based rerankers (weighted, RRF) or neural rerankers. Defaults from
4416
+ * VectorStoresConfig are used when parameters are not provided.
4417
+ *
4418
+ * Examples: # Weighted ranker with custom alpha
4419
+ * SearchRankingOptions(ranker="weighted", alpha=0.7)
4420
+ *
4421
+ * # RRF ranker with custom impact factor
4422
+ * SearchRankingOptions(ranker="rrf", impact_factor=50.0)
4423
+ *
4424
+ * # Use config defaults (just specify ranker type)
4425
+ * SearchRankingOptions(ranker="weighted") # Uses alpha from VectorStoresConfig
4426
+ *
4427
+ * # Score threshold filtering
4428
+ * SearchRankingOptions(ranker="weighted", score_threshold=0.5)
3581
4429
  */
3582
4430
  export interface RankingOptions {
4431
+ /**
4432
+ * Weight factor for weighted ranker
4433
+ */
4434
+ alpha?: number | null;
4435
+
4436
+ /**
4437
+ * Impact factor for RRF algorithm
4438
+ */
4439
+ impact_factor?: number | null;
4440
+
4441
+ /**
4442
+ * Model identifier for neural reranker
4443
+ */
4444
+ model?: string | null;
4445
+
3583
4446
  ranker?: string | null;
3584
4447
 
3585
4448
  score_threshold?: number | null;
4449
+
4450
+ /**
4451
+ * Weights for combining vector, keyword, and neural scores. Keys: 'vector',
4452
+ * 'keyword', 'neural'
4453
+ */
4454
+ weights?: { [key: string]: number } | null;
3586
4455
  }
3587
4456
  }
3588
4457
 
@@ -3607,8 +4476,6 @@ export namespace ResponseCreateParams {
3607
4476
  export interface OpenAIResponseInputToolMcp {
3608
4477
  server_label: string;
3609
4478
 
3610
- server_url: string;
3611
-
3612
4479
  /**
3613
4480
  * Filter configuration for restricting which MCP tools can be used.
3614
4481
  */
@@ -3616,6 +4483,8 @@ export namespace ResponseCreateParams {
3616
4483
 
3617
4484
  authorization?: string | null;
3618
4485
 
4486
+ connector_id?: string | null;
4487
+
3619
4488
  headers?: { [key: string]: unknown } | null;
3620
4489
 
3621
4490
  /**
@@ -3623,6 +4492,8 @@ export namespace ResponseCreateParams {
3623
4492
  */
3624
4493
  require_approval?: 'always' | 'never' | OpenAIResponseInputToolMcp.ApprovalFilter;
3625
4494
 
4495
+ server_url?: string | null;
4496
+
3626
4497
  type?: 'mcp';
3627
4498
  }
3628
4499
 
@@ -3649,14 +4520,23 @@ export namespace ResponseCreateParams {
3649
4520
  }
3650
4521
 
3651
4522
  export interface ResponseCreateParamsNonStreaming extends ResponseCreateParamsBase {
4523
+ /**
4524
+ * Whether to stream the response.
4525
+ */
3652
4526
  stream?: false | null;
3653
4527
  }
3654
4528
 
3655
4529
  export interface ResponseCreateParamsStreaming extends ResponseCreateParamsBase {
4530
+ /**
4531
+ * Whether to stream the response.
4532
+ */
3656
4533
  stream: true;
3657
4534
  }
3658
4535
 
3659
4536
  export interface ResponseListParams extends OpenAICursorPageParams {
4537
+ /**
4538
+ * The model to filter responses by.
4539
+ */
3660
4540
  model?: string | null;
3661
4541
 
3662
4542
  /**