hanzoai 8.5.621 → 8.5.640

package/dist/api/ai-api.js

@@ -29,6 +29,72 @@ const base_1 = require("../base");
  */
 const AiApiAxiosParamCreator = function (configuration) {
     return {
+        /**
+         * Asks one model one prompt and answers the reply, which model served it and who paid. The call is the caller\'s own POST /v1/chat/completions, run as the caller, so it is gated, billed and refused exactly as that route is; a refusal names its code and what lifts it, never an amount. An empty model asks the deployment\'s default Hanzo model.
+         * @summary Asks one model one prompt and answers the reply, which model served it and who paid.
+         * @param {AiChatIn} aiChatIn
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        aiChat: async (aiChatIn, options = {}) => {
+            // verify required parameter 'aiChatIn' is not null or undefined
+            (0, common_1.assertParamExists)('aiChat', 'aiChatIn', aiChatIn);
+            const localVarPath = `/v1/ai/mcp/chat`;
+            // use dummy base URL string because the URL constructor only accepts absolute URLs.
+            const localVarUrlObj = new URL(localVarPath, common_1.DUMMY_BASE_URL);
+            let baseOptions;
+            if (configuration) {
+                baseOptions = configuration.baseOptions;
+            }
+            const localVarRequestOptions = { method: 'POST', ...baseOptions, ...options };
+            const localVarHeaderParameter = {};
+            const localVarQueryParameter = {};
+            // authentication bearer required
+            // http bearer authentication required
+            await (0, common_1.setBearerAuthToObject)(localVarHeaderParameter, configuration);
+            localVarHeaderParameter['Content-Type'] = 'application/json';
+            (0, common_1.setSearchParams)(localVarUrlObj, localVarQueryParameter);
+            let headersFromBaseOptions = baseOptions && baseOptions.headers ? baseOptions.headers : {};
+            localVarRequestOptions.headers = { ...localVarHeaderParameter, ...headersFromBaseOptions, ...options.headers };
+            localVarRequestOptions.data = (0, common_1.serializeDataIfNeeded)(aiChatIn, localVarRequestOptions, configuration);
+            return {
+                url: (0, common_1.toPathString)(localVarUrlObj),
+                options: localVarRequestOptions,
+            };
+        },
+        /**
+         * Runs one decision — the caller\'s own POST /v1/decisions, gated and billed exactly as that route is — and answers the decision, which model served it and who paid. A refusal names its code and what lifts it, never an amount.
+         * @summary Runs one decision — the caller\'s own POST /v1/decisions, gated and billed exactly as that route is — and answers the decision, which model served it and who paid.
+         * @param {AiDecideIn} aiDecideIn
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        aiDecide: async (aiDecideIn, options = {}) => {
+            // verify required parameter 'aiDecideIn' is not null or undefined
+            (0, common_1.assertParamExists)('aiDecide', 'aiDecideIn', aiDecideIn);
+            const localVarPath = `/v1/ai/mcp/decisions`;
+            // use dummy base URL string because the URL constructor only accepts absolute URLs.
+            const localVarUrlObj = new URL(localVarPath, common_1.DUMMY_BASE_URL);
+            let baseOptions;
+            if (configuration) {
+                baseOptions = configuration.baseOptions;
+            }
+            const localVarRequestOptions = { method: 'POST', ...baseOptions, ...options };
+            const localVarHeaderParameter = {};
+            const localVarQueryParameter = {};
+            // authentication bearer required
+            // http bearer authentication required
+            await (0, common_1.setBearerAuthToObject)(localVarHeaderParameter, configuration);
+            localVarHeaderParameter['Content-Type'] = 'application/json';
+            (0, common_1.setSearchParams)(localVarUrlObj, localVarQueryParameter);
+            let headersFromBaseOptions = baseOptions && baseOptions.headers ? baseOptions.headers : {};
+            localVarRequestOptions.headers = { ...localVarHeaderParameter, ...headersFromBaseOptions, ...options.headers };
+            localVarRequestOptions.data = (0, common_1.serializeDataIfNeeded)(aiDecideIn, localVarRequestOptions, configuration);
+            return {
+                url: (0, common_1.toPathString)(localVarUrlObj),
+                options: localVarRequestOptions,
+            };
+        },
         /**
          * Reads the caller\'s plan usage: for each class of model the plan includes, the share used of the billing period and of its short window, who pays for the next request, and whether the caller is in limited mode and why; the share used of the session and the day; and the ways on. Shares only, never amounts.
          * @summary Reads the caller\'s plan usage: for each class of model the plan includes, the share used of the billing period and of its short window, who pays for the next request, and whether the caller is in limited mode and why; the share used of the session and the day; and the ways on.
@@ -89,6 +155,54 @@ const AiApiAxiosParamCreator = function (configuration) {
                 options: localVarRequestOptions,
             };
         },
+        /**
+         * Searches the model catalog — the one GET /v1/models lists — by text, class, family or capability, and answers each match with its class, family, list price (per token and per million; variable for a router billed at the answering model\'s price), context window and capabilities.
+         * @summary Searches the model catalog — the one GET /v1/models lists — by text, class, family or capability, and answers each match with its class, family, list price (per token and per million; variable for a router billed at the answering model\'s price), context window and capabilities.
+         * @param {string} [q] Q matches the model\'s id, name, description or owner, ignoring case.
+         * @param {string} [_class] Class keeps one class: premium (third-party frontier models), ours (Hanzo\'s priced models) or free.
+         * @param {string} [family] Family keeps one Hanzo family: enso, zen, kai or zoo.
+         * @param {string} [capability] Capability keeps the models that have it: tools, vision or reasoning, or a modality they take or give (text, image, audio, file, video, decision).
+         * @param {number} [limit] Limit is the most models to answer, 1 to 500; 0 answers 50.
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        aiModels: async (q, _class, family, capability, limit, options = {}) => {
+            const localVarPath = `/v1/ai/mcp/models`;
+            // use dummy base URL string because the URL constructor only accepts absolute URLs.
+            const localVarUrlObj = new URL(localVarPath, common_1.DUMMY_BASE_URL);
+            let baseOptions;
+            if (configuration) {
+                baseOptions = configuration.baseOptions;
+            }
+            const localVarRequestOptions = { method: 'GET', ...baseOptions, ...options };
+            const localVarHeaderParameter = {};
+            const localVarQueryParameter = {};
+            // authentication bearer required
+            // http bearer authentication required
+            await (0, common_1.setBearerAuthToObject)(localVarHeaderParameter, configuration);
+            if (q !== undefined) {
+                localVarQueryParameter['q'] = q;
+            }
+            if (_class !== undefined) {
+                localVarQueryParameter['class'] = _class;
+            }
+            if (family !== undefined) {
+                localVarQueryParameter['family'] = family;
+            }
+            if (capability !== undefined) {
+                localVarQueryParameter['capability'] = capability;
+            }
+            if (limit !== undefined) {
+                localVarQueryParameter['limit'] = limit;
+            }
+            (0, common_1.setSearchParams)(localVarUrlObj, localVarQueryParameter);
+            let headersFromBaseOptions = baseOptions && baseOptions.headers ? baseOptions.headers : {};
+            localVarRequestOptions.headers = { ...localVarHeaderParameter, ...headersFromBaseOptions, ...options.headers };
+            return {
+                url: (0, common_1.toPathString)(localVarUrlObj),
+                options: localVarRequestOptions,
+            };
+        },
         /**
          * Sets the payer\'s choice to keep using a model on credits once the plan\'s included usage of it is spent. A pooled org wallet is its org admin\'s to set; a person\'s own wallet is theirs. Every change is on the audit trail, before and after; no trail, no change. Answers the limits as they read after it.
          * @summary Sets the payer\'s choice to keep using a model on credits once the plan\'s included usage of it is spent.
@@ -6350,10 +6464,13 @@ const AiApiAxiosParamCreator = function (configuration) {
         /**
          * Attaches a per-request outcome reward to the routing decision that served request_id — the enso training loop\'s quality signal. Org-scoped via the same session-OR-Bearer principal the usage read uses (RequirePrincipal): the reward lands only on the caller\'s OWN org\'s event, so a request_id from another org (or unknown) is a 404 — cross-org writes are impossible and unknown ids are indistinguishable from foreign ones. Idempotent: a repeat overwrites. The body carries NO prompt text — only {request_id, reward|rating}.
          * @summary Attaches a per-request outcome reward to the routing decision that served request_id — the enso training loop\'s quality signal.
+         * @param {AiRoutingRewardRequest} aiRoutingRewardRequest
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        postAiFeedback: async (options = {}) => {
+        postAiFeedback: async (aiRoutingRewardRequest, options = {}) => {
+            // verify required parameter 'aiRoutingRewardRequest' is not null or undefined
+            (0, common_1.assertParamExists)('postAiFeedback', 'aiRoutingRewardRequest', aiRoutingRewardRequest);
             const localVarPath = `/v1/ai/feedback`;
             // use dummy base URL string because the URL constructor only accepts absolute URLs.
             const localVarUrlObj = new URL(localVarPath, common_1.DUMMY_BASE_URL);
@@ -6367,9 +6484,11 @@ const AiApiAxiosParamCreator = function (configuration) {
             // authentication bearer required
             // http bearer authentication required
             await (0, common_1.setBearerAuthToObject)(localVarHeaderParameter, configuration);
+            localVarHeaderParameter['Content-Type'] = 'application/json';
             (0, common_1.setSearchParams)(localVarUrlObj, localVarQueryParameter);
             let headersFromBaseOptions = baseOptions && baseOptions.headers ? baseOptions.headers : {};
             localVarRequestOptions.headers = { ...localVarHeaderParameter, ...headersFromBaseOptions, ...options.headers };
+            localVarRequestOptions.data = (0, common_1.serializeDataIfNeeded)(aiRoutingRewardRequest, localVarRequestOptions, configuration);
             return {
                 url: (0, common_1.toPathString)(localVarUrlObj),
                 options: localVarRequestOptions,
@@ -8421,10 +8540,12 @@ const AiApiAxiosParamCreator = function (configuration) {
          * Implements the OpenAI-compatible chat completions API
          * @summary Implements the OpenAI-compatible chat completions API
          * @param {OpenaiChatCompletionRequest} openaiChatCompletionRequest
+         * @param {number} [xMaxCost] The most this request may cost, in USD per 1,000 tokens. Routing picks only models under it; an org\'s own ceiling fills it when it is absent, and the lower of the two holds.
+         * @param {number} [xMaxLatencyMs] The slowest model this request accepts, in milliseconds.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        postChat: async (openaiChatCompletionRequest, options = {}) => {
+        postChat: async (openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options = {}) => {
             // verify required parameter 'openaiChatCompletionRequest' is not null or undefined
             (0, common_1.assertParamExists)('postChat', 'openaiChatCompletionRequest', openaiChatCompletionRequest);
             const localVarPath = `/v1/chat`;
@@ -8441,6 +8562,16 @@ const AiApiAxiosParamCreator = function (configuration) {
             // http bearer authentication required
             await (0, common_1.setBearerAuthToObject)(localVarHeaderParameter, configuration);
             localVarHeaderParameter['Content-Type'] = 'application/json';
+            if (xMaxCost != null) {
+                localVarHeaderParameter['X-Max-Cost'] = typeof xMaxCost === 'string'
+                    ? xMaxCost
+                    : JSON.stringify(xMaxCost);
+            }
+            if (xMaxLatencyMs != null) {
+                localVarHeaderParameter['X-Max-Latency-Ms'] = typeof xMaxLatencyMs === 'string'
+                    ? xMaxLatencyMs
+                    : JSON.stringify(xMaxLatencyMs);
+            }
             (0, common_1.setSearchParams)(localVarUrlObj, localVarQueryParameter);
             let headersFromBaseOptions = baseOptions && baseOptions.headers ? baseOptions.headers : {};
             localVarRequestOptions.headers = { ...localVarHeaderParameter, ...headersFromBaseOptions, ...options.headers };
@@ -8454,10 +8585,12 @@ const AiApiAxiosParamCreator = function (configuration) {
          * Implements the OpenAI-compatible chat completions API
          * @summary Implements the OpenAI-compatible chat completions API
          * @param {OpenaiChatCompletionRequest} openaiChatCompletionRequest
+         * @param {number} [xMaxCost] The most this request may cost, in USD per 1,000 tokens. Routing picks only models under it; an org\'s own ceiling fills it when it is absent, and the lower of the two holds.
+         * @param {number} [xMaxLatencyMs] The slowest model this request accepts, in milliseconds.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        postChatCompletions: async (openaiChatCompletionRequest, options = {}) => {
+        postChatCompletions: async (openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options = {}) => {
             // verify required parameter 'openaiChatCompletionRequest' is not null or undefined
             (0, common_1.assertParamExists)('postChatCompletions', 'openaiChatCompletionRequest', openaiChatCompletionRequest);
             const localVarPath = `/v1/chat/completions`;
@@ -8474,6 +8607,16 @@ const AiApiAxiosParamCreator = function (configuration) {
             // http bearer authentication required
             await (0, common_1.setBearerAuthToObject)(localVarHeaderParameter, configuration);
             localVarHeaderParameter['Content-Type'] = 'application/json';
+            if (xMaxCost != null) {
+                localVarHeaderParameter['X-Max-Cost'] = typeof xMaxCost === 'string'
+                    ? xMaxCost
+                    : JSON.stringify(xMaxCost);
+            }
+            if (xMaxLatencyMs != null) {
+                localVarHeaderParameter['X-Max-Latency-Ms'] = typeof xMaxLatencyMs === 'string'
+                    ? xMaxLatencyMs
+                    : JSON.stringify(xMaxLatencyMs);
+            }
             (0, common_1.setSearchParams)(localVarUrlObj, localVarQueryParameter);
             let headersFromBaseOptions = baseOptions && baseOptions.headers ? baseOptions.headers : {};
             localVarRequestOptions.headers = { ...localVarHeaderParameter, ...headersFromBaseOptions, ...options.headers };
@@ -8520,10 +8663,12 @@ const AiApiAxiosParamCreator = function (configuration) {
          * Implements the OpenAI-compatible chat completions API
          * @summary Implements the OpenAI-compatible chat completions API
          * @param {OpenaiChatCompletionRequest} openaiChatCompletionRequest
+         * @param {number} [xMaxCost] The most this request may cost, in USD per 1,000 tokens. Routing picks only models under it; an org\'s own ceiling fills it when it is absent, and the lower of the two holds.
+         * @param {number} [xMaxLatencyMs] The slowest model this request accepts, in milliseconds.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        postCompletions: async (openaiChatCompletionRequest, options = {}) => {
+        postCompletions: async (openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options = {}) => {
             // verify required parameter 'openaiChatCompletionRequest' is not null or undefined
             (0, common_1.assertParamExists)('postCompletions', 'openaiChatCompletionRequest', openaiChatCompletionRequest);
             const localVarPath = `/v1/completions`;
@@ -8540,6 +8685,16 @@ const AiApiAxiosParamCreator = function (configuration) {
             // http bearer authentication required
             await (0, common_1.setBearerAuthToObject)(localVarHeaderParameter, configuration);
             localVarHeaderParameter['Content-Type'] = 'application/json';
+            if (xMaxCost != null) {
+                localVarHeaderParameter['X-Max-Cost'] = typeof xMaxCost === 'string'
+                    ? xMaxCost
+                    : JSON.stringify(xMaxCost);
+            }
+            if (xMaxLatencyMs != null) {
+                localVarHeaderParameter['X-Max-Latency-Ms'] = typeof xMaxLatencyMs === 'string'
+                    ? xMaxLatencyMs
+                    : JSON.stringify(xMaxLatencyMs);
+            }
             (0, common_1.setSearchParams)(localVarUrlObj, localVarQueryParameter);
             let headersFromBaseOptions = baseOptions && baseOptions.headers ? baseOptions.headers : {};
             localVarRequestOptions.headers = { ...localVarHeaderParameter, ...headersFromBaseOptions, ...options.headers };
@@ -8767,10 +8922,12 @@ const AiApiAxiosParamCreator = function (configuration) {
         /**
          * Implements POST /v1/responses. The converted request is completed by the chat path, which is handed a sink saying where the answer goes: a stream is translated as it is produced, a whole body is translated entire.
          * @summary Implements POST /v1/responses.
+         * @param {number} [xMaxCost] The most this request may cost, in USD per 1,000 tokens. Routing picks only models under it; an org\'s own ceiling fills it when it is absent, and the lower of the two holds.
+         * @param {number} [xMaxLatencyMs] The slowest model this request accepts, in milliseconds.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        postResponses: async (options = {}) => {
+        postResponses: async (xMaxCost, xMaxLatencyMs, options = {}) => {
             const localVarPath = `/v1/responses`;
             // use dummy base URL string because the URL constructor only accepts absolute URLs.
             const localVarUrlObj = new URL(localVarPath, common_1.DUMMY_BASE_URL);
@@ -8784,6 +8941,16 @@ const AiApiAxiosParamCreator = function (configuration) {
             // authentication bearer required
             // http bearer authentication required
             await (0, common_1.setBearerAuthToObject)(localVarHeaderParameter, configuration);
+            if (xMaxCost != null) {
+                localVarHeaderParameter['X-Max-Cost'] = typeof xMaxCost === 'string'
+                    ? xMaxCost
+                    : JSON.stringify(xMaxCost);
+            }
+            if (xMaxLatencyMs != null) {
+                localVarHeaderParameter['X-Max-Latency-Ms'] = typeof xMaxLatencyMs === 'string'
+                    ? xMaxLatencyMs
+                    : JSON.stringify(xMaxLatencyMs);
+            }
             (0, common_1.setSearchParams)(localVarUrlObj, localVarQueryParameter);
             let headersFromBaseOptions = baseOptions && baseOptions.headers ? baseOptions.headers : {};
             localVarRequestOptions.headers = { ...localVarHeaderParameter, ...headersFromBaseOptions, ...options.headers };
@@ -10068,6 +10235,32 @@ exports.AiApiAxiosParamCreator = AiApiAxiosParamCreator;
 const AiApiFp = function (configuration) {
     const localVarAxiosParamCreator = (0, exports.AiApiAxiosParamCreator)(configuration);
     return {
+        /**
+         * Asks one model one prompt and answers the reply, which model served it and who paid. The call is the caller\'s own POST /v1/chat/completions, run as the caller, so it is gated, billed and refused exactly as that route is; a refusal names its code and what lifts it, never an amount. An empty model asks the deployment\'s default Hanzo model.
+         * @summary Asks one model one prompt and answers the reply, which model served it and who paid.
+         * @param {AiChatIn} aiChatIn
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        async aiChat(aiChatIn, options) {
+            const localVarAxiosArgs = await localVarAxiosParamCreator.aiChat(aiChatIn, options);
+            const localVarOperationServerIndex = configuration?.serverIndex ?? 0;
+            const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.aiChat']?.[localVarOperationServerIndex]?.url;
+            return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
+        },
+        /**
+         * Runs one decision — the caller\'s own POST /v1/decisions, gated and billed exactly as that route is — and answers the decision, which model served it and who paid. A refusal names its code and what lifts it, never an amount.
+         * @summary Runs one decision — the caller\'s own POST /v1/decisions, gated and billed exactly as that route is — and answers the decision, which model served it and who paid.
+         * @param {AiDecideIn} aiDecideIn
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        async aiDecide(aiDecideIn, options) {
+            const localVarAxiosArgs = await localVarAxiosParamCreator.aiDecide(aiDecideIn, options);
+            const localVarOperationServerIndex = configuration?.serverIndex ?? 0;
+            const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.aiDecide']?.[localVarOperationServerIndex]?.url;
+            return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
+        },
         /**
          * Reads the caller\'s plan usage: for each class of model the plan includes, the share used of the billing period and of its short window, who pays for the next request, and whether the caller is in limited mode and why; the share used of the session and the day; and the ways on. Shares only, never amounts.
          * @summary Reads the caller\'s plan usage: for each class of model the plan includes, the share used of the billing period and of its short window, who pays for the next request, and whether the caller is in limited mode and why; the share used of the session and the day; and the ways on.
@@ -10093,6 +10286,23 @@ const AiApiFp = function (configuration) {
             const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.aiMCPTools']?.[localVarOperationServerIndex]?.url;
             return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
         },
+        /**
+         * Searches the model catalog — the one GET /v1/models lists — by text, class, family or capability, and answers each match with its class, family, list price (per token and per million; variable for a router billed at the answering model\'s price), context window and capabilities.
+         * @summary Searches the model catalog — the one GET /v1/models lists — by text, class, family or capability, and answers each match with its class, family, list price (per token and per million; variable for a router billed at the answering model\'s price), context window and capabilities.
+         * @param {string} [q] Q matches the model\'s id, name, description or owner, ignoring case.
+         * @param {string} [_class] Class keeps one class: premium (third-party frontier models), ours (Hanzo\'s priced models) or free.
+         * @param {string} [family] Family keeps one Hanzo family: enso, zen, kai or zoo.
+         * @param {string} [capability] Capability keeps the models that have it: tools, vision or reasoning, or a modality they take or give (text, image, audio, file, video, decision).
+         * @param {number} [limit] Limit is the most models to answer, 1 to 500; 0 answers 50.
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        async aiModels(q, _class, family, capability, limit, options) {
+            const localVarAxiosArgs = await localVarAxiosParamCreator.aiModels(q, _class, family, capability, limit, options);
+            const localVarOperationServerIndex = configuration?.serverIndex ?? 0;
+            const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.aiModels']?.[localVarOperationServerIndex]?.url;
+            return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
+        },
         /**
          * Sets the payer\'s choice to keep using a model on credits once the plan\'s included usage of it is spent. A pooled org wallet is its org admin\'s to set; a person\'s own wallet is theirs. Every change is on the audit trail, before and after; no trail, no change. Answers the limits as they read after it.
          * @summary Sets the payer\'s choice to keep using a model on credits once the plan\'s included usage of it is spent.
@@ -12627,11 +12837,12 @@ const AiApiFp = function (configuration) {
         /**
          * Attaches a per-request outcome reward to the routing decision that served request_id — the enso training loop\'s quality signal. Org-scoped via the same session-OR-Bearer principal the usage read uses (RequirePrincipal): the reward lands only on the caller\'s OWN org\'s event, so a request_id from another org (or unknown) is a 404 — cross-org writes are impossible and unknown ids are indistinguishable from foreign ones. Idempotent: a repeat overwrites. The body carries NO prompt text — only {request_id, reward|rating}.
          * @summary Attaches a per-request outcome reward to the routing decision that served request_id — the enso training loop\'s quality signal.
+         * @param {AiRoutingRewardRequest} aiRoutingRewardRequest
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        async postAiFeedback(options) {
-            const localVarAxiosArgs = await localVarAxiosParamCreator.postAiFeedback(options);
+        async postAiFeedback(aiRoutingRewardRequest, options) {
+            const localVarAxiosArgs = await localVarAxiosParamCreator.postAiFeedback(aiRoutingRewardRequest, options);
             const localVarOperationServerIndex = configuration?.serverIndex ?? 0;
             const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.postAiFeedback']?.[localVarOperationServerIndex]?.url;
             return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
@@ -13461,11 +13672,13 @@ const AiApiFp = function (configuration) {
          * Implements the OpenAI-compatible chat completions API
          * @summary Implements the OpenAI-compatible chat completions API
          * @param {OpenaiChatCompletionRequest} openaiChatCompletionRequest
+         * @param {number} [xMaxCost] The most this request may cost, in USD per 1,000 tokens. Routing picks only models under it; an org\'s own ceiling fills it when it is absent, and the lower of the two holds.
+         * @param {number} [xMaxLatencyMs] The slowest model this request accepts, in milliseconds.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        async postChat(openaiChatCompletionRequest, options) {
-            const localVarAxiosArgs = await localVarAxiosParamCreator.postChat(openaiChatCompletionRequest, options);
+        async postChat(openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options) {
+            const localVarAxiosArgs = await localVarAxiosParamCreator.postChat(openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options);
             const localVarOperationServerIndex = configuration?.serverIndex ?? 0;
             const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.postChat']?.[localVarOperationServerIndex]?.url;
             return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
@@ -13474,11 +13687,13 @@ const AiApiFp = function (configuration) {
          * Implements the OpenAI-compatible chat completions API
          * @summary Implements the OpenAI-compatible chat completions API
          * @param {OpenaiChatCompletionRequest} openaiChatCompletionRequest
+         * @param {number} [xMaxCost] The most this request may cost, in USD per 1,000 tokens. Routing picks only models under it; an org\'s own ceiling fills it when it is absent, and the lower of the two holds.
+         * @param {number} [xMaxLatencyMs] The slowest model this request accepts, in milliseconds.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        async postChatCompletions(openaiChatCompletionRequest, options) {
-            const localVarAxiosArgs = await localVarAxiosParamCreator.postChatCompletions(openaiChatCompletionRequest, options);
+        async postChatCompletions(openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options) {
+            const localVarAxiosArgs = await localVarAxiosParamCreator.postChatCompletions(openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options);
             const localVarOperationServerIndex = configuration?.serverIndex ?? 0;
             const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.postChatCompletions']?.[localVarOperationServerIndex]?.url;
             return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
@@ -13500,11 +13715,13 @@ const AiApiFp = function (configuration) {
          * Implements the OpenAI-compatible chat completions API
          * @summary Implements the OpenAI-compatible chat completions API
          * @param {OpenaiChatCompletionRequest} openaiChatCompletionRequest
+         * @param {number} [xMaxCost] The most this request may cost, in USD per 1,000 tokens. Routing picks only models under it; an org\'s own ceiling fills it when it is absent, and the lower of the two holds.
+         * @param {number} [xMaxLatencyMs] The slowest model this request accepts, in milliseconds.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        async postCompletions(openaiChatCompletionRequest, options) {
-            const localVarAxiosArgs = await localVarAxiosParamCreator.postCompletions(openaiChatCompletionRequest, options);
+        async postCompletions(openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options) {
+            const localVarAxiosArgs = await localVarAxiosParamCreator.postCompletions(openaiChatCompletionRequest, xMaxCost, xMaxLatencyMs, options);
             const localVarOperationServerIndex = configuration?.serverIndex ?? 0;
             const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.postCompletions']?.[localVarOperationServerIndex]?.url;
             return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
@@ -13600,11 +13817,13 @@ const AiApiFp = function (configuration) {
         /**
          * Implements POST /v1/responses. The converted request is completed by the chat path, which is handed a sink saying where the answer goes: a stream is translated as it is produced, a whole body is translated entire.
          * @summary Implements POST /v1/responses.
+         * @param {number} [xMaxCost] The most this request may cost, in USD per 1,000 tokens. Routing picks only models under it; an org\'s own ceiling fills it when it is absent, and the lower of the two holds.
+         * @param {number} [xMaxLatencyMs] The slowest model this request accepts, in milliseconds.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        async postResponses(options) {
-            const localVarAxiosArgs = await localVarAxiosParamCreator.postResponses(options);
+        async postResponses(xMaxCost, xMaxLatencyMs, options) {
+            const localVarAxiosArgs = await localVarAxiosParamCreator.postResponses(xMaxCost, xMaxLatencyMs, options);
             const localVarOperationServerIndex = configuration?.serverIndex ?? 0;
             const localVarOperationServerBasePath = base_1.operationServerMap['AiApi.postResponses']?.[localVarOperationServerIndex]?.url;
             return (axios, basePath) => (0, common_1.createRequestFunction)(localVarAxiosArgs, axios_1.default, base_1.BASE_PATH, configuration)(axios, localVarOperationServerBasePath || basePath);
@@ -14099,6 +14318,26 @@ exports.AiApiFp = AiApiFp;
 const AiApiFactory = function (configuration, basePath, axios) {
     const localVarFp = (0, exports.AiApiFp)(configuration);
     return {
+        /**
+         * Asks one model one prompt and answers the reply, which model served it and who paid. The call is the caller\'s own POST /v1/chat/completions, run as the caller, so it is gated, billed and refused exactly as that route is; a refusal names its code and what lifts it, never an amount. An empty model asks the deployment\'s default Hanzo model.
+         * @summary Asks one model one prompt and answers the reply, which model served it and who paid.
+         * @param {AiApiAiChatRequest} requestParameters Request parameters.
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        aiChat(requestParameters, options) {
+            return localVarFp.aiChat(requestParameters.aiChatIn, options).then((request) => request(axios, basePath));
+        },
+        /**
+         * Runs one decision — the caller\'s own POST /v1/decisions, gated and billed exactly as that route is — and answers the decision, which model served it and who paid. A refusal names its code and what lifts it, never an amount.
+         * @summary Runs one decision — the caller\'s own POST /v1/decisions, gated and billed exactly as that route is — and answers the decision, which model served it and who paid.
+         * @param {AiApiAiDecideRequest} requestParameters Request parameters.
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        aiDecide(requestParameters, options) {
+            return localVarFp.aiDecide(requestParameters.aiDecideIn, options).then((request) => request(axios, basePath));
+        },
         /**
          * Reads the caller\'s plan usage: for each class of model the plan includes, the share used of the billing period and of its short window, who pays for the next request, and whether the caller is in limited mode and why; the share used of the session and the day; and the ways on. Shares only, never amounts.
          * @summary Reads the caller\'s plan usage: for each class of model the plan includes, the share used of the billing period and of its short window, who pays for the next request, and whether the caller is in limited mode and why; the share used of the session and the day; and the ways on.
@@ -14118,6 +14357,16 @@ const AiApiFactory = function (configuration, basePath, axios) {
         aiMCPTools(requestParameters = {}, options) {
             return localVarFp.aiMCPTools(requestParameters.names, options).then((request) => request(axios, basePath));
         },
+        /**
+         * Searches the model catalog — the one GET /v1/models lists — by text, class, family or capability, and answers each match with its class, family, list price (per token and per million; variable for a router billed at the answering model\'s price), context window and capabilities.
+         * @summary Searches the model catalog — the one GET /v1/models lists — by text, class, family or capability, and answers each match with its class, family, list price (per token and per million; variable for a router billed at the answering model\'s price), context window and capabilities.
+         * @param {AiApiAiModelsRequest} requestParameters Request parameters.
+         * @param {*} [options] Override http request option.
+         * @throws {RequiredError}
+         */
+        aiModels(requestParameters = {}, options) {
+            return localVarFp.aiModels(requestParameters.q, requestParameters._class, requestParameters.family, requestParameters.capability, requestParameters.limit, options).then((request) => request(axios, basePath));
+        },
         /**
          * Sets the payer\'s choice to keep using a model on credits once the plan\'s included usage of it is spent. A pooled org wallet is its org admin\'s to set; a person\'s own wallet is theirs. Every change is on the audit trail, before and after; no trail, no change. Answers the limits as they read after it.
          * @summary Sets the payer\'s choice to keep using a model on credits once the plan\'s included usage of it is spent.
@@ -15968,11 +16217,12 @@ const AiApiFactory = function (configuration, basePath, axios) {
         /**
          * Attaches a per-request outcome reward to the routing decision that served request_id — the enso training loop\'s quality signal. Org-scoped via the same session-OR-Bearer principal the usage read uses (RequirePrincipal): the reward lands only on the caller\'s OWN org\'s event, so a request_id from another org (or unknown) is a 404 — cross-org writes are impossible and unknown ids are indistinguishable from foreign ones. Idempotent: a repeat overwrites. The body carries NO prompt text — only {request_id, reward|rating}.
          * @summary Attaches a per-request outcome reward to the routing decision that served request_id — the enso training loop\'s quality signal.
+         * @param {AiApiPostAiFeedbackRequest} requestParameters Request parameters.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        postAiFeedback(options) {
-            return localVarFp.postAiFeedback(options).then((request) => request(axios, basePath));
+        postAiFeedback(requestParameters, options) {
+            return localVarFp.postAiFeedback(requestParameters.aiRoutingRewardRequest, options).then((request) => request(axios, basePath));
         },
         /**
          * Create one file.
@@ -16597,7 +16847,7 @@ const AiApiFactory = function (configuration, basePath, axios) {
          * @throws {RequiredError}
          */
         postChat(requestParameters, options) {
-            return localVarFp.postChat(requestParameters.openaiChatCompletionRequest, options).then((request) => request(axios, basePath));
+            return localVarFp.postChat(requestParameters.openaiChatCompletionRequest, requestParameters.xMaxCost, requestParameters.xMaxLatencyMs, options).then((request) => request(axios, basePath));
         },
         /**
          * Implements the OpenAI-compatible chat completions API
@@ -16607,7 +16857,7 @@ const AiApiFactory = function (configuration, basePath, axios) {
          * @throws {RequiredError}
          */
         postChatCompletions(requestParameters, options) {
-            return localVarFp.postChatCompletions(requestParameters.openaiChatCompletionRequest, options).then((request) => request(axios, basePath));
+            return localVarFp.postChatCompletions(requestParameters.openaiChatCompletionRequest, requestParameters.xMaxCost, requestParameters.xMaxLatencyMs, options).then((request) => request(axios, basePath));
         },
         /**
          * Serves one completion to a caller with no account.
@@ -16627,7 +16877,7 @@ const AiApiFactory = function (configuration, basePath, axios) {
          * @throws {RequiredError}
          */
         postCompletions(requestParameters, options) {
-            return localVarFp.postCompletions(requestParameters.openaiChatCompletionRequest, options).then((request) => request(axios, basePath));
+            return localVarFp.postCompletions(requestParameters.openaiChatCompletionRequest, requestParameters.xMaxCost, requestParameters.xMaxLatencyMs, options).then((request) => request(axios, basePath));
         },
         /**
          * Implements POST /v1/decisions (the Decisions API).  Body: {\"model\": \"kai\", \"state\": \"...\"|{...}|[...], \"questions\": {\"<name>\": {\"type\": \"choice\"|\"noul\"|\"score\", \"instructions\": ..., \"criteria\": ...}}}. model is kai, Kai\'s versioned id kai-<12 hex of the weights\' sha256> — priced as kai and sent as asked — or Jev by OpenRouter\'s vendor ids, typesafe/jev-1.13 and ~typesafe/jev-latest, which reach Jev itself and bill at Jev\'s list price. No Jev id is ever answered by Kai: a bare one, such as jev-latest, is an unknown model. model is required; state and questions are required unless the request names a handle, which carries neither. instructions is optional and any JSON. A choice names at least 2 labels and a score at least 1 level, bounded by the token budget rather than a count; questions holds 1 to 100.  observe holds the state under an id, and a later request naming that id as its handle decides over it again. An id is 1 to 128 characters of A-Z, a-z, 0-9, \'.\', \'_\' and \'-\'. A handle belongs to the org that observed it: no other org\'s request can name it.  Response: {\"id\",\"model\",\"provider\",\"answers\":{\"<name>\":{\"type\",...}}, \"usage\":{\"input_tokens\",\"output_tokens\"},\"routing\",\"state_hash\",\"latency_ms\"}. usage.input_tokens is the billed count — the request\'s text counted once, the state once and each question\'s instructions and options once; a decision over a handle bills the state once, when it was observed.  The body may be sent gzip, deflate, br or zstd encoded; decoded, it is bounded as sent.  Refusals are {\"error\":{\"code\",\"message\"}}: 400 malformed JSON or unknown model, 401 no valid credential, 402 insufficient balance, 403 a key kind that may not call this (pk-), 415 any other Content-Encoding, 422 an invalid question or handle id, a state beyond the checkpoint\'s reach (code state_too_long) or a body past 16 MiB (code request_too_long), 429 rate limited or queue full, 502 the service failed, 503 the model is known and not served, 529 overloaded. 402, 429 and 529 carry Retry-After and Retry-After-Ms, and every answer carries X-Request-Id. Billed on the answer\'s input tokens at the model\'s price.
@@ -16699,11 +16949,12 @@ const AiApiFactory = function (configuration, basePath, axios) {
         /**
          * Implements POST /v1/responses. The converted request is completed by the chat path, which is handed a sink saying where the answer goes: a stream is translated as it is produced, a whole body is translated entire.
          * @summary Implements POST /v1/responses.
+         * @param {AiApiPostResponsesRequest} requestParameters Request parameters.
          * @param {*} [options] Override http request option.
          * @throws {RequiredError}
          */
-        postResponses(options) {
-            return localVarFp.postResponses(options).then((request) => request(axios, basePath));
+        postResponses(requestParameters = {}, options) {
+            return localVarFp.postResponses(requestParameters.xMaxCost, requestParameters.xMaxLatencyMs, options).then((request) => request(axios, basePath));
         },
         /**
          * Implements POST /v1/videos/generations — the ASYNC create.  Body: {\"model\": \"...\", \"prompt\": \"...\", \"size\"?: \"1280x720\", \"seconds\"?: int}  It authenticates the caller, resolves the model to its upstream provider via the shared routing table (zen3-video* / wan2-2-t2v-a14b → the spark-video backend), reserves the per-video budget (the balance gate), creates ONE upstream job, registers it in the in-pod store, and returns the OpenAI-shaped video object with status \"queued\" IMMEDIATELY. The client then polls GET /v1/videos/{id} and downloads GET /v1/videos/{id}/content. Nothing is billed here — the debit lands on completion.
@@ -17047,6 +17298,28 @@ exports.AiApiFactory = AiApiFactory;
  * @extends {BaseAPI}
  */
 class AiApi extends base_1.BaseAPI {
+    /**
+     * Asks one model one prompt and answers the reply, which model served it and who paid. The call is the caller\'s own POST /v1/chat/completions, run as the caller, so it is gated, billed and refused exactly as that route is; a refusal names its code and what lifts it, never an amount. An empty model asks the deployment\'s default Hanzo model.
+     * @summary Asks one model one prompt and answers the reply, which model served it and who paid.
+     * @param {AiApiAiChatRequest} requestParameters Request parameters.
+     * @param {*} [options] Override http request option.
+     * @throws {RequiredError}
+     * @memberof AiApi
+     */
+    aiChat(requestParameters, options) {
+        return (0, exports.AiApiFp)(this.configuration).aiChat(requestParameters.aiChatIn, options).then((request) => request(this.axios, this.basePath));
+    }
+    /**
+     * Runs one decision — the caller\'s own POST /v1/decisions, gated and billed exactly as that route is — and answers the decision, which model served it and who paid. A refusal names its code and what lifts it, never an amount.
+     * @summary Runs one decision — the caller\'s own POST /v1/decisions, gated and billed exactly as that route is — and answers the decision, which model served it and who paid.
+     * @param {AiApiAiDecideRequest} requestParameters Request parameters.
+     * @param {*} [options] Override http request option.
+     * @throws {RequiredError}
+     * @memberof AiApi
+     */
+    aiDecide(requestParameters, options) {
+        return (0, exports.AiApiFp)(this.configuration).aiDecide(requestParameters.aiDecideIn, options).then((request) => request(this.axios, this.basePath));
+    }
     /**
      * Reads the caller\'s plan usage: for each class of model the plan includes, the share used of the billing period and of its short window, who pays for the next request, and whether the caller is in limited mode and why; the share used of the session and the day; and the ways on. Shares only, never amounts.
      * @summary Reads the caller\'s plan usage: for each class of model the plan includes, the share used of the billing period and of its short window, who pays for the next request, and whether the caller is in limited mode and why; the share used of the session and the day; and the ways on.
@@ -17068,6 +17341,17 @@ class AiApi extends base_1.BaseAPI {
     aiMCPTools(requestParameters = {}, options) {
         return (0, exports.AiApiFp)(this.configuration).aiMCPTools(requestParameters.names, options).then((request) => request(this.axios, this.basePath));
     }
+    /**
+     * Searches the model catalog — the one GET /v1/models lists — by text, class, family or capability, and answers each match with its class, family, list price (per token and per million; variable for a router billed at the answering model\'s price), context window and capabilities.
+     * @summary Searches the model catalog — the one GET /v1/models lists — by text, class, family or capability, and answers each match with its class, family, list price (per token and per million; variable for a router billed at the answering model\'s price), context window and capabilities.
+     * @param {AiApiAiModelsRequest} requestParameters Request parameters.
+     * @param {*} [options] Override http request option.
+     * @throws {RequiredError}
+     * @memberof AiApi
+     */
+    aiModels(requestParameters = {}, options) {
+        return (0, exports.AiApiFp)(this.configuration).aiModels(requestParameters.q, requestParameters._class, requestParameters.family, requestParameters.capability, requestParameters.limit, options).then((request) => request(this.axios, this.basePath));
+    }
     /**
      * Sets the payer\'s choice to keep using a model on credits once the plan\'s included usage of it is spent. A pooled org wallet is its org admin\'s to set; a person\'s own wallet is theirs. Every change is on the audit trail, before and after; no trail, no change. Answers the limits as they read after it.
      * @summary Sets the payer\'s choice to keep using a model on credits once the plan\'s included usage of it is spent.
@@ -19113,12 +19397,13 @@ class AiApi extends base_1.BaseAPI {
     /**
      * Attaches a per-request outcome reward to the routing decision that served request_id — the enso training loop\'s quality signal. Org-scoped via the same session-OR-Bearer principal the usage read uses (RequirePrincipal): the reward lands only on the caller\'s OWN org\'s event, so a request_id from another org (or unknown) is a 404 — cross-org writes are impossible and unknown ids are indistinguishable from foreign ones. Idempotent: a repeat overwrites. The body carries NO prompt text — only {request_id, reward|rating}.
      * @summary Attaches a per-request outcome reward to the routing decision that served request_id — the enso training loop\'s quality signal.
+     * @param {AiApiPostAiFeedbackRequest} requestParameters Request parameters.
      * @param {*} [options] Override http request option.
      * @throws {RequiredError}
      * @memberof AiApi
      */
-    postAiFeedback(options) {
-        return (0, exports.AiApiFp)(this.configuration).postAiFeedback(options).then((request) => request(this.axios, this.basePath));
+    postAiFeedback(requestParameters, options) {
+        return (0, exports.AiApiFp)(this.configuration).postAiFeedback(requestParameters.aiRoutingRewardRequest, options).then((request) => request(this.axios, this.basePath));
     }
     /**
      * Create one file.
@@ -19808,7 +20093,7 @@ class AiApi extends base_1.BaseAPI {
      * @memberof AiApi
      */
     postChat(requestParameters, options) {
-        return (0, exports.AiApiFp)(this.configuration).postChat(requestParameters.openaiChatCompletionRequest, options).then((request) => request(this.axios, this.basePath));
+        return (0, exports.AiApiFp)(this.configuration).postChat(requestParameters.openaiChatCompletionRequest, requestParameters.xMaxCost, requestParameters.xMaxLatencyMs, options).then((request) => request(this.axios, this.basePath));
     }
     /**
      * Implements the OpenAI-compatible chat completions API
@@ -19819,7 +20104,7 @@ class AiApi extends base_1.BaseAPI {
      * @memberof AiApi
      */
     postChatCompletions(requestParameters, options) {
-        return (0, exports.AiApiFp)(this.configuration).postChatCompletions(requestParameters.openaiChatCompletionRequest, options).then((request) => request(this.axios, this.basePath));
+        return (0, exports.AiApiFp)(this.configuration).postChatCompletions(requestParameters.openaiChatCompletionRequest, requestParameters.xMaxCost, requestParameters.xMaxLatencyMs, options).then((request) => request(this.axios, this.basePath));
     }
     /**
      * Serves one completion to a caller with no account.
@@ -19841,7 +20126,7 @@ class AiApi extends base_1.BaseAPI {
      * @memberof AiApi
      */
     postCompletions(requestParameters, options) {
-        return (0, exports.AiApiFp)(this.configuration).postCompletions(requestParameters.openaiChatCompletionRequest, options).then((request) => request(this.axios, this.basePath));
+        return (0, exports.AiApiFp)(this.configuration).postCompletions(requestParameters.openaiChatCompletionRequest, requestParameters.xMaxCost, requestParameters.xMaxLatencyMs, options).then((request) => request(this.axios, this.basePath));
     }
     /**
      * Implements POST /v1/decisions (the Decisions API).  Body: {\"model\": \"kai\", \"state\": \"...\"|{...}|[...], \"questions\": {\"<name>\": {\"type\": \"choice\"|\"noul\"|\"score\", \"instructions\": ..., \"criteria\": ...}}}. model is kai, Kai\'s versioned id kai-<12 hex of the weights\' sha256> — priced as kai and sent as asked — or Jev by OpenRouter\'s vendor ids, typesafe/jev-1.13 and ~typesafe/jev-latest, which reach Jev itself and bill at Jev\'s list price. No Jev id is ever answered by Kai: a bare one, such as jev-latest, is an unknown model. model is required; state and questions are required unless the request names a handle, which carries neither. instructions is optional and any JSON. A choice names at least 2 labels and a score at least 1 level, bounded by the token budget rather than a count; questions holds 1 to 100.  observe holds the state under an id, and a later request naming that id as its handle decides over it again. An id is 1 to 128 characters of A-Z, a-z, 0-9, \'.\', \'_\' and \'-\'. A handle belongs to the org that observed it: no other org\'s request can name it.  Response: {\"id\",\"model\",\"provider\",\"answers\":{\"<name>\":{\"type\",...}}, \"usage\":{\"input_tokens\",\"output_tokens\"},\"routing\",\"state_hash\",\"latency_ms\"}. usage.input_tokens is the billed count — the request\'s text counted once, the state once and each question\'s instructions and options once; a decision over a handle bills the state once, when it was observed.  The body may be sent gzip, deflate, br or zstd encoded; decoded, it is bounded as sent.  Refusals are {\"error\":{\"code\",\"message\"}}: 400 malformed JSON or unknown model, 401 no valid credential, 402 insufficient balance, 403 a key kind that may not call this (pk-), 415 any other Content-Encoding, 422 an invalid question or handle id, a state beyond the checkpoint\'s reach (code state_too_long) or a body past 16 MiB (code request_too_long), 429 rate limited or queue full, 502 the service failed, 503 the model is known and not served, 529 overloaded. 402, 429 and 529 carry Retry-After and Retry-After-Ms, and every answer carries X-Request-Id. Billed on the answer\'s input tokens at the model\'s price.
@@ -19920,12 +20205,13 @@ class AiApi extends base_1.BaseAPI {
     /**
      * Implements POST /v1/responses. The converted request is completed by the chat path, which is handed a sink saying where the answer goes: a stream is translated as it is produced, a whole body is translated entire.
      * @summary Implements POST /v1/responses.
+     * @param {AiApiPostResponsesRequest} requestParameters Request parameters.
      * @param {*} [options] Override http request option.
      * @throws {RequiredError}
      * @memberof AiApi
      */
-    postResponses(options) {
-        return (0, exports.AiApiFp)(this.configuration).postResponses(options).then((request) => request(this.axios, this.basePath));
+    postResponses(requestParameters = {}, options) {
+        return (0, exports.AiApiFp)(this.configuration).postResponses(requestParameters.xMaxCost, requestParameters.xMaxLatencyMs, options).then((request) => request(this.axios, this.basePath));
     }
     /**
      * Implements POST /v1/videos/generations — the ASYNC create.  Body: {\"model\": \"...\", \"prompt\": \"...\", \"size\"?: \"1280x720\", \"seconds\"?: int}  It authenticates the caller, resolves the model to its upstream provider via the shared routing table (zen3-video* / wan2-2-t2v-a14b → the spark-video backend), reserves the per-video budget (the balance gate), creates ONE upstream job, registers it in the in-pod store, and returns the OpenAI-shaped video object with status \"queued\" IMMEDIATELY. The client then polls GET /v1/videos/{id} and downloads GET /v1/videos/{id}/content. Nothing is billed here — the debit lands on completion.