@openrouter/sdk 1.1.17 → 1.1.18

package/esm/funcs/benchmarksGetBenchmarks.d.ts

@@ -13,7 +13,7 @@ import { Result } from "../types/fp.js";
  * List Benchmarks
  *
  * @remarks
- * Unified benchmark endpoint that aggregates scores from multiple benchmark sources (Artificial Analysis, Design Arena). Filter by source to reproduce the exact shapes from the legacy per-source endpoints, or use task_type to find models suited for specific workloads. Authenticate with any valid OpenRouter API key. Rate-limited to 30 requests/minute per key and 500 requests/day per account.
+ * Unified benchmark endpoint that aggregates scores from multiple benchmark sources (Artificial Analysis, Design Arena, and OpenRouter's own tau-bench and GPQA evals). Filter by source to reproduce the exact shapes from the legacy per-source endpoints, or use task_type to find models suited for specific workloads. Authenticate with any valid OpenRouter API key. Rate-limited to 30 requests/minute per key and 500 requests/day per account.
  */
 export declare function benchmarksGetBenchmarks(client: OpenRouterCore, request?: operations.GetBenchmarksRequest | undefined, options?: RequestOptions): APIPromise<Result<models.UnifiedBenchmarksResponse, errors.BadRequestResponseError | errors.UnauthorizedResponseError | errors.TooManyRequestsResponseError | errors.InternalServerResponseError | OpenRouterError | ResponseValidationError | ConnectionError | RequestAbortedError | RequestTimeoutError | InvalidRequestError | UnexpectedClientError | SDKValidationError>>;
 //# sourceMappingURL=benchmarksGetBenchmarks.d.ts.map

package/esm/funcs/benchmarksGetBenchmarks.js

@@ -17,7 +17,7 @@ import { APIPromise } from "../types/async.js";
  * List Benchmarks
  *
  * @remarks
- * Unified benchmark endpoint that aggregates scores from multiple benchmark sources (Artificial Analysis, Design Arena). Filter by source to reproduce the exact shapes from the legacy per-source endpoints, or use task_type to find models suited for specific workloads. Authenticate with any valid OpenRouter API key. Rate-limited to 30 requests/minute per key and 500 requests/day per account.
+ * Unified benchmark endpoint that aggregates scores from multiple benchmark sources (Artificial Analysis, Design Arena, and OpenRouter's own tau-bench and GPQA evals). Filter by source to reproduce the exact shapes from the legacy per-source endpoints, or use task_type to find models suited for specific workloads. Authenticate with any valid OpenRouter API key. Rate-limited to 30 requests/minute per key and 500 requests/day per account.
  */
 export function benchmarksGetBenchmarks(client, request, options) {
     return new APIPromise($do(client, request, options));

package/esm/lib/config.d.ts

@@ -49,8 +49,8 @@ export declare function serverURLFromOptions(options: SDKOptions): URL | null;
 export declare const SDK_METADATA: {
     readonly language: "typescript";
     readonly openapiDocVersion: "1.0.0";
-    readonly sdkVersion: "1.1.17";
+    readonly sdkVersion: "1.1.18";
     readonly genVersion: "2.914.0";
-    readonly userAgent: "speakeasy-sdk/typescript 1.1.17 2.914.0 1.0.0 @openrouter/sdk";
+    readonly userAgent: "speakeasy-sdk/typescript 1.1.18 2.914.0 1.0.0 @openrouter/sdk";
 };
 //# sourceMappingURL=config.d.ts.map

package/esm/lib/config.js

@@ -26,8 +26,8 @@ export function serverURLFromOptions(options) {
 export const SDK_METADATA = {
     language: "typescript",
     openapiDocVersion: "1.0.0",
-    sdkVersion: "1.1.17",
+    sdkVersion: "1.1.18",
     genVersion: "2.914.0",
-    userAgent: "speakeasy-sdk/typescript 1.1.17 2.914.0 1.0.0 @openrouter/sdk",
+    userAgent: "speakeasy-sdk/typescript 1.1.18 2.914.0 1.0.0 @openrouter/sdk",
 };
 //# sourceMappingURL=config.js.map

package/esm/models/index.d.ts

@@ -523,6 +523,7 @@ export * from "./unifiedbenchmarkpricing.js";
 export * from "./unifiedbenchmarksaaitem.js";
 export * from "./unifiedbenchmarksdaitem.js";
 export * from "./unifiedbenchmarksmeta.js";
+export * from "./unifiedbenchmarksoritem.js";
 export * from "./unifiedbenchmarksresponse.js";
 export * from "./unprocessableentityresponseerrordata.js";
 export * from "./updatebyokkeyrequest.js";

package/esm/models/index.js

@@ -527,6 +527,7 @@ export * from "./unifiedbenchmarkpricing.js";
 export * from "./unifiedbenchmarksaaitem.js";
 export * from "./unifiedbenchmarksdaitem.js";
 export * from "./unifiedbenchmarksmeta.js";
+export * from "./unifiedbenchmarksoritem.js";
 export * from "./unifiedbenchmarksresponse.js";
 export * from "./unprocessableentityresponseerrordata.js";
 export * from "./updatebyokkeyrequest.js";

package/esm/models/operations/getbenchmarks.d.ts

@@ -27,6 +27,7 @@ export type GetBenchmarksGlobals = {
 export declare const Source: {
     readonly ArtificialAnalysis: "artificial-analysis";
     readonly DesignArena: "design-arena";
+    readonly Openrouter: "openrouter";
 };
 /**
  * Benchmark source to query. Determines the shape of the returned items. When omitted, returns results from all sources.

package/esm/models/operations/getbenchmarks.js

@@ -11,6 +11,7 @@ import * as openEnums from "../../types/enums.js";
 export const Source = {
     ArtificialAnalysis: "artificial-analysis",
     DesignArena: "design-arena",
+    Openrouter: "openrouter",
 };
 /**
  * Filter results by task type. For Artificial Analysis, maps to the corresponding index. For Design Arena, maps to the matching category.

package/esm/models/unifiedbenchmarksmeta.d.ts

@@ -8,6 +8,7 @@ import { SDKValidationError } from "./errors/sdkvalidationerror.js";
 export declare const SourceEnum: {
     readonly ArtificialAnalysis: "artificial-analysis";
     readonly DesignArena: "design-arena";
+    readonly Openrouter: "openrouter";
 };
 /**
  * The source filter applied, or null when all sources are returned.

package/esm/models/unifiedbenchmarksmeta.js

@@ -12,6 +12,7 @@ import * as openEnums from "../types/enums.js";
 export const SourceEnum = {
     ArtificialAnalysis: "artificial-analysis",
     DesignArena: "design-arena",
+    Openrouter: "openrouter",
 };
 /**
  * Dataset version.

package/esm/models/unifiedbenchmarksoritem.d.ts

@@ -0,0 +1,59 @@
+import * as z from "zod/v4";
+import { OpenEnum } from "../types/enums.js";
+import { Result as SafeParseResult } from "../types/fp.js";
+import { SDKValidationError } from "./errors/sdkvalidationerror.js";
+/**
+ * OpenRouter benchmark evaluation type.
+ */
+export declare const BenchmarkType: {
+    readonly GpqaDiamond: "gpqa_diamond";
+    readonly TauBenchVerifiedAirline: "tau_bench_verified_airline";
+};
+/**
+ * OpenRouter benchmark evaluation type.
+ */
+export type BenchmarkType = OpenEnum<typeof BenchmarkType>;
+export type UnifiedBenchmarksORItem = {
+    /**
+     * Aggregate accuracy score from 0 to 1. Higher is better.
+     */
+    accuracy: number;
+    /**
+     * Standard deviation of run accuracy, or null for a single run.
+     */
+    accuracyStddev: number | null;
+    /**
+     * Average cost per task in USD, or null if unavailable.
+     */
+    avgCostPerTask: number | null;
+    /**
+     * OpenRouter benchmark evaluation type.
+     */
+    benchmarkType: BenchmarkType;
+    /**
+     * Human-readable model name.
+     */
+    displayName: string;
+    /**
+     * Timestamp of the most recent public benchmark run.
+     */
+    lastRunTimestamp: string;
+    /**
+     * Stable OpenRouter model identifier.
+     */
+    modelPermaslug: string;
+    /**
+     * Benchmark source discriminator.
+     */
+    source: "openrouter";
+    /**
+     * Total benchmark tasks across runs.
+     */
+    totalTasks: number;
+};
+/** @internal */
+export declare const BenchmarkType$inboundSchema: z.ZodType<BenchmarkType, unknown>;
+/** @internal */
+export declare const UnifiedBenchmarksORItem$inboundSchema: z.ZodType<UnifiedBenchmarksORItem, unknown>;
+export declare function unifiedBenchmarksORItemFromJSON(jsonString: string): SafeParseResult<UnifiedBenchmarksORItem, SDKValidationError>;
+//# sourceMappingURL=unifiedbenchmarksoritem.d.ts.map

package/esm/models/unifiedbenchmarksoritem.js

@@ -0,0 +1,43 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ * @generated-id: be2fd0591b6b
+ */
+import * as z from "zod/v4";
+import { remap as remap$ } from "../lib/primitives.js";
+import { safeParse } from "../lib/schemas.js";
+import * as openEnums from "../types/enums.js";
+/**
+ * OpenRouter benchmark evaluation type.
+ */
+export const BenchmarkType = {
+    GpqaDiamond: "gpqa_diamond",
+    TauBenchVerifiedAirline: "tau_bench_verified_airline",
+};
+/** @internal */
+export const BenchmarkType$inboundSchema = openEnums.inboundSchema(BenchmarkType);
+/** @internal */
+export const UnifiedBenchmarksORItem$inboundSchema = z.object({
+    accuracy: z.number(),
+    accuracy_stddev: z.nullable(z.number()),
+    avg_cost_per_task: z.nullable(z.number()),
+    benchmark_type: BenchmarkType$inboundSchema,
+    display_name: z.string(),
+    last_run_timestamp: z.string(),
+    model_permaslug: z.string(),
+    source: z.literal("openrouter"),
+    total_tasks: z.int(),
+}).transform((v) => {
+    return remap$(v, {
+        "accuracy_stddev": "accuracyStddev",
+        "avg_cost_per_task": "avgCostPerTask",
+        "benchmark_type": "benchmarkType",
+        "display_name": "displayName",
+        "last_run_timestamp": "lastRunTimestamp",
+        "model_permaslug": "modelPermaslug",
+        "total_tasks": "totalTasks",
+    });
+});
+export function unifiedBenchmarksORItemFromJSON(jsonString) {
+    return safeParse(jsonString, (x) => UnifiedBenchmarksORItem$inboundSchema.parse(JSON.parse(x)), `Failed to parse 'UnifiedBenchmarksORItem' from JSON`);
+}
+//# sourceMappingURL=unifiedbenchmarksoritem.js.map

package/esm/models/unifiedbenchmarksresponse.d.ts

@@ -5,9 +5,10 @@ import { SDKValidationError } from "./errors/sdkvalidationerror.js";
 import { UnifiedBenchmarksAAItem } from "./unifiedbenchmarksaaitem.js";
 import { UnifiedBenchmarksDAItem } from "./unifiedbenchmarksdaitem.js";
 import { UnifiedBenchmarksMeta } from "./unifiedbenchmarksmeta.js";
-export type UnifiedBenchmarksResponseData = UnifiedBenchmarksAAItem | UnifiedBenchmarksDAItem | discriminatedUnionTypes.Unknown<"source">;
+import { UnifiedBenchmarksORItem } from "./unifiedbenchmarksoritem.js";
+export type UnifiedBenchmarksResponseData = UnifiedBenchmarksAAItem | UnifiedBenchmarksDAItem | UnifiedBenchmarksORItem | discriminatedUnionTypes.Unknown<"source">;
 export type UnifiedBenchmarksResponse = {
-    data: Array<UnifiedBenchmarksAAItem | UnifiedBenchmarksDAItem | discriminatedUnionTypes.Unknown<"source">>;
+    data: Array<UnifiedBenchmarksAAItem | UnifiedBenchmarksDAItem | UnifiedBenchmarksORItem | discriminatedUnionTypes.Unknown<"source">>;
     meta: UnifiedBenchmarksMeta;
 };
 /** @internal */

package/esm/models/unifiedbenchmarksresponse.js

@@ -8,10 +8,12 @@ import { discriminatedUnion } from "../types/discriminatedUnion.js";
 import { UnifiedBenchmarksAAItem$inboundSchema, } from "./unifiedbenchmarksaaitem.js";
 import { UnifiedBenchmarksDAItem$inboundSchema, } from "./unifiedbenchmarksdaitem.js";
 import { UnifiedBenchmarksMeta$inboundSchema, } from "./unifiedbenchmarksmeta.js";
+import { UnifiedBenchmarksORItem$inboundSchema, } from "./unifiedbenchmarksoritem.js";
 /** @internal */
 export const UnifiedBenchmarksResponseData$inboundSchema = discriminatedUnion("source", {
     ["artificial-analysis"]: UnifiedBenchmarksAAItem$inboundSchema,
     ["design-arena"]: UnifiedBenchmarksDAItem$inboundSchema,
+    openrouter: UnifiedBenchmarksORItem$inboundSchema,
 });
 export function unifiedBenchmarksResponseDataFromJSON(jsonString) {
     return safeParse(jsonString, (x) => UnifiedBenchmarksResponseData$inboundSchema.parse(JSON.parse(x)), `Failed to parse 'UnifiedBenchmarksResponseData' from JSON`);
@@ -21,6 +23,7 @@ export const UnifiedBenchmarksResponse$inboundSchema = z.object({
     data: z.array(discriminatedUnion("source", {
         ["artificial-analysis"]: UnifiedBenchmarksAAItem$inboundSchema,
         ["design-arena"]: UnifiedBenchmarksDAItem$inboundSchema,
+        openrouter: UnifiedBenchmarksORItem$inboundSchema,
     })),
     meta: UnifiedBenchmarksMeta$inboundSchema,
 });

package/esm/sdk/benchmarks.d.ts

@@ -6,7 +6,7 @@ export declare class Benchmarks extends ClientSDK {
      * List Benchmarks
      *
      * @remarks
-     * Unified benchmark endpoint that aggregates scores from multiple benchmark sources (Artificial Analysis, Design Arena). Filter by source to reproduce the exact shapes from the legacy per-source endpoints, or use task_type to find models suited for specific workloads. Authenticate with any valid OpenRouter API key. Rate-limited to 30 requests/minute per key and 500 requests/day per account.
+     * Unified benchmark endpoint that aggregates scores from multiple benchmark sources (Artificial Analysis, Design Arena, and OpenRouter's own tau-bench and GPQA evals). Filter by source to reproduce the exact shapes from the legacy per-source endpoints, or use task_type to find models suited for specific workloads. Authenticate with any valid OpenRouter API key. Rate-limited to 30 requests/minute per key and 500 requests/day per account.
      */
     getBenchmarks(request?: operations.GetBenchmarksRequest | undefined, options?: RequestOptions): Promise<models.UnifiedBenchmarksResponse>;
 }

package/esm/sdk/benchmarks.js

@@ -10,7 +10,7 @@ export class Benchmarks extends ClientSDK {
      * List Benchmarks
      *
      * @remarks
-     * Unified benchmark endpoint that aggregates scores from multiple benchmark sources (Artificial Analysis, Design Arena). Filter by source to reproduce the exact shapes from the legacy per-source endpoints, or use task_type to find models suited for specific workloads. Authenticate with any valid OpenRouter API key. Rate-limited to 30 requests/minute per key and 500 requests/day per account.
+     * Unified benchmark endpoint that aggregates scores from multiple benchmark sources (Artificial Analysis, Design Arena, and OpenRouter's own tau-bench and GPQA evals). Filter by source to reproduce the exact shapes from the legacy per-source endpoints, or use task_type to find models suited for specific workloads. Authenticate with any valid OpenRouter API key. Rate-limited to 30 requests/minute per key and 500 requests/day per account.
      */
     async getBenchmarks(request, options) {
         return unwrapAsync(benchmarksGetBenchmarks(this, request, options));

package/jsr.json

@@ -2,7 +2,7 @@
 
 {
   "name": "@openrouter/sdk",
-  "version": "1.1.17",
+  "version": "1.1.18",
   "exports": {
     ".": "./src/index.ts",    
     "./models/errors": "./src/models/errors/index.ts",    

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@openrouter/sdk",
-  "version": "1.1.17",
+  "version": "1.1.18",
   "author": "OpenRouter",
   "description": "The OpenRouter TypeScript SDK is a type-safe toolkit for building AI applications with access to 400+ language models through a unified API.",
   "keywords": [
@@ -73,15 +73,15 @@
     "lint": "eslint --cache --max-warnings=0 src",
     "build": "tsc",
     "prepublishOnly": "npm run build",
-    "postinstall": "node scripts/check-types.js || true",
-    "prepare": "npm run build",
-    "test": "vitest --run --project unit",
     "test:e2e": "vitest --run --project e2e",
-    "test:transit": "exit 0",
     "test:watch": "vitest --watch --project unit",
     "compile": "tsc",
+    "postinstall": "node scripts/check-types.js || true",
+    "test:transit": "exit 0",
     "typecheck": "tsc --noEmit",
-    "typecheck:transit": "exit 0"
+    "typecheck:transit": "exit 0",
+    "prepare": "npm run build",
+    "test": "vitest --run --project unit"
   },
   "peerDependencies": {},
   "devDependencies": {