{ "openapi": "3.1.0", "info": { "title": "Openbenchmarks for Agents Public API", "version": "0.2.0", "description": "Read-only public API serving the same benchmark data the website renders. No authentication required. CORS is open.", "license": { "name": "CC-BY-4.0", "url": "https://creativecommons.org/licenses/by/4.0/" }, "contact": { "name": "Openbenchmarks", "url": "https://openbenchmarks.com" } }, "servers": [ { "url": "https://openbenchmarks.com" } ], "externalDocs": { "description": "Openbenchmarks for Agents documentation, methodology, and metric definitions.", "url": "https://openbenchmarks.com" }, "tags": [ { "name": "benchmarks", "description": "Read-only endpoints exposing benchmark data. Identical to what the website renders.", "externalDocs": { "description": "Methodology and metric definitions.", "url": "https://openbenchmarks.com/technographics#methodology" } } ], "x-mcp-servers": [ { "name": "openbenchmarks", "url": "https://mcp.openbenchmarks.com/mcp", "transport": "http", "protocol_version": "2025-03-26", "description": "Openbenchmarks MCP server - connect Claude, ChatGPT, Cursor and other MCP clients to the public benchmark data via tools, prompts, and resources. OAuth 2.1 with dynamic client registration.", "documentation": "https://openbenchmarks.com/llms.txt", "capabilities": { "tools": true, "prompts": true, "resources": true, "tasks": true }, "auth": { "type": "oauth2", "flow": "authorization_code", "pkce": "S256", "scopes": [ "mcp" ], "dynamic_registration": true, "authorization_endpoint": "https://mcp.openbenchmarks.com/oauth/authorize", "token_endpoint": "https://mcp.openbenchmarks.com/oauth/token", "registration_endpoint": "https://mcp.openbenchmarks.com/oauth/register", "metadata_url": "https://mcp.openbenchmarks.com/.well-known/oauth-authorization-server" } } ], "paths": { "/api/benchmarks": { "get": { "tags": [ "benchmarks" ], "summary": "List all benchmarks", "description": "Returns the index of all published benchmarks with summary winners and links to each detail endpoint. Safe to call without authentication.", "operationId": "listBenchmarks", "responses": { "200": { "description": "Index of benchmarks.", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/BenchmarkIndex" }, "example": { "site": { "name": "Openbenchmarks for Agents", "url": "https://openbenchmarks.com", "description": "Open benchmarks for Agents - verified benchmarks for build vs buy decisions on AI agent tooling." }, "last_updated": "2026-06-04T00:00:00.000Z", "benchmarks": [ { "slug": "technographics", "name": "Technographics Benchmark", "page_url": "https://openbenchmarks.com/technographics", "api_url": "https://openbenchmarks.com/api/benchmarks/technographics", "provider_count": 5, "canonical_categories": [ "engineering", "data", "sales", "marketing", "finance", "hr", "support", "ops" ], "winners": { "highest_category_coverage": { "provider": "OpenFunnel", "category_coverage_pct": 100, "categories_covered": 8 } } }, { "slug": "lookalikes", "name": "Lookalike Benchmark", "page_url": "https://openbenchmarks.com/lookalikes", "api_url": "https://openbenchmarks.com/api/benchmarks/lookalikes", "provider_count": 5 } ] } } } } } } }, "/api/benchmarks/{slug}": { "get": { "tags": [ "benchmarks" ], "summary": "Get a benchmark by slug", "description": "Returns full per-provider results for a benchmark. Safe to call without authentication.", "operationId": "getBenchmark", "parameters": [ { "name": "slug", "in": "path", "required": true, "schema": { "type": "string", "enum": [ "technographics", "lookalikes" ] }, "description": "Benchmark slug. Published benchmarks: 'technographics' and 'lookalikes'.", "example": "technographics" } ], "responses": { "200": { "description": "Full benchmark data.", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/Benchmark" }, "example": { "slug": "technographics", "name": "Technographics Benchmark", "last_updated": "2026-06-04T00:00:00.000Z", "license": "https://creativecommons.org/licenses/by/4.0/", "categories": [ "engineering", "data", "sales", "marketing", "finance", "hr", "support", "ops" ], "results": [ { "rank": 1, "provider": "OpenFunnel", "provider_slug": "openfunnel", "companies_observed": 534064, "total_companies": 0, "categories_covered": 8, "canonical_categories": 8, "category_coverage_pct": 100, "distinct_tools": 40, "distinct_company_tool_pairs": 534064 } ], "methodology_url": "https://openbenchmarks.com/technographics#methodology" } } } }, "404": { "description": "Benchmark slug not recognized.", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/Error" }, "example": { "error": "benchmark_not_found", "message": "No benchmark with slug 'unknown'. See /api/benchmarks for the list.", "available": [ "technographics", "lookalikes" ] } } } } } } } }, "components": { "schemas": { "BenchmarkIndex": { "type": "object", "properties": { "site": { "type": "object", "properties": { "name": { "type": "string" }, "url": { "type": "string", "format": "uri" }, "description": { "type": "string" } }, "required": [ "name", "url" ] }, "last_updated": { "type": "string", "format": "date-time" }, "benchmarks": { "type": "array", "items": { "$ref": "#/components/schemas/BenchmarkSummary" } }, "docs": { "type": "object", "properties": { "openapi": { "type": "string", "format": "uri" }, "llms": { "type": "string", "format": "uri" }, "methodology": { "type": "string", "format": "uri" } } }, "mcp": { "type": "object", "description": "Native Model Context Protocol endpoint exposing Openbenchmarks data as tools, prompts, and resources. OAuth 2.1 with dynamic client registration.", "properties": { "name": { "type": "string" }, "url": { "type": "string", "format": "uri" }, "transport": { "type": "string", "enum": [ "http" ] }, "protocol_version": { "type": "string" }, "capabilities": { "type": "object", "properties": { "tools": { "type": "boolean" }, "prompts": { "type": "boolean" }, "resources": { "type": "boolean" }, "tasks": { "type": "boolean" } } }, "auth": { "type": "object", "properties": { "type": { "type": "string", "enum": [ "oauth2" ] }, "scopes": { "type": "array", "items": { "type": "string" } }, "dynamic_registration": { "type": "boolean" }, "metadata_url": { "type": "string", "format": "uri" } } }, "discovery": { "type": "string", "format": "uri" }, "documentation": { "type": "string", "format": "uri" }, "description": { "type": "string" } } } }, "required": [ "site", "last_updated", "benchmarks" ] }, "BenchmarkSummary": { "type": "object", "properties": { "slug": { "type": "string" }, "name": { "type": "string" }, "description": { "type": "string" }, "page_url": { "type": "string", "format": "uri" }, "api_url": { "type": "string", "format": "uri" }, "metrics": { "type": "array", "items": { "$ref": "#/components/schemas/Metric" } }, "provider_count": { "type": "integer", "minimum": 0 }, "canonical_categories": { "type": "array", "items": { "type": "string" } }, "winners": { "type": "object" } }, "required": [ "slug", "name", "page_url", "api_url" ] }, "Benchmark": { "type": "object", "properties": { "slug": { "type": "string" }, "name": { "type": "string" }, "description": { "type": "string" }, "page_url": { "type": "string", "format": "uri" }, "last_updated": { "type": "string", "format": "date-time" }, "license": { "type": "string", "format": "uri" }, "metrics": { "type": "array", "items": { "$ref": "#/components/schemas/Metric" } }, "categories": { "type": "array", "items": { "type": "string" } }, "active_dataset": { "type": "string", "nullable": true }, "results": { "type": "array", "description": "Per-provider results. Item shape depends on the benchmark: technographics (coverage / breadth) vs lookalikes (Precision@10/50/100).", "items": { "oneOf": [ { "$ref": "#/components/schemas/TechnographicsProviderResult" }, { "$ref": "#/components/schemas/LookalikeProviderResult" } ] } }, "methodology_url": { "type": "string", "format": "uri" } }, "required": [ "slug", "name", "results" ] }, "TechnographicsProviderResult": { "type": "object", "properties": { "rank": { "type": "integer", "minimum": 1 }, "provider": { "type": "string" }, "provider_slug": { "type": "string" }, "companies_observed": { "type": "integer", "nullable": true, "description": "Distinct companies the vendor surfaced at least one detection for across the matrix." }, "total_companies": { "type": "integer", "nullable": true }, "categories_covered": { "type": "integer", "nullable": true, "description": "Count of the 8 canonical departments the vendor returned at least one tool for." }, "canonical_categories": { "type": "integer", "nullable": true }, "category_coverage_pct": { "type": "number", "nullable": true, "description": "HEADLINE ranking metric. Of 8 canonical departments, the % the vendor returned at least one tool for." }, "distinct_tools": { "type": "integer", "nullable": true, "description": "Count of canonical tools (out of 40) the vendor surfaced at least one company for." }, "distinct_company_tool_pairs": { "type": "integer", "nullable": true, "description": "Total distinct (company, tool) pairs the vendor returned. Breadth metric." }, "total_observations": { "type": "integer", "nullable": true }, "avg_tools_per_company": { "type": "number", "nullable": true }, "consensus_agreement_pct": { "type": "number", "nullable": true, "description": "% of this provider's tuples also reported by at least one other provider. Cross-provider corroboration signal." }, "uniqueness_pct": { "type": "number", "nullable": true, "description": "% of this provider's tuples that no other provider returned. Could be unique coverage OR unique noise." }, "total_cost_usd": { "type": "number", "nullable": true }, "cost_per_company_usd": { "type": "number", "nullable": true }, "cost_per_tool_usd": { "type": "number", "nullable": true, "description": "Total estimated request cost divided by distinct (company, tool) tuples surfaced. Null when public pricing is unavailable." }, "avg_latency_ms": { "type": "number", "nullable": true, "description": "Mean per-snapshot request latency in milliseconds." }, "category_breakdown": { "type": "object", "additionalProperties": { "type": "object", "properties": { "distinct_tools": { "type": "integer" }, "companies": { "type": "integer" } } } } }, "required": [ "rank", "provider", "provider_slug" ] }, "LookalikeProviderResult": { "type": "object", "description": "Per-vendor row on the lookalike benchmark. Precision is fixed-cutoff (10 / 50 / 100); Precision@100 is the headline.", "properties": { "rank": { "type": "integer", "minimum": 1 }, "provider": { "type": "string" }, "provider_slug": { "type": "string" }, "avg_precision_at_10": { "type": "number", "nullable": true, "description": "Mean Precision@10 across judged seeds (share of the top 10 returned the LLM judge scored relevant)." }, "avg_precision_at_50": { "type": "number", "nullable": true, "description": "Mean Precision@50 across judged seeds." }, "avg_precision_at_100": { "type": "number", "nullable": true, "description": "HEADLINE metric. Mean Precision@100 across judged seeds." }, "total_relevant": { "type": "integer", "nullable": true, "description": "Total relevant lookalikes the judge scored across the cohort. Reach / tiebreak metric." }, "seeds_judged": { "type": "integer", "nullable": true, "description": "Seeds where the vendor returned enough results and the judge scored every one." }, "total_seeds": { "type": "integer", "nullable": true }, "cost_per_relevant_usd": { "type": "number", "nullable": true, "description": "Estimated request spend divided by total relevant. Null when public pricing is unavailable." }, "avg_latency_ms": { "type": "number", "nullable": true } }, "required": [ "rank", "provider", "provider_slug" ] }, "Metric": { "type": "object", "properties": { "key": { "type": "string" }, "label": { "type": "string" }, "unit": { "type": "string", "enum": [ "percent", "usd", "milliseconds", "count" ] }, "direction": { "type": "string", "enum": [ "higher_is_better", "lower_is_better", "neutral" ] }, "definition": { "type": "string" } }, "required": [ "key", "label", "unit", "direction" ] }, "Error": { "type": "object", "properties": { "error": { "type": "string" }, "message": { "type": "string" }, "available": { "type": "array", "items": { "type": "string" } }, "retry_after_seconds": { "type": "integer" } }, "required": [ "error", "message" ] } } } }