> ## Documentation Index > Fetch the complete documentation index at: https://benchgen.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Create a benchmark > One request, no files to build: send a title, a description and 5-2000 items. The platform validates them, builds the benchmark with its question-answer or multiple-choice importer (which generates the ingestion, scoring and LLM-judge programs) and queues creation. Items with `choices` make a multiple-choice benchmark; otherwise it is question-answer. The benchmark is created unpublished. Requires the `benchmark:create` scope. For API tokens and agents, creation is limited per user across every creation path (default 10 per rolling 24 hours and 3 per minute); a dry run does not count, and the web app and platform admins are not limited. ## OpenAPI ````yaml POST /api/competitions/create_from_spec/ openapi: 3.0.3 info: title: BenchGen Platform API version: 1.0.0 description: >- One API for the whole BenchGen platform: model catalogue and serving, fine-tuning and benchmarks, datasets (knowledge), and billing. ## Authentication Every request uses the same credential: a platform API token sent as `Authorization: Bearer bgn_...`. Create tokens in the web app under Profile Settings > Platform API tokens (the secret is shown exactly once), or via `POST /api/tokens/` with an interactive session. Revoking a token disables it platform-wide within 60 seconds. ## Scopes A token carries scopes chosen at creation; a request outside the token's scopes gets `403` with an explanatory message. | scope | grants | |---|---| | `models:read` | read model catalogues, job status, logs, GPU info | | `models:write` | deploy, train, merge, stop models and jobs | | `benchmark:read` | read benchmark runs and results | | `benchmark:run` | launch benchmark runs | | `benchmark:create` | create benchmarks in your account (drafts, Excel, bundles, specs); counts toward the creation limit | | `benchmark:manage` | edit and delete benchmarks you own or collaborate on | | `benchmark:publish` | publish and unpublish benchmarks you own or collaborate on | | `knowledge:read` | read your datasets and fine-tuning data | | `knowledge:write` | create, edit and delete datasets and fine-tuning data | | `billing:read` | read your balance and usage | | `agents:chat` | chat with your own agents through the API | | `agents:manage` | manage your agents, knowledge bases and channels | | `admin` | everything the account can do (staff accounts only) | ## For agents This document plus `/api/llms.txt` are the machine-readable entry points. Responses are JSON. Errors use conventional status codes; the body carries `error` or `message`. Knowledge endpoints return `[{"data": [...], "meta": {...}}]`. The complete auto-generated schema of every endpoint (including internal ones) lives at `/api/public-docs.json` (Swagger 2.0); this document is the curated, stable, supported surface. contact: url: https://benchgen.com servers: - url: https://api.benchgen.com security: - platformToken: [] tags: - name: auth description: Token introspection for services and integrations - name: tokens description: Manage your platform API tokens - name: models description: Model catalogue and serving - name: finetune description: Fine-tuning jobs, inference deployments, GPUs - name: knowledge description: Datasets and fine-tuning data (knowledge API) - name: billing description: Balance and usage - name: agents description: Chat with your agents (OpenAI-compatible facade) - name: benchmark description: Public benchmark (competition) listings. paths: /api/competitions/create_from_spec/: post: tags: - benchmark summary: Create a benchmark from a JSON spec description: >- One request, no files to build: send a title, a description and 5-2000 items. The platform validates them, builds the benchmark with its question-answer or multiple-choice importer (which generates the ingestion, scoring and LLM-judge programs) and queues creation. Items with `choices` make a multiple-choice benchmark; otherwise it is question-answer. The benchmark is created unpublished. Requires the `benchmark:create` scope. For API tokens and agents, creation is limited per user across every creation path (default 10 per rolling 24 hours and 3 per minute); a dry run does not count, and the web app and platform admins are not limited. parameters: - name: dry_run in: query required: false schema: type: boolean description: 'Validate only: answers 200 with a summary and creates nothing' requestBody: required: true content: application/json: schema: type: object required: - title - items properties: title: type: string example: Unit Conversion Basics description: type: string type: type: string enum: - question_answer - multiple_choice description: Optional; inferred from the items when omitted items: type: array minItems: 5 maxItems: 2000 items: type: object required: - question - answer properties: question: type: string answer: type: string description: >- Short text answer, or for choices the letter (A, B, ...) or the exact text of the correct choice choices: type: array items: type: string description: 2-10 options; makes the spec multiple_choice context: type: string description: Optional passage for question_answer items group: type: string description: Optional topic or category used to group results responses: '200': description: 'dry_run: the spec is valid; nothing was created' '201': description: Creation queued content: application/json: schema: type: object properties: valid: type: boolean kind: type: string title: type: string status_id: type: integer description: Poll GET /api/competitions/{status_id}/creation_status/ '400': description: Input the caller can fix; `errors` lists every problem content: application/json: schema: type: object properties: valid: type: boolean example: false errors: type: array items: type: string example: - items[2].answer is empty '403': description: >- Token lacks benchmark:create, or the deployment restricts creation to certain roles (`code: role_required`) '429': description: Creation limit reached; wait for the `Retry-After` header headers: Retry-After: schema: type: integer description: Seconds until another creation is allowed content: application/json: schema: type: object properties: code: type: string enum: - quota_exceeded - rate_limited detail: type: string limit: type: integer used: type: integer description: quota_exceeded only resets_at: type: string format: date-time description: quota_exceeded only window_seconds: type: integer description: rate_limited only security: - platformToken: [] components: securitySchemes: platformToken: type: http scheme: bearer bearerFormat: bgn_ opaque token description: >- Platform API token created under Profile Settings > Platform API tokens. Scopes are fixed at creation. ````