Endpoints overview

About

Prism exposes 97 endpoints across 20+ categories. All inference endpoints live under /v1/ and follow the OpenAI API format. Admin endpoints live under /-/ and require an admin token.

Base URL

All endpoints are relative to your Prism gateway URL:

https://gateway.futureagi.com

Inference endpoints use the /v1/ prefix and accept your virtual API key (sk-prism-...) as a Bearer token. Admin endpoints use the /-/ prefix and require the admin token.

---

Chat and completions

The primary endpoints for generating text with LLMs.

Method	Path	Description
POST	/v1/chat/completions	Chat completion (streaming and non-streaming)
POST	/v1/completions	Text completion (legacy)
POST	/v1/count_tokens	Count tokens for a set of messages

---

Embeddings, reranking, and search

Method	Path	Description
POST	/v1/embeddings	Generate text embeddings
POST	/v1/rerank	Rerank text passages by relevance
POST	/v1/search	Search API
POST	/v1/ocr	Optical character recognition

---

Audio

Method	Path	Description
POST	/v1/audio/speech	Text-to-speech
POST	/v1/audio/speech/stream	Streaming text-to-speech
POST	/v1/audio/transcriptions	Speech-to-text (Whisper)
POST	/v1/audio/translations	Translate audio to English

---

Images and video

Method	Path	Description
POST	/v1/images/generations	Generate images from prompts
POST	/v1/videos	Submit video generation job
GET	/v1/videos	List video jobs
GET	/v1/videos/{video_id}	Get video job status
DELETE	/v1/videos/{video_id}	Cancel video job

---

Files

Method	Path	Description
POST	/v1/files	Upload a file
GET	/v1/files	List files
GET	/v1/files/{file_id}	Get file metadata
GET	/v1/files/{file_id}/content	Download file content
DELETE	/v1/files/{file_id}	Delete a file

---

Vector stores

Used with the Assistants API for file-based retrieval.

Method	Path	Description
POST	/v1/vector_stores	Create vector store
GET	/v1/vector_stores	List vector stores
GET	/v1/vector_stores/{id}	Get vector store
POST	/v1/vector_stores/{id}	Update vector store
DELETE	/v1/vector_stores/{id}	Delete vector store
POST	/v1/vector_stores/{id}/search	Search a vector store
POST	/v1/vector_stores/{id}/files	Add file to vector store
GET	/v1/vector_stores/{id}/files	List files in vector store
DELETE	/v1/vector_stores/{id}/files/{file_id}	Remove file from vector store
POST	/v1/vector_stores/{id}/file_batches	Batch add files

---

Assistants API

Full proxy for the OpenAI Assistants API. Create assistants, manage threads, send messages, and execute runs.

Assistants

Method	Path	Description
POST	/v1/assistants	Create assistant
GET	/v1/assistants	List assistants
GET	/v1/assistants/{id}	Get assistant
POST	/v1/assistants/{id}	Update assistant
DELETE	/v1/assistants/{id}	Delete assistant

Threads

Method	Path	Description
POST	/v1/threads	Create thread
GET	/v1/threads/{id}	Get thread
POST	/v1/threads/{id}	Update thread
DELETE	/v1/threads/{id}	Delete thread

Messages

Method	Path	Description
POST	/v1/threads/{id}/messages	Add message
GET	/v1/threads/{id}/messages	List messages
GET	/v1/threads/{id}/messages/{msg_id}	Get message
POST	/v1/threads/{id}/messages/{msg_id}	Update message
DELETE	/v1/threads/{id}/messages/{msg_id}	Delete message

Runs

Method	Path	Description
POST	/v1/threads/{id}/runs	Create run
GET	/v1/threads/{id}/runs	List runs
GET	/v1/threads/{id}/runs/{run_id}	Get run
POST	/v1/threads/{id}/runs/{run_id}	Update run
POST	/v1/threads/{id}/runs/{run_id}/cancel	Cancel run
POST	/v1/threads/{id}/runs/{run_id}/submit_tool_outputs	Submit tool outputs
GET	/v1/threads/{id}/runs/{run_id}/steps	List run steps
GET	/v1/threads/{id}/runs/{run_id}/steps/{step_id}	Get run step
POST	/v1/threads/runs	Create thread and run in one call

---

Responses API

Method	Path	Description
POST	/v1/responses	Create response
GET	/v1/responses/{id}	Get response
DELETE	/v1/responses/{id}	Delete response

---

Async inference

Method	Path	Description
GET	/v1/async/{job_id}	Get async job status and result
DELETE	/v1/async/{job_id}	Cancel async job

Async jobs are created by sending a regular chat completion request with async mode enabled. The batch API is available via admin endpoints below.

---

Scheduled completions

Method	Path	Description
POST	/v1/scheduled	Schedule a completion for later
GET	/v1/scheduled	List scheduled jobs
GET	/v1/scheduled/{job_id}	Get scheduled job
DELETE	/v1/scheduled/{job_id}	Cancel scheduled job

---

Realtime (WebSocket)

Method	Path	Description
GET	/v1/realtime	Upgrade to WebSocket for real-time audio/video streaming

---

Native format passthrough

For clients that prefer a provider’s native API format instead of the OpenAI format.

Method	Path	Description
POST	/v1/messages	Anthropic Messages API (native format)
POST	/v1/messages/count_tokens	Anthropic token counting
POST	/v1beta/models/{model}:generateContent	Google GenAI generate content
POST	/v1beta/models/{model}:streamGenerateContent	Google GenAI streaming

---

Models

Method	Path	Description
GET	/v1/models	List all available models
GET	/v1/models/{model}	Get model details

---

MCP (Model Context Protocol)

Prism acts as an MCP server, aggregating tools from upstream MCP tool servers.

Method	Path	Description
POST	/mcp	MCP protocol endpoint
GET	/mcp	MCP SSE streaming endpoint

Management

Method	Path	Description
GET	/-/mcp/status	MCP server status and stats
GET	/-/mcp/tools	List available tools
GET	/-/mcp/resources	List MCP resources
GET	/-/mcp/prompts	List MCP prompts
POST	/-/mcp/test	Test tool execution

---

A2A (Agent-to-Agent)

Method	Path	Description
GET	/.well-known/agent.json	Agent capabilities card
POST	/a2a	A2A protocol messages
GET	/v1/agents	List registered A2A agents

---

Admin: key management

Requires admin token.

Method	Path	Description
POST	/-/keys	Create API key
GET	/-/keys	List keys
GET	/-/keys/{key_id}	Get key details
PUT	/-/keys/{key_id}	Update key
DELETE	/-/keys/{key_id}	Revoke key
POST	/-/keys/{key_id}/credits	Add credits to key

---

Admin: organization config

Method	Path	Description
GET	/-/orgs/{org_id}/config	Get org config
PUT	/-/orgs/{org_id}/config	Set org config
DELETE	/-/orgs/{org_id}/config	Delete org config
GET	/-/orgs/configs	List all org configs
POST	/-/orgs/configs/bulk	Bulk load configs

---

Admin: operations

Method	Path	Description
GET	/-/cluster/nodes	List cluster nodes
POST	/-/admin/providers/{id}/rotate	Start key rotation
GET	/-/admin/providers/{id}/rotation	Get rotation status
POST	/-/admin/providers/{id}/rotate/promote	Promote rotated key
POST	/-/admin/providers/{id}/rotate/rollback	Rollback rotation
POST	/-/batches	Submit batch job
GET	/-/batches/{batch_id}	Get batch status
POST	/-/batches/{batch_id}/cancel	Cancel batch
GET	/-/shadow/stats	Shadow testing statistics

---

Health and diagnostics

Method	Path	Description
GET	/healthz	Liveness probe
GET	/livez	Liveness probe (alias)
GET	/readyz	Readiness probe
POST	/-/reload	Reload config from file
GET	/-/config	Server config summary
GET	/-/metrics	Prometheus metrics
GET	/-/health/providers	Provider health status
GET	/-/health/providers/{org_id}	Org-specific provider health

---

Next Steps