PrivateMind PrivateMind API Explorer

PrivateMind API Explorer

The Try it console calls api.privatemind.com/v1 straight from your browser. Paste a key into the Auth panel to authorise calls. More on the spec & auth

PrivateMind API

OpenAI-compatible inference API for chat, responses, embeddings, vision, tools, audio, images, batch and reranking, plus the Anthropic Messages API.

The PrivateMind API is OpenAI-compatible. For chat completions and embeddings you can use the official openai client libraries by pointing base_url at https://api.privatemind.com/v1 and using a PrivateMind API key. Endpoints that have no OpenAI equivalent (rerank, voices, container execs) are called over plain HTTP or the client's generic request methods.

Authentication uses a bearer token of the form ACCESS_KEY_ID:SECRET (the two halves joined by a colon, passed as one opaque token). Manage keys under Settings → API Keys in the app. See the Authentication guide.

The same keys and models also serve the Anthropic Messages API (/messages, /messages/count_tokens, /messages/batches), for clients built on the Anthropic SDKs: point them at https://api.privatemind.com and send the key as x-api-key or a bearer token. /models and /files answer in the Anthropic shape when the request carries anthropic-version. See the Messages API guide.

/responses serves the OpenAI Responses API on the same keys and models, for clients such as Codex CLI and the OpenAI Agents SDK. It stores no responses: send the whole conversation in input on every request. See the Responses API guide.

This document is the machine-readable contract for the public API surface. It is hand-maintained alongside the human docs and is the file behind the interactive API explorer.

Chat

  • POST/chat/completionsCreate a chat completion
  • POST/chat/count_tokensCount input tokens

Responses

  • POST/responsesCreate a response

Embeddings

  • POST/embeddingsCreate embeddings

Rerank

  • POST/rerankRerank documents by relevance

Model catalog

  • GET/modelsList available models
  • GET/models/{model_id}Retrieve one model

Audio

  • POST/audio/speechSynthesise speech from text
  • POST/audio/transcriptionsTranscribe audio to text
  • GET/voicesList text-to-speech voices

Images

  • POST/images/generationsGenerate an image from a prompt

Usage

  • GET/usageList your usage rows
  • GET/usage/summaryAggregate your usage by group

Batch

  • GET/batchesList batches
  • POST/batchesCreate a batch
  • GET/batches/{batch_id}Retrieve a batch
  • POST/batches/{batch_id}/cancelCancel a batch

Files

  • GET/filesList files
  • POST/filesUpload a file
  • GET/files/{file_id}Retrieve file metadata
  • POST/files/{file_id}Change who can see a file
  • DELETE/files/{file_id}Delete a file
  • GET/files/{file_id}/contentDownload file content
  • GET/files/usageStorage usage

Sandboxes

  • GET/containersList sandboxes
  • POST/containersCreate a sandbox
  • GET/containers/{container_id}Retrieve a sandbox
  • DELETE/containers/{container_id}Delete a sandbox
  • GET/containers/{container_id}/filesList a sandbox's files
  • POST/containers/{container_id}/filesUpload a file into a sandbox
  • GET/containers/{container_id}/files/{file_id}Retrieve a sandbox file
  • DELETE/containers/{container_id}/files/{file_id}Delete a sandbox file
  • GET/containers/{container_id}/files/{file_id}/contentDownload a sandbox file
  • GET/containers/{container_id}/execsList a sandbox's execs
  • POST/containers/{container_id}/execsRun code in a sandbox
  • GET/containers/{container_id}/execs/{exec_id}Retrieve an exec
  • POST/containers/{container_id}/execs/{exec_id}/cancelCancel an exec

Messages

  • POST/messagesCreate a message
  • POST/messages/count_tokensCount message tokens

Message batches

  • GET/messages/batchesList message batches
  • POST/messages/batchesCreate a message batch
  • GET/messages/batches/{message_batch_id}Retrieve a message batch
  • DELETE/messages/batches/{message_batch_id}Delete a message batch
  • POST/messages/batches/{message_batch_id}/cancelCancel a message batch
  • GET/messages/batches/{message_batch_id}/resultsStream message batch results