Mansa Platform Docs
Console

API overview

Use the Mansa API to add multilingual chat, translation, transcription, speech, and web tools to your application through one REST API.

The API is available at https://api.mymansa.ai. New to Mansa? Follow the Quickstart to create a key, add credits, and send your first request.

Prerequisites

To use the Mansa API, you need:

  • A Mansa account
  • A Mansa API key stored on your server
  • A funded Mansa Credits balance

Manage keys and credits in the Mansa console.

Available APIs

APIWhat it doesEndpoint
ChatGenerate multilingual assistant responses, stream output, and use web tools.POST /v1/chat
TranslateTranslate text between supported languages with control over tone and context.POST /v1/translate
TranscribeConvert supported audio recordings into text.POST /v1/transcribe
TTSGenerate WAV speech in a supported language and voice.POST /v1/audio/speech
Web searchSearch the public web and return a summary with source URLs.POST /v1/web-search
Web fetchExtract readable content from a public URL you already know.POST /v1/web-fetch

Open an API guide for its request fields, supported languages, examples, and response details.

Authentication

Every public /v1 endpoint requires a Mansa API key as a Bearer token:

Authorization: Bearer <MANSA_API_KEY>

Create a key on the API keys page and keep it in a server-side environment variable or secret manager. Never expose it in browser code, mobile applications, public repositories, or URLs. See Authentication for rotation and legacy-key guidance.

Request and response format

Send request bodies as JSON with Content-Type: application/json. Successful JSON responses place the result under data:

{
  "data": {
    "message": "Akwaaba! How can I help?"
  }
}

TTS is the exception: a successful request returns binary WAV audio instead of JSON. Non-success responses use a shared RFC 7807-inspired error shape with a stable context value for application logic. See Errors.

Billing and limits

Requests debit your Mansa Credits balance according to the API used. Chat usage is metered by input and output tokens; other capabilities use the units described in their guides. See Pricing.

The current default limit is 120 requests per minute per API key and per IP address. Chat supports up to 24 concurrent streams. Handle 429 responses with bounded exponential backoff and jitter. See Rate limits for recovery guidance.

Next steps

  • Quickstart — send your first authenticated Chat request
  • Authentication — create, store, rotate, and revoke API keys
  • Pricing — understand Mansa Credits and usage costs
  • Errors — handle validation, authentication, credit, and service failures