Taalas API

Welcome to the Taalas API — a high-performance LLM inference API running on our HC1 hardcore model silicon.

Quick Start

All API requests require a Bearer token in the Authorization header:

Authorization: Bearer YOUR_API_KEY

Endpoints

Chat Completions (OpenAI-compatible)

POST /v1/chat/completions

{
  "model": "three_bit_numerics",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Hello!"}
  ],
  "temperature": 0.7,
  "max_tokens": 512,
  "stream": false
}

Text Completions (OpenAI-compatible)

POST /v1/completions

{
  "model": "three_bit_numerics",
  "prompt": "Once upon a time",
  "temperature": 0.7,
  "max_tokens": 256
}

Streaming

Set "stream": true in your request body to receive server-sent events (SSE). The response format follows the OpenAI streaming convention.

Available Models

GET /v1/models

Returns the list of available models and their metadata.

Health Check

GET /health

Use this endpoint to monitor service availability. It does not require authentication.

Rate Limits

API keys are subject to per-minute rate limits. If you exceed the limit, you'll receive a 429 Too Many Requests response. Wait briefly and retry.

API Docs

Report a Bug

Found an issue? Submit a bug report at /bug-report.