ᛞ

The full stack of
decentralized AI

Every capability. One API key. 1,243 models from 124 providers — ten services behind a single endpoint.

ᚱᚢτᚾᛊ
ᚱ

Inference

SN58 Handshake + SN64 Chutes
POST /v1/chat/completions

1,243+ models across every major open-source family. OpenAI-compatible format. Smart routing picks the fastest available miner. Streaming, function calling, JSON mode — all supported.

# inference — same as OpenAI
from openai import OpenAI

client = OpenAI(
    base_url="https://api.opentau.ai/v1",
    api_key="otau_your-key"
)

response = client.chat.completions.create(
    model="meta-llama/llama-3.3-70b-instruct",
    messages=[{
        "role": "user",
        "content": "Explain Bittensor"
    }]
)
ᚢ

Fine-Tuning Live

SN56 Gradients
POST /v1/fine-tuning/jobs

$5/hr — 5x cheaper than OpenAI. QLoRA and full fine-tune on any open-source model. Upload your dataset, train on decentralized GPUs, deploy the result to your inference endpoint automatically.

# fine-tune for $5/hr
job = client.fine_tuning.jobs.create(
    model="meta-llama/llama-3.1-8b-instruct",
    training_file="file-abc123",
    hyperparameters={
        "method": "qlora",
        "n_epochs": 3,
        "learning_rate": 2e-4
    }
)
# → job.id = "ftjob-..."
ᛊ

Search Live

8 providers — Desearch SN22 + Tavily + Brave + Exa
POST /v1/search

Web, X/Twitter, Reddit, and ArXiv search via 8 decentralized and hybrid providers. AI-powered search with citations, summaries, and source links. No tracking, no filter bubbles.

# search the web + X/Twitter
curl https://api.opentau.ai/v1/search \
  -H "Authorization: Bearer otau_your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "latest Bittensor subnet launches",
    "model": "desearch/ai-search",
    "max_results": 10
  }'
ᛟ

Image Generation Live

4 providers — Replicate + Stability AI + fal.ai
POST /v1/images/generations

Generate images from text prompts via FLUX, Stable Diffusion, and more. Multiple sizes, batch generation, fast inference on dedicated GPU clusters. Same format as OpenAI Images API.

# generate an image with FLUX
curl https://api.opentau.ai/v1/images/generations \
  -H "Authorization: Bearer otau_your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A mystical rune circle glowing green",
    "model": "replicate/black-forest-labs/flux-schnell",
    "n": 1,
    "size": "1024x1024"
  }'
ᚾ

Code Execution Live

5 providers — E2B sandboxed environments
POST /v1/code/execute

Run Python, JavaScript, Bash, and more in secure sandboxed environments. Perfect for AI agents that need to execute code, run data analysis, or test hypotheses autonomously.

# execute Python in a sandbox
curl https://api.opentau.ai/v1/code/execute \
  -H "Authorization: Bearer otau_your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "language": "python",
    "code": "import math; print(math.pi ** 2)",
    "model": "e2b/python"
  }'
ᛃ

Web Scraping Live

9 providers — Webtools + Firecrawl + Apify
POST /v1/scrape

Fetch any webpage as clean markdown or structured HTML. Extract content from URLs for RAG pipelines, data collection, or research automation. Handles JavaScript-rendered pages.

# scrape a webpage as markdown
curl https://api.opentau.ai/v1/scrape \
  -H "Authorization: Bearer otau_your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://docs.bittensor.com",
    "format": "markdown",
    "model": "webtools/fetch-markdown"
  }'
ᚨ

Data APIs Live

47 providers — CoinGecko, Wolfram, Weather, SEC, Maps
POST /v1/data

Access 47 structured data providers through a single endpoint. Crypto prices, math computation, weather, financial filings, geocoding, and more. Feed real-time data directly into your AI pipelines.

# query live exchange rates
curl https://api.opentau.ai/v1/data \
  -H "Authorization: Bearer otau_your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "livedata/exchange-rate",
    "query": "TAO/USD"
  }'
ᛗ

Audio Live

1 provider — Whisper + Suno
POST /v1/audio

Speech-to-text transcription via Whisper and audio generation via Suno. Process audio input or generate spoken content programmatically through one unified endpoint.

# transcribe or generate audio
curl https://api.opentau.ai/v1/audio \
  -H "Authorization: Bearer otau_your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "whisper/large-v3",
    "input": "base64-encoded-audio-data"
  }'
ᛟᚨτᛃᚾ

How OpenTau compares

Feature OpenRouter Together AI OpenTau
Inference (1,243+ models) ✓ ✓ ✓
Fine-tuning -- ✓ ✓
Search (8 providers) -- -- ✓
Image generation -- ✓ ✓
Code execution -- -- ✓
Web scraping (9 providers) -- -- ✓
Data APIs (47 providers) -- -- ✓
Audio (transcription + gen) -- -- ✓
Cross-subnet pipelines -- -- ✓
Censorship-resistant -- -- ✓
124 providers -- -- ✓
Free tier ✓ -- ✓