Understand Targon

AI Evolved

Operating on Bittensor's Subnet 4, Targon provides seamless access to a decentralized network of AI models through our optimized API, powered by a network of high-performance miners

Network Architecture

Built on Bittensor's Subnet 4, our API seamlessly connects users to a decentralized network of miners, backed by a robust validator system.

API Layer

  • OpenAI-compatible interface for seamless integration
  • Smart request routing to high-performing miners

Bittensor Network

  • Decentralized network of validators and miners
  • Performance-based incentives for reliable responses
99.9%
Network Uptime
300+
Avg Tokens Per Second
OpenAI Compatibility
from openai import OpenAI

client = OpenAI(
    base_url="https://api.targon.com/v1",
    api_key="your-api-key"
)

response = await client.chat.completions.create(
    model="NousResearch/Meta-Llama-3.1-70B-Instruct",
    stream=True,
    messages=[
        {"role": "system", "content": "You are an assistant."},
        {"role": "user", "content": "Hello assitant!"},
    ],
)

Model Support

Targon supports various large language models through a standardized OpenAI-compatible interface.

Supported Models

  • •nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
  • •NousResearch/Meta-Llama-3.1-70B-Instruct
  • •And Many More

Model Hosting

Experience worry-free model deployment with our flexible leasing system. Choose between one-time 7-day leases or automatic weekly renewals for continuous operation.

Leasing Options

  • •Choose between one-time 7-day leases or automatic weekly renewals
  • •Your model is protected during the lease period and cannot be replaced
  • •Keep your model running smoothly with hassle-free lease renewals

Current Model Timeline

View full timeline

Ready to Get Started?

Browse our available models or lease your own with our comprehensive guides.

Browse ModelsLease a Model