Decentralized AI Infrastructure
The Future ofAI Computing
Access powerful AI models through a decentralized, verifiable network. No middlemen. No markup. Just compute.
Access GPU-accelerated AI inference through a decentralized network of verified providers. TEE-secured execution ensures your data stays private.
Decentralized data availability layer with erasure coding and proof-of-random-access. Store training data and model weights on-chain.
High-throughput data availability sampling for AI workloads. Guaranteed data retrieval with cryptographic proofs.
Browse verified AI providers and their models. Transparent pricing, TEE-secured execution, instant access.
Explore available AI models and providers. Compare pricing, latency, and capabilities — no wallet needed.
Choose your model and provider. Sign in and deposit funds to your compute ledger.
Send inference requests through the decentralized network. Pay only for what you use, token by token.
Every response is cryptographically verified through TEE attestation. Trustless, transparent, provable.
Drop-in replacement for any OpenAI SDK. Switch your base URL and start using decentralized AI in minutes.
from openai import OpenAI client = OpenAI( base_url="https://router-api-staging.0g.ai/v1", api_key="<YOUR_API_KEY>",) response = client.chat.completions.create( model="<MODEL_ID>", messages=[ {"role": "user", "content": "Hello!"} ], stream=True,) for chunk in response: if chunk.choices[0].delta.content: print(chunk.choices[0].delta.content, end="")Decentralized compute eliminates middleman markup. Compare per-1K-token costs for equivalent model quality.
Approximate per-1K-token pricing for comparable model quality. Actual prices vary by provider.
No. You can explore all available models, compare pricing, and view provider details without connecting a wallet. Account verification is only required when you want to send Router inference requests.
Trusted Execution Environments (TEEs) create hardware-isolated enclaves that process your inference requests. The TEE generates cryptographic attestation proofs that verify the computation was performed correctly, without exposing your data.
The network uses 0G tokens (A0GI) for all transactions. You deposit tokens into your compute ledger, and costs are deducted per-token as you use AI models. Unused funds can be withdrawn at any time.
Yes. 0G Compute exposes an OpenAI-compatible API, so you can use any existing OpenAI SDK (Python, JavaScript, etc.) by simply changing the base URL. No code changes needed.
Providers set their own per-token pricing. The marketplace creates natural competition that drives prices down. You can compare prices across providers and choose the best value for your needs.
The decentralized architecture means multiple providers serve the same models. If one goes offline, you can seamlessly switch to another provider. Your funds remain safe in your on-chain ledger.