No fake proxies, no silent substitutions. The model you call is the model you get — request is verifiably routed to the model's official channel, or attested against the exact GPU it's deployed on.
Absolute privacy, on demand, for sensitive workloads. Processed in hardware-isolated enclaves — neither 0G nor the provider sees it.
Open provider competition sets the price — no single-vendor markup, no top-up fees, no premium for verifiable, private inference.
Premium, verifiable models — while staying remarkably competitive on price.
Three tiers of routing — full coverage by default, dial up to verified or private when the task calls for it.
Every model and provider on the network, including community-hosted channels — best price and availability across the full catalog.
Everything in Verified, plus community channels
Routes only to verifiable providers — TeeML enclaves, or official channels verified via TeeTLS — so you always know which model ran. Privacy depends on the provider it lands on.
Everything in Private, plus official providers
Runs sealed inside TEE GPU enclaves and stays cryptographically verifiable — no one can see your data, not even us.
TEE (private enclave) providers only
A header for a single call, or a default on your API key — no new endpoint, no SDK changes.
Set this header on any call. Omit it and the key's default applies.
X-0G-Provider-Trust-Mode:verified|privateChoose a trust mode when you create an API key. Every request it makes uses that mode unless a header overrides it.
On verified and private requests, each inference is cryptographically signed, with a Proof ID coming soon. Confirm — independently — that the right model ran on the right hardware, untampered.
Trust is no longer a promise. It's a check you can run yourself.
Closed-source models: 0G only connects to authoritative, officially-backed model channels, and uses TeeTLS to verify the routing process.
Private models: the provider must deploy the model on a GPU equipped with TEE hardware. Once deployed, an attestation is generated and can no longer be altered.
Other open-source models: 0G uses SPML to verify the provider's GPU and model deployment.
On verified and private requests, each response is cryptographically signed, with a Proof ID coming soon. Paste it into the verifier (or call the verification endpoint) and it independently checks the attestation signature, the model hash, the enclave measurement, and the on-chain routing record. You don't have to trust us — you run the check yourself.
verified or private on any model, opt in explicitly (see below).verified for verifiable providers only, or private for TEE-isolated providers only. Omit the header for Standard.<br/><br/>Option 2 — API key default. Set a key's trust mode to Verified or Private in the dashboard, and every request it makes uses that mode unless a header overrides it.<br/><br/>Path: Dashboard → API Key → ⋯ → EditThe network uses 0G tokens for all transactions. You deposit into your compute ledger and costs are deducted per-token as you use models. There are no deposit or top-up fees, and unused funds remain available for service use.
Because the network is decentralized, multiple providers serve the same models. If one becomes unavailable, your request is automatically routed to another healthy provider serving that model. Your funds stay safe in your on-chain ledger throughout.
Providers set their own per-token rates, and the marketplace creates natural competition that drives prices down. You can compare prices across providers for the same model and route to the best value — frontier models come discounted, and open models go even lower.