Every model.
Private by default.
An OpenAI-compatible endpoint. Point any OpenAI SDK at it — every model private by default, verifiably confidential when it matters. Paid per token in USDC.
Private by default.
Confidential when it matters.
Namespaced model IDs (provider/model). The live catalog, pricing, and each model’s private/confidential tag ride GET /v1/models — public, no key.
Zero data retention — providers are contractually bound not to store or train on your prompts.
Runs in a verified GPU-TEE; every response carries a signed receipt anchored on Sui you can check yourself.
Or pay per call
over x402.
Private Inference is a first-party service on the gateway, so an agent can pay per call from its wallet’s USDC — gasless, no key. The gateway handles 402 → pay → retry. The agent-native path.
$ t2 pay https://x402.t2000.ai/t2000/v1/chat/completions \ --data '{"model":"zai/glm-5.2","messages":[…]}' \ --max-price 0.10
The only Sui-native verifiable
confidential inference.
Anchored on Sui and read straight from a fullnode — so we can’t forge it.
Private by default
Every model is zero data retention — prompts and outputs are never stored, logged, or trained on.
Confidential tier
phala/* models run in a verified GPU-TEE. The gateway attests the upstream before forwarding — and fails closed if it can't.
Signed, Sui-anchored receipts
Every confidential response commits its hash on-chain — tamper-evident and publicly timestamped.
Verify it yourself
t2 verify checks the Sui anchor, receipt signature, and TDX quote on your machine. Check the proofs yourself.
Point any OpenAI client at it.
The API speaks the OpenAI Chat Completions format. Set two environment variables and most tools repoint with zero code changes — private by default across every model.
export OPENAI_BASE_URL="https://api.t2000.ai/v1" export OPENAI_API_KEY="sk-…"
One agent stack.
Every model, private.
One base URL.
Sign in at the console to mint a key + add credit — or fund from the CLI wallet in one command: t2 agent onboard --fund 5.