No-Log LLM API: Content-Stateless Real-Time Inference
A no-log LLM API you can verify: content-stateless real-time inference, explicit encrypted Batch retention, attested TEE routes, and 220+ model routes.
No prompt logs, enforced by code you can read and attestation you can check.
Most no-log claims are a sentence in a privacy policy. TrustedRouter never logs prompt or output content. Ordinary synchronous and streaming inference does not retain it. The opt-in Batch API temporarily retains enclave-encrypted artifacts for up to 30 days. On TEE routes that behavior runs inside a hardware-attested confidential-compute gateway: the source is open, the image digest is published, and a live attestation signed by the CPU vendor's root key is one curl away.
The API is OpenAI-compatible. Point base_url at https://api.trustedrouter.com/v1 and keep your SDK, your model ids, and your code. You get 220+ model routes across 30+ providers, including Claude, GPT, Gemini, DeepSeek, Kimi, Qwen, and GLM, with automatic rollover when a provider goes down.
NONCE=$(openssl rand -hex 16)
curl "https://api.trustedrouter.com/attestation?nonce=$NONCE"
# Returns a JWT signed by the CPU vendor's root key.
# image_digest must match the published open-source build.Content is not stored
TrustedRouter never logs prompt or output content. Ordinary synchronous and streaming inference does not retain it. The opt-in Batch API temporarily retains enclave-encrypted artifacts for up to 30 days. Usage metadata is separate: token counts and billing records are kept so your account and spend tracking work.
Three kinds of no-log, labeled per route
Open routes run through the attested gateway to any provider. Zero-Data-Retention routes add providers contractually bound to keep nothing; they rely on the provider honoring that contract. TEE routes run end-to-end in confidential compute, where even the provider cannot read your prompt. Full breakdown at zero data retention.
Anonymous-friendly access
Prepaid credits via Stripe, bring your own provider key, or pay per request with x402. Continue with MetaMask signs you in with a wallet and works without a traditional account. Built to be friendly to pseudonymous use; we do not promise anonymity.
Attestation beats a privacy policy.
Roleplay chats, character cards, journals, drafts: the text people send to an LLM API is personal. The standard advice for no-log inference is to run models locally, which caps you at whatever your hardware can hold. TrustedRouter's answer is a gateway you can inspect: open source, a published image digest, and attestation bound to the live TLS session so nothing can be replayed at you.
The catalog holds up its end. DeepSeek, Kimi, Qwen, GLM, and MiniMax routes are served by non-Chinese hosting providers on attested infrastructure with ZDR and TEE tiers, so you get frontier Chinese open-weights without sending your data to the model vendor. Pick a route with the model chooser, compare latency measured from real routed requests at the latency benchmarks, and see per-model rates at pricing.
Current routes, prices, privacy, and measured performance.
Catalog facts come from the routes currently configured in TrustedRouter. Performance uses the same cached metadata snapshot as the public leaderboard. Prompts and outputs are not part of these measurements.
| Model | Providers | Context | Input | Output | Privacy | Measured route |
|---|---|---|---|---|---|---|
Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 |
|
1,000,000 | $5.25/1M | $26.25/1M | varies 5 cited scores | 2886 ms TTFT anthropic · 58 tok/s · 100.00% available · n=5 |
OpenAI: GPT-5.5openai/gpt-5.5 |
|
1,050,000 | $5.25/1M | $31.5/1M | ZDR 3 cited scores | 2180 ms TTFT openai · 100.00% available · n=5 |
Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash |
+1
|
1,048,576 | $1.575/1M | $9.45/1M | ZDR | 1997 ms TTFT google-ai-studio · 100.00% available · n=17 |
MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code |
+8
|
262,144 | $0.735/1M to $0.9975/1M | $3.675/1M to $4.2/1M | ZDR 5 cited scores | 1485 ms TTFT inceptron · 100.00% available · n=45 |
Z.ai: GLM 5.2z-ai/glm-5.2 |
+22
|
1,048,576 | $0.714/1M to $1.575/1M | $1.575/1M to $5.5125/1M | E2EE 4 cited scores | 4152 ms TTFT parasail · 29 tok/s · 97.44% available · n=231 |
MiniMax: MiniMax M3minimax/minimax-m3 |
+8
|
1,048,576 | $0.2835/1M to $0.63/1M | $1.155/1M to $2.52/1M | ZDR 4 cited scores | 1785 ms TTFT minimax · 69 tok/s · 100.00% available · n=41 |
Browse every modelReview provider policiesOpen the full leaderboardSnapshot 2026-08-07T04:33:46Z
Questions
Does TrustedRouter log my prompts?
TrustedRouter never logs prompt or output content. Ordinary synchronous and streaming inference does not retain it. The opt-in Batch API temporarily retains enclave-encrypted artifacts for up to 30 days. Usage metadata is separate: token counts and billing records are kept so your account and spend tracking work. On TEE routes, the request path runs inside a hardware-attested confidential-compute gateway with open source and a published image digest. Privacy you can check beats privacy you're promised.
How do I verify the no-log claim myself?
Request a live attestation: curl the /attestation endpoint with a fresh nonce and you get back a JWT signed by the CPU vendor's root key. Confirm the image_digest matches the published open-source artifact. The attestation binds to the live TLS session, so an old response cannot be replayed at you. The full procedure is documented on the security page.
Do all routes have the same privacy level?
No. Every route carries a tier. Open routes run through the attested gateway to any provider. Zero-Data-Retention routes use providers contractually bound to keep nothing, so they rely on the provider honoring that contract. TEE routes run end-to-end in confidential compute, where even the provider cannot read your prompt.
Can I use TrustedRouter without a traditional account?
Yes. Continue with MetaMask signs you in with a wallet and works without a traditional account, and x402 supports pay-per-request crypto payments. Prepaid credits via Stripe and bring-your-own-key access are also available. This is anonymous-friendly access built for pseudonymous use; TrustedRouter does not promise anonymity guarantees.