Qwen3.8 27B Uncensored
inferway/qwen3.8-27b-uncensored
Rate limits and quotas
Anonymous playground8 / s burst · 90 / min · 500 / hour · 1,000 / dayApplies only to anonymous requests without an API key. Per IP: 500,000 tokens per rolling 24 hours.
Registered free accounts8 / s burst · 90 / min · 500 / hour · 1,000 / daySame request caps as above, with 4 concurrent requests and up to 2,000,000 tokens a day (input plus output, reset at 00:00 UTC). No per-request token cap.
Funded accountsAccount limitsPaid requests use your account’s limits, visible in the console.
Non-streaming timeout~100 s first byteNon-streaming requests pass through a CDN. Use stream=true for long outputs.
Guest playgroundNo key requiredThe box below calls the same gateway under the free-use caps. No sign-up, no card.
Concurrency64 / key, defaultCounted per API key. Paid accounts default to 64 per key; registered free accounts get 4.
Try it
Ask anything. The response streams token by token from the same gateway your code will call.
Copy setup for your AI assistant
Compatible with OpenAI SDKs and Python/Node clients: sets base_url and model id, reads API key from an environment variable, and writes a streaming smoke test.Data and hardware
Data handlingIn-memory text inference
Ordinary self-hosted text inference runs in memory and is not written to disk; see Privacy Policy for H3 ciphertext and optional processing exceptions. Only request metadata — token counts, timestamps, latency, routing — is retained, for up to 90 days, for billing and reliability.
Transparency hubHardwareHardware we operate
This model is served at NVFP4 on hardware we operate, with no silent downgrade under load. Execution happens only in the published inference countries, Canada and the United States, and the region each request ran in is recorded on that request.
Transparency hub