awesome-free-ai-coding · every free model · every provider · filter the list

AMD Radeon Cloud Token Factory free tier

🔌 LLM APIs with free tier · no card · provisional — added on 2026-11-18, a regular row from the first probe it passes on or after 2026-12-02 · live — last verified by a probe on 2026-11-18 · developer.amd.com.cn · back to the whole list

What you get

Free shared model API on AMD Radeon GPUs, with a recurring account-specific daily usage allowance

Free models

No free model family is currently listed here; callable ids, where the row has them, are under Connect.

Limits, in the vendor’s words

Shared API usage: Daily points allowance: see your account for values per account. 00:00 Beijing time

Typical gateway limit: 20 requests/minute per account (typical; current limits are returned in response headers)

Typical platform limit: 30 requests/minute; 8 requests at once per key (typical; current limits are returned in response headers)

China-site AMD AI Developer Program account and API key required. Registration asks for name, email and phone, with email or phone verification; some accounts need approval before receiving a key. Points meter shared API usage, carry no monetary value and are separate from GPU-instance credits. All models share the daily cap and rate limits; response headers and the usage endpoint report current allowances. For evaluation, prototyping and development; production workloads and reselling access are prohibited. Prompts and outputs are logged for two days, with longer retention for abuse or legal investigations; metadata is retained longer.

Where it is offered

The vendor names no country it keeps the offer from (source, read 2026-11-18).

What happens to what you send

What you send is not used to train models. In the vendor’s words: “We do not train models on your data” (source).

Connect

Try it from your terminal with your key in AMD_TOKEN_FACTORY_API_KEY — it goes from your machine to the vendor and nowhere else:

curl -s https://developer.amd.com.cn/radeon/api/v1/chat/completions \
  -H "Authorization: Bearer $AMD_TOKEN_FACTORY_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"model":"MiMo-V2.6-Flash","messages":[{"role":"user","content":"2+2? MAKE NO MISTAKES."}]}'

Evidence

History

Each line is a change to what this page publishes, dated the day it reached the list, in UTC.


Generated from registry.yaml and re-verified twice a week; the full list, the Atom feed and the machinery are at https://github.com/mvalentsev/awesome-free-ai-coding.