This is the repo from the video, plus the exact settings that connect it to Claude Code, Cursor and Codex. Everything here is on a permanent free tier, not a trial with credits that run out.
The repo
github.com/open-free-llm-api/awesome-freellm-apis
It is a directory of free LLM APIs with the rate limit, the context window and a direct link to get a key for each one. It is updated continuously, so treat it as the source of truth rather than anything you copy out of it once.
Providers listed include Google Gemini, NVIDIA NIM, Groq, Mistral, Cohere, Cerebras, DeepSeek, Hugging Face, SambaNova, xAI and OpenRouter
1. Pick a provider before you pick a model
Four things decide whether a free tier is actually usable for what you are doing. Check them in this order:
- Rate limit. Requests per minute and per day. A generous daily cap is useless if the per-minute limit stalls an agent mid-run.
- Context window. Agent work eats context far faster than chat does.
- Whether it needs a card or a phone number. Some free tiers verify, some do not.
- OpenAI SDK compatibility. If the endpoint speaks the OpenAI format, every tool below will accept it with no code changes.
2. Point Claude Code at it
Claude Code reads two environment variables. Set them and it will route through the provider instead of the default endpoint:
- ANTHROPIC_BASE_URL, set to the provider's base URL
- ANTHROPIC_AUTH_TOKEN, set to your key from that provider
Set them in your shell profile if you want it permanent, or inline for a single session so you can switch providers per project.
Free, for your email
Get the full setup steps + the copy-paste setup prompt
The rest of this guide (4 more sections) opens right here. The setup prompt, one paste into Claude that sets it all up, comes by email only. Use your real address.
Unlocks: 3. Point Cursor at it · 4. Point Codex at it · What to watch for · The one page version
50,000+ follow where these guides come from. No spam, reply “unsubscribe” any time.