GPT-OSS free API
OpenAI open-weight GPT-OSS. Free API on Groq (openai/gpt-oss-120b and openai/gpt-oss-20b, 30 RPM and 1,000 requests/day) — Groq's replacement for its retired Llama models. Also runs locally on Ollama; the default tag is about 14 GB with a 128K context window.
Free providers
| Provider | RPM | Daily | Model ID | Checked |
|---|---|---|---|---|
| groq | 30 | 1000 | openai/gpt-oss-120b | 2026-10-10 |
Example (groq)
curl https://api.groq.com/openai/v1/chat/completions \
-H "Authorization: Bearer $GROQ_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-oss-120b",
"messages": [{"role": "user", "content": "Hello"}]
}' FAQ
- Is GPT-OSS free via API?
- Yes — free hosted routes are listed on this page from groq. Limits change; check the table and the provider pages for the latest RPM and daily caps.
- Which free provider should I use for GPT-OSS?
- Use groq for the free route on this page.
- What are the free rate limits for GPT-OSS?
- Published free-tier figures on this page include 30 RPM and 1000 requests/day. Confirm on the linked provider page before production use.
- Can I run GPT-OSS locally instead of using a free API?
- Yes. Pull it with Ollama using tag gpt-oss and point your editor at http://localhost:11434/v1.
- How do I stack multiple free API keys?
- See our free API stack guide to combine Groq, OpenRouter, and Gemini keys, then compare limits on the free API comparison table.
Limits checked against official docs on 2026-10-10. Rates change without notice. Not affiliated with providers unless noted. Verify before production use.