7.4 billion tokens per month. 34 free LLM providers. 635 free model endpoints. All behind one /v1 endpoint, plus any custom OpenAI-compatible endpoint. Smart routing, automatic failover, encrypted keys. Personal experimentation only.
TypeScript
docs: credit @praveenkumarpranjal for #47 (#49)
Tashfeen18eb04be990b
diff --git a/README.md b/README.mdindex a34863cc..c2f5c842 100644--- a/README.md+++ b/README.md@@ -297,6 +297,7 @@ Thanks to everyone who's helped improve FreeLLMAPI:- [@deadc](https://github.com/deadc) — preserve Gemini `thoughtSignature` so multi-turn function calling stops 400-ing (#32); router model-first key-exhaustion tests + per-model `limits` hoist (#42)- [@zhangyu1324](https://github.com/zhangyu1324) — requested Ollama Cloud integration, now V10 catalog (#14 / #41)- [@jtbrennan-git](https://github.com/jtbrennan-git) — security review (#35) and Phase 1 hardening: parameterized analytics queries, sort-preset whitelist, timing-safe API key compare, mid-stream error sanitization+- [@praveenkumarpranjal](https://github.com/praveenkumarpranjal) — guard Gemini SSE `JSON.parse` so a malformed frame no longer aborts the whole stream, plus first streaming tests for the Google provider (#47)## Terms of Service review