Forge

Forge / Free APIs & Models

Groq

Freemium

Speed-first hosted inference. Use as the failover and for tight agent loops that feel laggy on Gemini free.

No cardMM-friendlyAPI
Official site

Free tier

Free developer tier with daily request/token caps. Extremely fast.

Limits

Daily caps. Model list on the free tier is a subset.

Best for

  • Low-latency chat
  • Agent tool loops
  • API failover

How to use it well

  1. Create a free key. Point any OpenAI-compatible client at Groq’s base URL.
  2. Pick a current free model id from their docs — ids rotate.
  3. Cache prompts. Free TPM disappears if you resend the whole repo every turn.

Myanmar notes

Signup is card-free. Good complement when Google throttles evening peak in Asia.