Welcome

153 free models across 13 providers — explore, compare, and build without a credit card.

Open Playground

Stream responses from Llama 4, DeepSeek, Gemini & more

Compare models

Same prompt, 3 models, instant tok/s + TTFT comparison

Browse models

201+ models — filter by provider, speed & context

Continue learning

9 lessons · 3 paths · From foundations to production AI

Contact & Support

Talk to the team · partnership · bug reports · press

153

Free models available

Integrated providers

850+ tok/s

Peak speed on Groq

2M tokens

Largest context window

Latest AI News

Live · NewsAPI

Fetching latest AI headlines…

Trending AI Repos

Live · GitHub

Fetching trending repositories…

Top-rated free models

Browse all 201

FREE

Meta

Llama 3.3 70B (Versatile)

Meta's flagship 70B served via Groq — best free balance of quality and speed.

97 86

FREE

Meta

Llama 3.1 8B Instant

Fast lightweight Llama 3.1 on Groq's LPU — ideal for high-volume tasks.

99 70

FREE

Meta

Llama 3.2 1B

Ultra-tiny 1B model — blazing speed for simple classification/extraction tasks.

100 45

FREE

Meta

Llama 3.2 3B

Compact 3B with strong quality-speed trade-off on Groq's LPU hardware.

100 58

Questions, feedback, or a partnership idea?

Replies in < 24h

Our team reads every message. Whether it's a bug, a feature idea, enterprise inquiry, or just a hello — we'd love to hear from you.

hello@llmatlas.comDiscord Community@llmatlassecurity@llmatlas.com