Reading · by Liquid AI
Liquid LFM
Run tiny on-device helpers that see, hear and talk back.
text → text, data · text, image, video, speech → text, speech · image → text
- Way in
- Official APINo sign-up, no key
- To start
- Freeopen weights (LFM Open License v1.0: free commercial use below $10M annual revenue)
- Runs
- On your MacAlso through Ollama, mlx-community
- Checked
- Nov 3, 2026Prices as of that day
What it can do 3
Run a fast small agent/tool-calling model alongside a bigger onelfm3:1.7b 1.1 GB; lfm2.5:8b (8B-A1B MoE) 5.2 GBtext → text, datafree (local)
Understand text, images, video and audio and answer in text and speech (LFM3-Omni-2B)2.0 GB q4_K_Mtext, image, video, speech → text, speechfree (local)
Describe images on device (LFM3-VL-2B)1.4-1.5 GBimage → textfree (local)
Does the same job
Prices and limits as of Nov 3, 2026, from Liquid AI’s own pages. They change; check before you rely on them.Not affiliated with Liquid AI. Not advice.Something wrong?