# LFM3 Audio 2B requirements

Updated: 2026-11-01

2B Speech recognition, released 2026-10.

- Peak memory: ~4.61 GB at Q8_0, audio.cpp ASR
- Tools: audio.cpp (experimental)
- License: LFM Open License v1.0 (commercial use: conditional)
- Runs on 29 of 49 tracked devices

Sources: https://huggingface.co/LiquidAI/LFM3-Audio-2B, https://huggingface.co/LiquidAI/LFM3-Audio-2B-GGUF, https://huggingface.co/LiquidAI/LFM3-Audio-2B/blob/main/LICENSE, https://ollama.com/library/lfm3-audio, https://github.com/0xShug0/audio.cpp/blob/cd770076152085af5b166521185bec76f9d47808/docs/models/lfm3_audio.md#performance
More: https://localmodel.run/model/lfm3-audio-2b