Liquid AI releases LFM2.5‑24B‑A2B with 2B active parameters
The 24-billion-parameter open-weight model activates 2 billion parameters per token and targets fast, private agents running locally.
AT A GLANCE
Key points
- Uses 24B total and 2B active parameters
- Offers a 131,072-token context window
- Open weights are available on Hugging Face
- Liquid AI reports 184 tokens per second on M5 Max
What is changing?
Liquid AI released LFM2.5‑24B‑A2B on October 13. The mixture-of-experts model has 24 billion total parameters and activates 2 billion per token; the company positions it for fast, private agents running on-device.
The context window increased from 32,768 tokens in the previous version to 131,072, while pretraining volume rose from 17 trillion to 33 trillion tokens. The base and post-trained versions are available as open weights on Hugging Face, with support for llama.cpp, MLX, vLLM, SGLang, and ONNX.
Liquid AI reports 184 tokens per second on an Apple M5 Max and up to 1.7 times the speed of comparable MoE models. These are provider measurements; actual speed should be retested for each hardware setup, quantization, context length, and agent toolchain.
Explore the original sources
Source date: 13 October 2026.
Published by AI Radar: 14 Oct 2026, 02:00 (Istanbul, UTC+3).
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…