Lfm2 5 24b A2b

Facts

Release: game · 2026-10-14 · 2026-10-14
Noted: a 24 billion parameter mixture-of-experts model with 245 million active parameters, pretrained on 33 trillion tokens and further trained for agent workflows · 2026-10-14
Noted: scored higher than Qwen3.6-35B-A3B, Gemma-4-26B-A4B-it, and Nemotron-3.5-Lightning (30B-A3B) on agent benchmarks including AutomationBench and BFCLv4 · 2026-10-14
Noted: Inference support already exists in llama.cpp, MLX, vLLM, SGLang, and ONNX, and the model is downloadable from Hugging Face as an open model · 2026-10-14

Connections

Produces of: liquid ai

Structured graph also available as JSON at /public/entities/lfm2-5-24b-a2b. CC BY 4.0.

All coverage

2h ago

Liquid AI Releases LFM2.5-24B-A2B for Local Agent Workloads

Liquid AI released LFM2.5-24B-A2B, a 24 billion parameter mixture-of-experts model with 245 million active parameters, pretrained on 33 trillion tokens and further trained for agent workflows. The company reports it scored higher than Qwen3.6-35B-A3B, Gemma-4-26B-A4B-it, and Nemotron-3.5-Lightning (30B-A3B) on agent benchmarks including AutomationBench and BFCLv4. Inference support already exists in llama.cpp, MLX, vLLM, SGLang, and ONNX, and the model is downloadable from Hugging Face as an open model.