Machine Learning Agents5 min reading time

Run fast, private agents on-device with LFM2.5-24B-A2B

Covered by 3 sourcesPublished 12 days ago

LFM2.5-24B-A2B is a 24B-parameter reasoning model activating only 2B parameters per token, enabling fast, private on-device agents that outperform larger models on agentic tasks. It supports long contexts up to 131K tokens and integrates with popular agent frameworks.

Covered by 3 sources

MarkTechPost
Unite.AI
Hugging Face

#agentic-software-development #hermes-agent #lfm2-5-24b-a2b #openclaw #pi #qwen3-6-35b-a3b

Run fast, private agents on-device with LFM2.5-24B-A2B

Covered by 3 sources

More in Machine Learning

Researchers from Stanford and UC Berkeley Introduce TokenGate: Token-Gated Attention Makes Long-Context Prefill Up to 11.3x Faster

Jack Dorsey’s Block made an AI model that talks to you like an awkward new friend

AI21 Labs Releases LIFT-Bench to Curb Reward Hacking in RL Training