The AI Loadout
←Back to models
Checked Active: 2026-10-19
Liquid AI

LFM2.5 24B-A2B (local)

A fast local model for agents on a 24 to 32 GB machine. It has 24B parameters but uses only about 2B per token, so it reasons and calls tools quickly.

Evaluation Analysis: Released Oct 13, 2026 as the largest and final model in Liquid AI's LFM2.5 family. It always reasons before it answers. Liquid AI reports 48.09 on LiveClaw against 38.19 for Qwen3.6-35B-A3B, 27.69 on AutomationBench against 8.72, and 184 tokens per second on an Apple M5 Max. It was trained inside agent harnesses including OpenClaw, Hermes Agent, Pi and OpenCode, and Ollama can start one directly: ollama launch opencode --model lfm2.5:24b. Text only. Commercial use is free for companies under $10M in annual revenue under the LFM Open License v1.0. Plan on RAM or VRAM of roughly the download size plus 2 to 4 GB for context.
Context Window128K
Input Price / 1MFree (local)
Output Price / 1MFree (local)
Release Date2026-10-13

Run it locally

ollama run lfm2.5:24b
Ollama tag
lfm2.5:24b (Q4_K_M; also 24b-a2b-q4_0, -q8_0, -bf16)
LM Studio name
Not in the LM Studio catalog yet (it lists the older LFM2-24B-A2B)
Download size by quantization
Q4_0 14 GB; Q4_K_M 15 GB; Q8_0 25 GB; BF16 48 GB
License
LFM Open License v1.0 (free commercial use under $10M annual revenue)

Memory needed is roughly the download size plus 2 to 4 GB for context. See how to run AI locally.

Specs and prices checked 2026-10-19 against the official page.

Used in these setups

Related skills and personas

Agent skills and bot personas that pair well with LFM2.5 24B-A2B (local).

Browse full directories:All Skills (>35)·All Personas (>15)

Want tested skills and personas installed in one command? See the AI Loadout packs, from $9.