Modelofficial release

LiquidAI/LFM2.5-24B-A2B

LFM2.5-24B-A2B →Hugging Face ↗
Parameters24B (2B active)
Context128K
Input → outputtext → text
Licenseother
PrecisionBF16
ArchitectureLfm2MoeForCausalLM
ReleasedOct 12, 2026
Downloads2.6K

Based on LiquidAI/LFM2.5-24B-A2B-Base

Other sizes and formats in the family

ModelParamsPrecisionDownloads
LiquidAI/LFM2.5-24B-A2B-GGUFgguf24B–17.8K

Compare side by side →

Built on this model

ModelParamsPrecisionDownloads
LiquidAI/LFM2.5-24B-A2B-GGUFgguf24B–17.8K

This release in the news

  1. Oct 12, 2026 · Model release
    LiquidAI/LFM2.5-24B-A2B

Recent news about LFM2.5-24B-A2B

  1. Oct 19, 2026 · Model release
    Nemotron 3.5 Bolt, Muse Spark Open, and a Fast LFM2.5 24B
  2. Oct 18, 2026 · Opinion / analysis
    Nine models, five coding tasks, one 16GB GPU
  3. Oct 15, 2026 · Opinion / analysis
    Very impressed with LFM2.5-24-A2B in Pi Agent
  4. Oct 14, 2026 · Model release
    Run fast, private agents on-device with LFM2.5-24B-A2B