← Back to Models

lfm2-24b-a2b

liquid/lfm2-24b-a2b

View on HuggingFace
Publisher
liquid
Architecture
lfm2_moe(sparse)
Parameters
24B
Quantization
4bit (4.0 bpw)
Format
mlx
Max Context
128k tokens
Type
llm
File Size
12.50 GB
Est. RAM
14.0 GB
Est. VRAM
14.0 GB(~2B active)
Capabilities
trained_for_tool_use

Benchmark Results

Compare Models