Akahsizrr/fuse-1-Lite-GGUF

fuse-1 Lite — GGUF

GGUF format of fuse-1 Lite for llama.cpp.

⚠️ Important: Custom llama.cpp Required

This GGUF uses the custom fuse3 architecture. Stock llama.cpp cannot load it. You need a llama.cpp fork with Fuse3 support.

See the integration guide for instructions on building llama.cpp with Fuse3 support.

Files

• fuse-1-Lite-f16.gguf — F16 (full precision)
• fuse-1-Lite-q4_k_m.gguf — Q4_K_M (4-bit quantized, ~4GB)
• fuse-1-Lite-q8_0.gguf — Q8_0 (8-bit quantized, ~7GB)

Usage (with custom llama.cpp)

./llama-cli -m fuse-1-Lite-f16.gguf -p "Write a Python function to check if a number is prime." -n 512 --temp 0.1

See the main model card for full architecture details.