Our 2-bit GGUF format for llama.cpp: Qwen3.5-4B, 2B, 0.8B, Qwen3-4B and Llama-3.1-8B, plus a GPU fit calculator. Runs with the v2.5 one-click build.
-
LLM GPU fit: Pingala-8 2-bit vs 4-bit
🧮1Will a model fit your GPU as Pingala-8 2-bit or Q4_K_M?
-
Genrius-AI/Qwen3-4B-Pingala8-2bit
Text Generation • 1B • Updated • 474 -
Genrius-AI/Qwen3.5-4B-Pingala8-2bit
Text Generation • 2B • Updated -
Genrius-AI/Qwen3.5-2B-Pingala8-2bit
Text Generation • 0.9B • Updated