Downloads · 30 days
1.3K
4% of all-time downloads
ikawrakow/various-2bit-sota-gguf
various-2bit-sota-gguf is a machine learning model from ikawrakow. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Various models in GGUF format quantized with a new 2-bit approach. Intended for use with llama.cpp. Requires llama.cpp PR 4773.
Downloads · 30 days
1.3K
4% of all-time downloads
All-time downloads
35K
Public
Repo size
151 GB
Likes
77
Public
Click a slice to open those files.
.gguf151 GB · 100%
From the Hugging Face model README
Various models in GGUF format quantized with a new 2-bit approach. Intended for use with llama.cpp. Requires llama.cpp PR 4773.
Update: PR 4773 has been merged into llama.cpp, but I have added new models that require PR 4856.
The new models are those that have around 2.3-2.4 bpw. They have a lower quantization error at the xpense of being ~10% larger.