README.md

metadata

tags:
  - uqff
  - mistral.rs
base_model: meta-llama/Llama-3.2-3B-Instruct
base_model_relation: quantized

meta-llama/Llama-3.2-3B-Instruct, UQFF quantization

Run with mistral.rs. Documentation: UQFF docs.

1. Flexible 🌀: Multiple quantization formats in one file format with one framework to run them all.
2. Reliable 🔒: Compatibility ensured with embedded and checked semantic versioning information from day 1.
3. Easy 🤗: Download UQFF models easily and quickly from Hugging Face, or use a local file.
4. Customizable 🛠️: Make and publish your own UQFF files in minutes.

Examples

Quantization type(s)	Example
FP8	./mistralrs-server -i plain -m EricB/Llama-3.2-3B-Instruct-UQFF --from-uqff llama3.2-3b-instruct-f8e4m3.uqff
HQQ4	./mistralrs-server -i plain -m EricB/Llama-3.2-3B-Instruct-UQFF --from-uqff llama3.2-3b-instruct-hqq4.uqff
HQQ8	./mistralrs-server -i plain -m EricB/Llama-3.2-3B-Instruct-UQFF --from-uqff llama3.2-3b-instruct-hqq8.uqff
Q3K	./mistralrs-server -i plain -m EricB/Llama-3.2-3B-Instruct-UQFF --from-uqff llama3.2-3b-instruct-q3k.uqff
Q4K	./mistralrs-server -i plain -m EricB/Llama-3.2-3B-Instruct-UQFF --from-uqff llama3.2-3b-instruct-q4k.uqff
Q5K	./mistralrs-server -i plain -m EricB/Llama-3.2-3B-Instruct-UQFF --from-uqff llama3.2-3b-instruct-q5k.uqff
Q8_0	./mistralrs-server -i plain -m EricB/Llama-3.2-3B-Instruct-UQFF --from-uqff llama3.2-3b-instruct-q8_0.uqff