mradermacher's picture
auto-patch README.md
2f42da0 verified
metadata
base_model: theprint/ReWiz-Qwen-2.5-14B
datasets:
  - theprint/ReWiz
language:
  - en
library_name: transformers
license: apache-2.0
quantized_by: mradermacher
tags:
  - text-generation-inference
  - transformers
  - unsloth
  - qwen2
  - trl
  - sft
  - theprint
  - rewiz

About

static quants of https://huggingface.co/theprint/ReWiz-Qwen-2.5-14B

weighted/imatrix quants are available at https://huggingface.co/mradermacher/ReWiz-Qwen-2.5-14B-i1-GGUF

Usage

If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files.

Provided Quants

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Link Type Size/GB Notes
PART 1 PART 2 Q2_K 11.6
PART 1 PART 2 Q3_K_S 13.4
PART 1 PART 2 Q3_K_M 14.8 lower quality
PART 1 PART 2 Q3_K_L 15.9
PART 1 PART 2 IQ4_XS 16.5
PART 1 PART 2 Q4_0_4_4 17.1 fast on arm, low quality
PART 1 PART 2 Q4_K_S 17.2 fast, recommended
PART 1 PART 2 Q4_K_M 18.1 fast, recommended
PART 1 PART 2 Q5_K_S 20.6
PART 1 PART 2 Q5_K_M 21.1
PART 1 PART 2 Q6_K 24.3 very good quality
PART 1 PART 2 Q8_0 31.5 fast, best quality

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):

image.png

And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9

FAQ / Model Request

See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized.

Thanks

I thank my company, nethype GmbH, for letting me use its servers and providing upgrades to my workstation to enable this work in my free time.