Transformers
GGUF
English
Inference Endpoints
imatrix
conversational
mradermacher's picture
auto-patch README.md
df9a8a7 verified
metadata
base_model: flammenai/Mahou-1.2-llama3-8B
datasets:
  - flammenai/Grill-preprod-v1_chatML
  - flammenai/Grill-preprod-v2_chatML
language:
  - en
library_name: transformers
license: llama3
quantized_by: mradermacher
tags: []

About

weighted/imatrix quants of https://huggingface.co/flammenai/Mahou-1.2-llama3-8B

static quants are available at https://huggingface.co/mradermacher/Mahou-1.2-llama3-8B-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
GGUF i1-IQ1_S 2.1 for the desperate
GGUF i1-IQ1_M 2.3 mostly desperate
GGUF i1-IQ2_XXS 2.5
GGUF i1-IQ2_XS 2.7
GGUF i1-IQ2_S 2.9
GGUF i1-IQ2_M 3.0
GGUF i1-Q2_K 3.3 IQ3_XXS probably better
GGUF i1-IQ3_XXS 3.4 lower quality
GGUF i1-IQ3_XS 3.6
GGUF i1-Q3_K_S 3.8 IQ3_XS probably better
GGUF i1-IQ3_S 3.8 beats Q3_K*
GGUF i1-IQ3_M 3.9
GGUF i1-Q3_K_M 4.1 IQ3_S probably better
GGUF i1-Q3_K_L 4.4 IQ3_M probably better
GGUF i1-IQ4_XS 4.5
GGUF i1-Q4_0 4.8 fast, low quality
GGUF i1-Q4_K_S 4.8 optimal size/speed/quality
GGUF i1-Q4_K_M 5.0 fast, recommended
GGUF i1-Q5_K_S 5.7
GGUF i1-Q5_K_M 5.8
GGUF i1-Q6_K 6.7 practically like static Q6_K

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.