CoolFace
Modelpublic

mradermacher/Chat2DB-SQL-7B-i1-GGUF

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face
0likes1.1kdownloads
Model Card

About

<!-- ### quantizeversion: 2 --> <!-- ### outputtensorquantised: 1 --> <!-- ### converttype: hf --> <!-- ### vocab_type: --> <!-- ### tags: nicoboss --> weighted/imatrix quants of https://huggingface.co/Chat2DB/Chat2DB-SQL-7B

<!-- provided-files --> static quants are available at https://huggingface.co/mradermacher/Chat2DB-SQL-7B-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)

LinkTypeSize/GBNotes
GGUFi1-IQ1_S1.6for the desperate
GGUFi1-IQ1_M1.8mostly desperate
GGUFi1-IQ2_XXS2.0
GGUFi1-IQ2_XS2.1
GGUFi1-IQ2_S2.3
GGUFi1-IQ2_M2.5
GGUFi1-Q2_K2.6IQ3_XXS probably better
GGUFi1-IQ3_XXS2.7lower quality
GGUFi1-IQ3_XS2.9
GGUFi1-IQ3_S3.0beats Q3_K*
GGUFi1-Q3KS3.0IQ3_XS probably better
GGUFi1-IQ3_M3.2
GGUFi1-Q3KM3.4IQ3_S probably better
GGUFi1-Q3KL3.7IQ3_M probably better
GGUFi1-IQ4_XS3.7
GGUFi1-Q404_43.9fast on arm, low quality
GGUFi1-Q404_83.9fast on arm+i8mm, low quality
GGUFi1-Q408_83.9fast on arm+sve, low quality
GGUFi1-Q4_03.9fast, low quality
GGUFi1-Q4KS4.0optimal size/speed/quality
GGUFi1-Q4KM4.2fast, recommended
GGUFi1-Q5KS4.8
GGUFi1-Q5KM4.9
GGUFi1-Q6_K5.6practically 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. Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.

<!-- end -->