RichardErkhov/Sao10K_-_Stheno-Mega-False-49B-L2-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Stheno-Mega-False-49B-L2 - GGUF

• Model creator: https://huggingface.co/Sao10K/
• Original model: https://huggingface.co/Sao10K/Stheno-Mega-False-49B-L2/

Name	Quant method	Size
Stheno-Mega-False-49B-L2.Q2_K.gguf	Q2_K	17.01GB
Stheno-Mega-False-49B-L2.IQ3_XS.gguf	IQ3_XS	18.8GB
Stheno-Mega-False-49B-L2.IQ3_S.gguf	IQ3_S	19.87GB
Stheno-Mega-False-49B-L2.Q3_K_S.gguf	Q3_K_S	19.87GB
Stheno-Mega-False-49B-L2.IQ3_M.gguf	IQ3_M	21.04GB
Stheno-Mega-False-49B-L2.Q3_K.gguf	Q3_K	22.31GB
Stheno-Mega-False-49B-L2.Q3_K_M.gguf	Q3_K_M	22.31GB
Stheno-Mega-False-49B-L2.Q3_K_L.gguf	Q3_K_L	24.46GB
Stheno-Mega-False-49B-L2.IQ4_XS.gguf	IQ4_XS	24.73GB
Stheno-Mega-False-49B-L2.Q4_0.gguf	Q4_0	25.97GB
Stheno-Mega-False-49B-L2.IQ4_NL.gguf	IQ4_NL	26.13GB
Stheno-Mega-False-49B-L2.Q4_K_S.gguf	Q4_K_S	26.14GB
Stheno-Mega-False-49B-L2.Q4_K.gguf	Q4_K	27.77GB
Stheno-Mega-False-49B-L2.Q4_K_M.gguf	Q4_K_M	27.77GB
Stheno-Mega-False-49B-L2.Q4_1.gguf	Q4_1	28.85GB
Stheno-Mega-False-49B-L2.Q5_0.gguf	Q5_0	31.72GB
Stheno-Mega-False-49B-L2.Q5_K_S.gguf	Q5_K_S	31.72GB
Stheno-Mega-False-49B-L2.Q5_K.gguf	Q5_K	32.64GB
Stheno-Mega-False-49B-L2.Q5_K_M.gguf	Q5_K_M	32.64GB
Stheno-Mega-False-49B-L2.Q5_1.gguf	Q5_1	34.59GB
Stheno-Mega-False-49B-L2.Q6_K.gguf	Q6_K	37.82GB
Stheno-Mega-False-49B-L2.Q8_0.gguf	Q8_0	48.98GB

Original model description:

license: llama2 language: - en

... A False-49B Llama2 Model, made from... splicing models together with Mergekit.

Purely experimental and for the memes.

Better than 70Bs? No. Better than 34bs? Doubt.

better than 13b? maybe. hopefully.

Why does it exist? Because I don't know.

https://huggingface.co/Sao10K/Stheno-Mega-False-49B-L2-GGUF