Llamacpp imatrix Quantizations of aya 23 8B Using llama.cpp release b2965 for quantization. Original model: https://huggingface.co/CohereForAI/aya 23 8B All quants made using imatrix option with dataset from here Prompt format Download a file (not the whole branch) from below: Filename Quant type File Size Description aya 23 8B Q8 0.gguf Q8 0 8.54GB Extremely high quality, generally unneeded but max available quant. aya 23 8B Q6 K.gguf Q6 K 6.59GB Very high quality, near perfect, recommended . aya 23 8B Q5 K M.gguf Q5 K M 5.80GB High quality, recommended . aya 23 8B Q5 K S.gguf Q5 K S 5.66GB High quality, recommended . aya 23 8B Q4 K M.gguf Q4 K M 5.05GB Good quality, uses about 4.83 bits per weight, recommended . aya 23 8B Q4 K S.gguf Q4 K S 4.82GB Slightly lower quality with more space savings, recommended . aya 23 8B IQ4 NL.gguf IQ4 NL 4.81GB Decent quality, slightly smaller than Q4 K S with similar performance recommended . aya 23 8B IQ4 XS.gguf IQ4 XS 4.60GB Decent quality, smaller than Q4 K S with similar performance, recommended . aya 23 8B Q3 K L.gguf Q3 K L 4.52GB Lower quality but usable, good for low RAM availability. aya 23 8B Q3 K M.gguf Q3 K M 4.22GB Even lower qualit…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy