Llamacpp imatrix Quantizations of v6 Finch 7B HF Using llama.cpp release b3751 for quantization. Original model: https://huggingface.co/RWKV/v6 Finch 7B HF All quants made using imatrix option with dataset from here Run them in LM Studio Prompt format No prompt format found, check original model page What's new: Fix BOS/EOS tokens Download a file (not the whole branch) from below: Filename Quant type File Size Split Description v6 Finch 7B HF f16.gguf f16 15.51GB false Full F16 weights. v6 Finch 7B HF Q8 0.gguf Q8 0 8.47GB false Extremely high quality, generally unneeded but max available quant. v6 Finch 7B HF Q6 K L.gguf Q6 K L 6.78GB false Uses Q8 0 for embed and output weights. Very high quality, near perfect, recommended . v6 Finch 7B HF Q6 K.gguf Q6 K 6.65GB false Very high quality, near perfect, recommended . v6 Finch 7B HF Q5 K L.gguf Q5 K L 5.85GB false Uses Q8 0 for embed and output weights. High quality, recommended . v6 Finch 7B HF Q5 K M.gguf Q5 K M 5.68GB false High quality, recommended . v6 Finch 7B HF Q5 K S.gguf Q5 K S 5.68GB false High quality, recommended . v6 Finch 7B HF Q4 K L.gguf Q4 K L 4.98GB false Uses Q8 0 for embed and output weights. Good quality, recomme…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy