Llamacpp imatrix Quantizations of Qwen2.5 72B Instruct Using llama.cpp release b3772 for quantization. Original model: https://huggingface.co/Qwen/Qwen2.5 72B Instruct All quants made using imatrix option with dataset from here Run them in LM Studio Prompt format What's new: Update context length settings and tokenizer Download a file (not the whole branch) from below: Filename Quant type File Size Split Description Qwen2.5 72B Instruct Q8 0.gguf Q8 0 77.26GB true Extremely high quality, generally unneeded but max available quant. Qwen2.5 72B Instruct Q6 K.gguf Q6 K 64.35GB true Very high quality, near perfect, recommended . Qwen2.5 72B Instruct Q5 K M.gguf Q5 K M 54.45GB true High quality, recommended . Qwen2.5 72B Instruct Q4 K M.gguf Q4 K M 47.42GB false Good quality, default size for must use cases, recommended . Qwen2.5 72B Instruct Q4 0.gguf Q4 0 41.38GB false Legacy format, generally not worth using over similarly sized formats Qwen2.5 72B Instruct Q3 K XL.gguf Q3 K XL 40.60GB false Uses Q8 0 for embed and output weights. Lower quality but usable, good for low RAM availability. Qwen2.5 72B Instruct IQ4 XS.gguf IQ4 XS 39.71GB false Decent quality, smaller than Q4 K S with sim…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy