Llamacpp imatrix Quantizations of Qwen2.5 3B Instruct Using llama.cpp release b3772 for quantization. Original model: https://huggingface.co/Qwen/Qwen2.5 3B Instruct All quants made using imatrix option with dataset from here Run them in LM Studio Prompt format What's new: Update tokenizer Download a file (not the whole branch) from below: Filename Quant type File Size Split Description Qwen2.5 3B Instruct f16.gguf f16 6.18GB false Full F16 weights. Qwen2.5 3B Instruct Q8 0.gguf Q8 0 3.29GB false Extremely high quality, generally unneeded but max available quant. Qwen2.5 3B Instruct Q6 K L.gguf Q6 K L 2.61GB false Uses Q8 0 for embed and output weights. Very high quality, near perfect, recommended . Qwen2.5 3B Instruct Q6 K.gguf Q6 K 2.54GB false Very high quality, near perfect, recommended . Qwen2.5 3B Instruct Q5 K L.gguf Q5 K L 2.30GB false Uses Q8 0 for embed and output weights. High quality, recommended . Qwen2.5 3B Instruct Q5 K M.gguf Q5 K M 2.22GB false High quality, recommended . Qwen2.5 3B Instruct Q5 K S.gguf Q5 K S 2.17GB false High quality, recommended . Qwen2.5 3B Instruct Q4 K L.gguf Q4 K L 2.01GB false Uses Q8 0 for embed and output weights. Good quality, recommende…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy