Llamacpp imatrix Quantizations of Qwen3 30B A3B by Qwen Using llama.cpp release b5200 for quantization. Original model: https://huggingface.co/Qwen/Qwen3 30B A3B All quants made using imatrix option with dataset from here (augmented with extra synthetic data I will post shortly) Run them in LM Studio Run them directly with llama.cpp, or any other llama.cpp based project Prompt format What's new: Improved imatrix dataset Download a file (not the whole branch) from below: Filename Quant type File Size Split Description Qwen3 30B A3B bf16.gguf bf16 61.10GB true Full BF16 weights. Qwen3 30B A3B Q8 0.gguf Q8 0 32.48GB false Extremely high quality, generally unneeded but max available quant. Qwen3 30B A3B Q6 K L.gguf Q6 K L 25.26GB false Uses Q8 0 for embed and output weights. Very high quality, near perfect, recommended . Qwen3 30B A3B Q6 K.gguf Q6 K 25.10GB false Very high quality, near perfect, recommended . Qwen3 30B A3B Q5 K L.gguf Q5 K L 21.94GB false Uses Q8 0 for embed and output weights. High quality, recommended . Qwen3 30B A3B Q5 K M.gguf Q5 K M 21.74GB false High quality, recommended . Qwen3 30B A3B Q5 K S.gguf Q5 K S 21.10GB false High quality, recommended . Qwen3 30B A3B Q4…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy