No machine on this network has measured this model yet. Everything below is what the catalog knows exists — where the weights come from, how large each build is, and whose work it is. Nothing here is a claim about what the model can do; that only appears once a machine has run it and signed the result.
hf.co/unsloth/Mistral-Medium-3.5-128B-GGUF:IQ2_XXSollama · IQ2_XXS · 32.5 GiB · quantizer (unsloth)
hf.co/unsloth/Mistral-Medium-3.5-128B-GGUF:IQ2_Mollama · IQ2_M · 41.1 GiB · quantizer (unsloth)
hf.co/unsloth/Mistral-Medium-3.5-128B-GGUF:IQ3_XXSollama · IQ3_XXS · 45.9 GiB · quantizer (unsloth)
bartowski/mistralai_Mistral-Medium-3.5-128B-GGUF:Q2_K_Lllamacpp · Q2_K_L · 47.9 GiB · quantizer (bartowski)
unsloth/Mistral-Medium-3.5-128B-GGUF:Q3_K_Sllamacpp · Q3_K_S · 50.6 GiB · quantizer (unsloth)
majentik/Mistral-Medium-3.5-128B-MLX-2bitmlx · 2BIT · 51.1 GiB · quantizer (majentik)
unsloth/Mistral-Medium-3.5-128B-GGUF:IQ4_XSllamacpp · IQ4_XS · 62.5 GiB · quantizer (unsloth)
unsloth/Mistral-Medium-3.5-128B-GGUF:Q4_0llamacpp · Q4_0 · 66.1 GiB · quantizer (unsloth)
hf.co/mradermacher/Mistral-Medium-3.5-128B-GGUF:Q4_K_Sollama · Q4_K_S · 66.4 GiB · quantizer (mradermacher)
majentik/Mistral-Medium-3.5-128B-MLX-MXFP4mlx · 4BIT · 69.1 GiB · quantizer (majentik)
hf.co/mradermacher/Mistral-Medium-3.5-128B-GGUF:Q4_K_Mollama · Q4_K_M · 69.8 GiB · quantizer (mradermacher)
unsloth/Mistral-Medium-3.5-128B-GGUF:Q4_K_Mllamacpp · Q4_K_M · 69.8 GiB · quantizer (unsloth)
zdy1995love/Mistral-Medium-3.5-128B-NVFP4vllm · W4A16 · 74.8 GiB · quantizer (zdy1995love)
NeuralNotwerk/Mistral-Medium-3.5-128B-NVFP4vllm · W4A4 · 74.8 GiB · quantizer (NeuralNotwerk)
nvidia/Mistral-Medium-3.5-128B-NVFP4vllm · NVFP4 · 88.7 GiB · quantizer (nvidia)
unsloth/Mistral-Medium-3.5-128B-GGUF:Q8_0llamacpp · Q8_0 · 123.7 GiB · quantizer (unsloth)
unsloth/Mistral-Medium-3.5-128B-GGUF:BF16llamacpp · BF16 · 232.9 GiB · quantizer (unsloth)
mistralai/Mistral-Medium-3.5-128Bvllm · FP8 · 248.9 GiB · vendor (mistralai)