M4 Max (40-core GPU, 64 GB)

14 records across 6 models (quants: Q4, Q4_G32, Q4_K - Medium, Q5, Q5_G32, Q6, Q8).

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
M4 Max (40-core GPU, 64 GB)llama-3-1-8b-instruct Q4_K - Mediumllamafile47.095953944444446557.1200590555555–2.1445155647222225––sourced–source
M4 Max (40-core GPU, 64 GB)llama-3-2-1b-instruct Q4_K - Mediumllamafile180.240404333333373857.499544666666–0.30436414811111107––sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 30B-A3B Q4MLX92.09822.59––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 30B-A3B Q5MLX84.89819.83––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 30B-A3B Q6MLX76.74817.63––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 30B-A3B Q8MLX52.58772.59––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 32B Q4_K - MediumOllama22–––128000–community–source
M4 Max (40-core GPU, 64 GB)Qwen 3 4B Q4_G32MLX149.072838.42––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 4B Q4MLX148.082976.68––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 4B Q5MLX143.22736.15––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 4B Q5_G32MLX142.982754.53––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 4B Q6MLX136.582735.69––2048–sourced–source
M4 Max (40-core GPU, 64 GB)Qwen 3 4B Q8MLX111.551780.63––2048–sourced–source
M4 Max (40-core GPU, 64 GB)qwen-2-5-14b-instruct Q4_K - Mediumllamafile25.874467866666663286.74860906666663–4.498645011088889––sourced–source

All rows cite their source. Provenance: community 40018 sourced 1085

← all hardware