M1 Ultra (64-core GPU, 128 GB)

3 records across 3 models (quants: Q4_K - Medium).

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
M1 Ultra (64-core GPU, 128 GB)llama-3-1-8b-instruct Q4_K - Mediumllamafile54.30790400000001667.5689454444446–1.6929770138888889––sourced–source
M1 Ultra (64-core GPU, 128 GB)llama-3-2-1b-instruct Q4_K - Mediumllamafile151.06786977777782977.086063277778–0.3675812592777778––sourced–source
M1 Ultra (64-core GPU, 128 GB)qwen-2-5-14b-instruct Q4_K - Mediumllamafile32.417089222222224371.7215576111111–3.126813851888889––sourced–source

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

← all hardware