M3 Ultra (80-core GPU, 512 GB)

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

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
M3 Ultra (80-core GPU, 512 GB)llama-3-1-8b-instruct Q4_K - Mediumllamafile62.675664777777771109.4627288333336–1.060675240722222––sourced–source
M3 Ultra (80-core GPU, 512 GB)llama-3-2-1b-instruct Q4_K - Mediumllamafile178.837921444444425600.99501711111–0.20634095138888886––sourced–source
M3 Ultra (80-core GPU, 512 GB)qwen-2-5-14b-instruct Q4_K - Mediumllamafile35.80064133333334577.4241827222222–2.0507156840555556––sourced–source

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

← all hardware