M1 Ultra (48-core GPU, 128 GB)

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

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
M1 Ultra (48-core GPU, 128 GB)llama-3-1-8b-instruct Q4_K - Mediumllamafile48.89361162962964534.3503048148149–2.1556520740370373––sourced–source
M1 Ultra (48-core GPU, 128 GB)llama-3-2-1b-instruct Q4_K - Mediumllamafile137.970232703703722582.382230222222–0.4251138565185186––sourced–source
M1 Ultra (48-core GPU, 128 GB)qwen-2-5-14b-instruct Q4_K - Mediumllamafile27.773420277777777289.8969081111111–4.044066250166667––sourced–source

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

← all hardware