RTX 3060 12GB

6 records across 3 models (quants: Q4_K, Q4_K_M).

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
RTX 3060 12GBgemma4 12B Q4_KM Q4_K_Mllama.cpp (CUDA)33.34609.26–––2026-06-09community–source
RTX 3060 12GBQwen3 14B 14BQ4_Kllama.cpp31.2972.6––4096–sourced–source
RTX 3060 12GBQwen3 14B 14BQ4_Kllama.cpp22.7678.2––16384–sourced–source
RTX 3060 12GBQwen3 8B 8BQ4_Kllama.cpp55.21696.8––4096–sourced–source
RTX 3060 12GBQwen3 8B 8BQ4_Kllama.cpp42.01119.2––16384–sourced–source
RTX 3060 12GBQwen3 8B 8BQ4_Kllama.cpp31.9764.7––32768–sourced–source

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

← all hardware