Qwen3-0.6B

14 records across 5 hardware.

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
RTX 3070Qwen3-0.6B unspecifiedvllm1676–––5122026-09-30community–source
RTX 3070Qwen3-0.6B unspecifiedvllm1671–––5122026-09-30community–source
B70 Arc ProQwen3-0.6B Q4_K_MVulkan475.7620217.98––5122026-10-02community–source
B70 Arc ProQwen3-0.6B Q4_K_MVulkan467.9617508.58––5122026-10-02community–source
B70 Arc ProQwen3-0.6B Q4_K_MVulkan456.9425346.93––5122026-10-02community–source
B70 Arc ProQwen3-0.6B Q4_K_MVulkan452.7021235.41––5122026-10-02community–source
RTX 4060 TiQwen3-0.6B Q4_K_MVulkan396.1328280.84––10242026-10-02community–source
RTX 4060 TiQwen3-0.6B Q4_K_MVulkan390.3324067.88––10242026-10-02community–source
RTX 4060 TiQwen3-0.6B Q4_K_MVulkan378.4416094.75––10242026-10-02community–source
RTX 4060 TiQwen3-0.6B Q4_K_MVulkan376.9217654.48––10242026-10-02community–source
M3 UltraQwen3-0.6B 8bitoMLX118.64629.0––327682026-10-18community–source
M2Qwen3-0.6B 8bitoMLX86.81749.0––10242026-10-06community–source
M3 UltraQwen3-0.6B unknownoMLX77.94278.0––327682026-10-08community–source
M2Qwen3-0.6B 8bitoMLX60.71317.0––40962026-10-06community–source

← all models