Ternary-Bonsai-2-27B

171 records across 25 hardware.

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
RTX 4070 Ti SUPERTernary-Bonsai-2-27B PTQ1_0llama.cpp (xyz-llama fork)156.3–––1638402026-09community–source
RTX 4070 Ti SUPERTernary-Bonsai-2-27B PTQ1_0llama.cpp (xyz-llama fork)144.2–––1638402026-09community–source
RTX 4070 Ti SUPERTernary-Bonsai-2-27B PTQ1_0llama.cpp (xyz-llama fork)141.3–––1638402026-09community–source
RTX 4070 Ti SUPERTernary-Bonsai-2-27B PTQ1_0llama.cpp (xyz-llama fork)139.5–––1638402026-09community–source
RTX 5090Ternary-Bonsai-2-27B Ternary PQ2_0llama.cpp (CUDA)129.93893––tg1282026-09-16community–source
RTX PRO 6000Ternary-Bonsai-2-27B Ternary PQ2_0llama.cpp (CUDA)124.84020––tg1282026-09-16community–source
RTX 5090Ternary-Bonsai-2-27B Ternary PTQ1_0llama.cpp (CUDA)120.51805––tg1282026-09-16community–source
RTX PRO 6000Ternary-Bonsai-2-27B Ternary PTQ1_0llama.cpp (CUDA)117.91972––tg1282026-09-16community–source
H100Ternary-Bonsai-2-27B Ternary PQ2_0llama.cpp (CUDA)113.92830––tg1282026-09-16community–source
RTX 4070Ternary-Bonsai-2-27B PTQ1_0llama.cpp (ada-surgery fork)1021173––40962026-10-08community–source
RTX 4090Ternary-Bonsai-2-27B Ternary PTQ1_0llama.cpp (CUDA)91.11645––tg1282026-09-16community–source
RTX 6000 AdaTernary-Bonsai-2-27B Ternary PTQ1_0llama.cpp (CUDA)90.41657––tg1282026-09-16community–source
H100Ternary-Bonsai-2-27B Ternary PTQ1_0llama.cpp (CUDA)86.91237––tg1282026-09-16community–source
RTX 4070Ternary-Bonsai-2-27B PTQ1_0llama.cpp (ada-surgery fork)831137––163842026-10-08community–source
RTX 6000 AdaTernary-Bonsai-2-27B Ternary PQ2_0llama.cpp (CUDA)82.82431––tg1282026-09-16community–source
L40STernary-Bonsai-2-27B Ternary PTQ1_0llama.cpp (CUDA)81.81543––tg1282026-09-16community–source
RTX 4090Ternary-Bonsai-2-27B Ternary PQ2_0llama.cpp (CUDA)81.23124––tg1282026-09-16community–source
RTX 4070Ternary-Bonsai-2-27B PTQ1_0llama.cpp (ada-surgery fork)781031––327682026-10-08community–source
L40STernary-Bonsai-2-27B Ternary PQ2_0llama.cpp (CUDA)74.42868––tg1282026-09-16community–source
A100Ternary-Bonsai-2-27B Ternary PQ2_0llama.cpp (CUDA)73.91328––tg1282026-09-16community–source
RTX 4070Ternary-Bonsai-2-27B PTQ1_0llama.cpp (ada-surgery fork)69867––655362026-10-08community–source
L4Ternary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)68.4––––2026-09community–source
L4Ternary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)67.8––––2026-09community–source
L4Ternary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)67.6––––2026-09community–source
RTX 4070Ternary-Bonsai-2-27B PTQ1_0llama.cpp (ada-surgery fork)61698––1146882026-10-08community–source
A100Ternary-Bonsai-2-27B Ternary PTQ1_0llama.cpp (CUDA)54.7706––tg1282026-09-16community–source
M5 MaxTernary-Bonsai-2-27B Ternary (pre-rotation build)llama.cpp (Metal)47.0765––tg1282026-09-16community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX44.8486.9––10242026-10-16community–source
RTX 5060 TiTernary-Bonsai-2-27B 27BPTQ1_0llama.cpp (CUDA)44.68464.81–––2026-09community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX44.3523.9––10242026-10-09community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX43.9487.5––10242026-10-17community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX43.6480.2––10242026-10-08community–source
L4Ternary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)42.9––––2026-09community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX42.8691.2––40962026-10-09community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX41.9634.5––40962026-10-16community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX41.3645.2––40962026-10-08community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX39.9589.8––40962026-10-17community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX39.7562.5––81922026-10-13community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX38.2663.5––81922026-10-08community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX38.2444.8––10242026-10-13communitycontradictionsource
M5 MaxTernary-Bonsai-2-27B 2bitoMLX37.5355.6––10242026-10-13communitycontradictionsource
M5 MaxTernary-Bonsai-2-27B 2bitoMLX37.4570.4––163842026-10-08community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX37.4477.0––40962026-10-13communitycontradictionsource
M4 MaxTernary-Bonsai-2-27B unknownoMLX36.2216.6––10242026-10-10community–source
M4 MaxTernary-Bonsai-2-27B unknownoMLX36.2216.6––10242026-10-10community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX35.9445.8––40962026-10-13communitycontradictionsource
M5 MaxTernary-Bonsai-2-27B unknownoMLX34.8552.5––163842026-10-13community–source
M3 MaxTernary-Bonsai-2-27B oQ2oMLX33.7218.6––10242026-10-16community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX32.5479.6––163842026-10-13communitycontradictionsource
L4Ternary-Bonsai-2-27B Ternary PTQ1_0llama.cpp (CUDA)32.1467––tg1282026-09-16community–source
L4Ternary-Bonsai-2-27B PTQ1_0llama.cpp (PrismML fork)31.6––––2026-09community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX31.5489.1––327682026-10-08community–source
M4 MaxTernary-Bonsai-2-27B unknownoMLX31.5173.1––40962026-10-10community–source
M4 MaxTernary-Bonsai-2-27B unknownoMLX31.5173.1––40962026-10-10community–source
L4Ternary-Bonsai-2-27B PTQ1_0llama.cpp (PrismML fork)31.4––––2026-09community–source
M2 UltraTernary-Bonsai-2-27B 2bitoMLX31.4263.9––10242026-10-15community–source
L4Ternary-Bonsai-2-27B PTQ1_0llama.cpp (PrismML fork)31.3––––2026-09community–source
L4Ternary-Bonsai-2-27B PTQ1_0llama.cpp (PrismML fork)30.6––––2026-09community–source
RTX 4070Ternary-Bonsai-2-27B PTQ1_0llama.cpp (ada-surgery fork)30649––1310722026-10-08community–source
L4Ternary-Bonsai-2-27B Ternary PQ2_0llama.cpp (CUDA)29.8777––tg1282026-09-16community–source
M3 MaxTernary-Bonsai-2-27B oQ2oMLX29.1232.7––81922026-10-11community–source
M2 UltraTernary-Bonsai-2-27B 2bitoMLX28.9262.1––81922026-10-15community–source
M5 ProTernary-Bonsai-2-27B Ternary (pre-rotation build)llama.cpp (Metal)28.7393––tg1282026-09-16community–source
M5 ProTernary-Bonsai-2-27B Ternary PQ2_0llama.cpp (Metal)28.138734.1–tg1282026-09-16community–source
M4 MaxTernary-Bonsai-2-27B unknownoMLX27.4155.5––163842026-10-10community–source
M4 MaxTernary-Bonsai-2-27B unknownoMLX27.4155.5––163842026-10-10community–source
M3 MaxTernary-Bonsai-2-27B oQ2oMLX26.4217.9––81922026-10-11community–source
M3 MaxTernary-Bonsai-2-27B 2bitoMLX26.1175.7––10242026-10-16community–source
M3 MaxTernary-Bonsai-2-27B oQ2oMLX25.5224.8––81922026-10-11communitycontradictionsource
M3 MaxTernary-Bonsai-2-27B 2bitoMLX24.7152.0––81922026-10-16community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX24.1363.5––655362026-10-08community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX23.7340.7––10242026-10-17community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX23.6286.0––10242026-10-17community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX23.6286.0––10242026-10-17community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX23.5279.4––10242026-10-09community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX23.4290.6––10242026-10-16community–source
M3 MaxTernary-Bonsai-2-27B oQ2oMLX23.2191.3––40962026-10-16community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX22.8336.2––40962026-10-09community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX22.7331.8––40962026-10-16community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX22.7344.9––40962026-10-17community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX22.6330.8––40962026-10-17community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX22.6330.8––40962026-10-17community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX22.3240.1––10242026-10-07community–source
M2 UltraTernary-Bonsai-2-27B 2bitoMLX21.4227.1––327682026-10-15community–source
M1 MaxTernary-Bonsai-2-27B 2bitoMLX21.1130.7––10242026-10-17community–source
M1 MaxTernary-Bonsai-2-27B 2bitoMLX21.1126.8––10242026-10-17community–source
M4 ProTernary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)20.7–––327682026-10community–source
M1 MaxTernary-Bonsai-2-27B 2bitoMLX20.4130.4––40962026-10-17community–source
M1 MaxTernary-Bonsai-2-27B 2bitoMLX20.4130.4––40962026-10-17community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX20.1278.9––40962026-10-07community–source
M5 MaxTernary-Bonsai-2-27B 2bitoMLX19.5283.3––655362026-10-11community–source
M3 MaxTernary-Bonsai-2-27B 2bitoMLX18.1126.4––40962026-10-06community–source
M4 ProTernary-Bonsai-2-27B Ternary (pre-rotation build)llama.cpp (Metal)18.0125––tg1282026-09-16community–source
RTX 4070Ternary-Bonsai-2-27B PTQ1_0llama.cpp (ada-surgery fork)17.9547––1843202026-10-08community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX17.4242.7––10242026-10-18communitycontradictionsource
M3 MaxTernary-Bonsai-2-27B 2bitoMLX16.9117.5––163842026-10-06community–source
M3 MaxTernary-Bonsai-2-27B 2bitoMLX16.8100.2––10242026-10-16communitycontradictionsource
M3 MaxTernary-Bonsai-2-27B 2bitoMLX16.5112.0––81922026-10-16communitycontradictionsource
M5 ProTernary-Bonsai-2-27B 2bitoMLX16.5250.1––40962026-10-18communitycontradictionsource
M5 MaxTernary-Bonsai-2-27B unknownoMLX16.4236.9––1310722026-10-13community–source
M1 MaxTernary-Bonsai-2-27B 2bitoMLX16.197.4––81922026-10-09community–source
M6Ternary-Bonsai-2-27B 2bitoMLX13.5194.3––40962026-10-11community–source
M6Ternary-Bonsai-2-27B 2bitoMLX13.5194.3––40962026-10-11community–source
M5Ternary-Bonsai-2-27B 2bitoMLX12.2155.1––10242026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX12.2155.1––10242026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX12.0154.7––10242026-10-08community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.9155.8––10242026-10-08community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.8158.1––40962026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.8158.1––40962026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.7152.2––10242026-10-08community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.7154.5––40962026-10-08community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.7155.5––40962026-10-08community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.7153.0––40962026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.6157.0––40962026-10-08community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.6152.2––40962026-10-17community–source
M3 ProTernary-Bonsai-2-27B 2bitoMLX11.386.6––10242026-10-09community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.1149.0––40962026-10-17community–source
M3 ProTernary-Bonsai-2-27B 2bitoMLX11.085.4––10242026-10-09community–source
M3 ProTernary-Bonsai-2-27B 2bitoMLX11.086.0––40962026-10-09community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.0156.2––40962026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX11.0152.3––40962026-10-17community–source
M1 ProTernary-Bonsai-2-27B 2bitoMLX10.965.3––40962026-10-11community–source
M5Ternary-Bonsai-2-27B 2bitoMLX10.9149.2––40962026-10-17community–source
M6Ternary-Bonsai-2-27B 2bitoMLX10.8159.7––163842026-10-11community–source
M6Ternary-Bonsai-2-27B 2bitoMLX10.8159.7––163842026-10-11community–source
M5Ternary-Bonsai-2-27B 2bitoMLX10.8153.1––81922026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX10.8153.1––81922026-10-17community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX10.7144.6––40962026-10-15communitycontradictionsource
M3 ProTernary-Bonsai-2-27B 2bitoMLX10.781.9––10242026-10-16community–source
M3 ProTernary-Bonsai-2-27B 2bitoMLX10.684.5––40962026-10-09community–source
M1 MaxTernary-Bonsai-2-27B 2bitoMLX10.6104.4––327682026-10-10community–source
M3 ProTernary-Bonsai-2-27B 2bitoMLX10.481.5––10242026-10-15community–source
M5Ternary-Bonsai-2-27B 2bitoMLX10.1126.7––10242026-10-08communitycontradictionsource
M5Ternary-Bonsai-2-27B 2bitoMLX10.1145.6––40962026-10-17communitycontradictionsource
M5Ternary-Bonsai-2-27B 2bitoMLX9.9149.1––40962026-10-17communitycontradictionsource
M5Ternary-Bonsai-2-27B 2bitoMLX9.6123.8––40962026-10-08communitycontradictionsource
M3 ProTernary-Bonsai-2-27B 2bitoMLX9.683.8––40962026-10-16communitycontradictionsource
M1 ProTernary-Bonsai-2-27B 2bitoMLX9.557.1––81922026-10-11community–source
M5Ternary-Bonsai-2-27B 2bitoMLX9.4129.0––163842026-10-08community–source
M5Ternary-Bonsai-2-27B 2bitoMLX9.3144.5––40962026-10-17communitycontradictionsource
M3 ProTernary-Bonsai-2-27B 2bitoMLX9.284.3––40962026-10-15communitycontradictionsource
M2 ProTernary-Bonsai-2-27B 2bitoMLX9.059.5––10242026-10-12community–source
M5 ProTernary-Bonsai-2-27B 2bitoMLX9.0131.6––163842026-10-15community–source
M5Ternary-Bonsai-2-27B 2bitoMLX9.0148.2––40962026-10-17communitycontradictionsource
M4Ternary-Bonsai-2-27B 2bitoMLX8.855.9––10242026-10-09community–source
M2 ProTernary-Bonsai-2-27B 2bitoMLX8.858.0––40962026-10-12community–source
M4Ternary-Bonsai-2-27B 2bitoMLX8.744.9––10242026-10-16community–source
M5Ternary-Bonsai-2-27B 2bitoMLX8.7141.3––163842026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX8.7141.3––163842026-10-17community–source
RTX 4070Ternary-Bonsai-2-27B PTQ1_0llama.cpp (ada-surgery fork)8.6436––2621442026-10-08community–source
M2 ProTernary-Bonsai-2-27B 2bitoMLX8.256.6––81922026-10-12community–source
M4Ternary-Bonsai-2-27B 2bitoMLX8.254.1––10242026-10-13community–source
RTX 5060 TiTernary-Bonsai-2-27B 27BPTQ1_0llama.cpp (CUDA)8.121.87––1548552026-09community–source
M4Ternary-Bonsai-2-27B 2bitoMLX8.155.6––40962026-10-09community–source
M4Ternary-Bonsai-2-27B 2bitoMLX8.153.9––40962026-10-13community–source
M4 ProTernary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)7.96–––327682026-10community–source
M4 ProTernary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)7.92–––327682026-10community–source
M4 ProTernary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)7.88–––327682026-10community–source
M2 ProTernary-Bonsai-2-27B 2bitoMLX7.753.1––163842026-10-12community–source
M3Ternary-Bonsai-2-27B 2bitoMLX7.446.6––10242026-10-12community–source
M3Ternary-Bonsai-2-27B 2bitoMLX7.446.8––10242026-10-12community–source
M4 ProTernary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)7.36–––327682026-10community–source
M1 MaxTernary-Bonsai-2-27B oQ6oMLX7.364.8––163842026-10-09community–source
M1 ProTernary-Bonsai-2-27B 2bitoMLX7.251.4––163842026-10-11community–source
M4 ProTernary-Bonsai-2-27B PQ2_0llama.cpp (PrismML fork)7.1–––327682026-10community–source
M3 ProTernary-Bonsai-2-27B 2bitoMLX7.086.7––10242026-10-09communitycontradictionsource
M3 ProTernary-Bonsai-2-27B 2bitoMLX6.386.2––40962026-10-09communitycontradictionsource
M5Ternary-Bonsai-2-27B 2bitoMLX6.3126.8––327682026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX6.3126.8––327682026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX3.7103.3––655362026-10-17community–source
M5Ternary-Bonsai-2-27B 2bitoMLX3.7103.3––655362026-10-17community–source

← all models