Qwen3.6 27B

Q6_K·27B params·GGUF
reasoning
checkpoint: unsloth/Qwen3.6-27B-GGUF:Q6_K
commit: 82d411acf4a0
weights 20.98 GiB

All runs (19)

HardwareBackendModeShapeConc.Gen tok/sPrefill tok/sTTFTTPOT (ms)Prompt tokOut tokTotalVRAM Δ
GeForce RTX 3090 · 24 GiB450 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-450wcodegen1
32.0
188.1371ms30.162100031.29s0.000 GiB
GeForce RTX 3090 · 24 GiB350 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-350wcodegen1
31.5
206.3326ms30.862100031.76s0.000 GiB
GeForce RTX 3090 · 24 GiB450 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-450wagent1
31.3
1124.6533ms30.259950015.96s0.000 GiB
GeForce RTX 3090 · 24 GiB350 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-350wagent1
30.8
1173.2511ms30.859950016.23s0.000 GiB
GeForce RTX 3090 · 24 GiB450 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-450wchat1
30.4
126.1238ms29.7301003.29s0.000 GiB
GeForce RTX 3090 · 24 GiB350 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-350wchat1
29.7
119.7251ms30.4301003.36s0.000 GiB
GeForce RTX 3090 · 24 GiB450 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-450wrag1
29.0
1105.0789ms30.08422006.90s0.000 GiB
GeForce RTX 3090 · 24 GiB350 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-350wrag1
28.5
1211.1804ms30.78422007.03s0.000 GiB
GeForce RTX 3090 · 24 GiB200 Wdrv 590
llama.cpp 59778f0 (cuda)baselinecodegen1
15.5
177.7364ms64.462100064.43s0.000 GiB
GeForce RTX 3090 · 24 GiB200 Wdrv 590
llama.cpp 59778f0 (cuda)baselinechat1
15.3
107.2281ms61.5301006.54s0.000 GiB
GeForce RTX 3090 · 24 GiB200 Wdrv 590
llama.cpp 59778f0 (cuda)baselineagent1
15.0
616.91.09s64.559950033.25s0.000 GiB
GeForce RTX 3090 · 24 GiB200 Wdrv 590
llama.cpp 59778f0 (cuda)baselinerag1
14.4
707.51.25s63.384220013.92s0.000 GiB
GeForce RTX 3090 · 24 GiB450 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-450wagent4
13.1
25.724.23s30.259950039.73s0.000 GiB
GeForce RTX 3090 · 24 GiB350 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baseline-pl-350wagent4
12.8
30.524.79s30.859950040.59s0.000 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unifieddrv 7
llama.cpp rocm-4f13cb7 (rocm)baselinecodegen1
9.3
129.9486ms106.5621000106.95s0.017 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unifieddrv 7
llama.cpp rocm-4f13cb7 (rocm)baselineagent1
9.3
2133.1281ms106.659950053.55s0.010 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unifieddrv 7
llama.cpp rocm-4f13cb7 (rocm)baselinechat1
9.1
77.7387ms106.33010010.93s0.004 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unifieddrv 7
llama.cpp rocm-4f13cb7 (rocm)baselinerag1
8.7
463.51.66s106.584220022.88s0.007 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unifieddrv 7
llama.cpp rocm-4f13cb7 (rocm)baselineagent4
3.9
10.580.46s106.5599500133.69s0.037 GiB

Environment

GeForce RTX 3090 · 24 GiB
cpuAMD EPYC 7302P 16-Core Processor
gpuNVIDIA GeForce RTX 3090
archNVIDIA
vram24 GiB (system 64.0 GiB)
power350 W / 450 W max(78% cap)
pcieGen 4 x16 / Gen 4 x16 max
clocksgfx 1965/2100 MHz · mem 9501 MHz
temp44°C idle · 64°C peak
peak draw336 W
backendllama.cpp cuda-4f13cb7 (cuda)
serverlemonade unknown
osUbuntu 24.04 LTS
kernel6.17.13-7-pve
driverNVIDIA 590.48.01 + CUDA 13.1
libc2.39
python3.12.3
containerizedtrue
llama.cppversion: 18 (4f13cb7) built with GNU 13.3.0 for Linux x86_64
build flagsGGML_CUDA=ON CMAKE_BUILD_TYPE=Release
runs/cell5
warmups2
endpoint/v1/chat/completions
streamingtrue
GeForce RTX 3090 · 24 GiB
cpuAMD EPYC 7302P 16-Core Processor
gpuNVIDIA GeForce RTX 3090
archNVIDIA
vram24 GiB (system 64.0 GiB)
power450 W / 450 W max
pcieGen 4 x16 / Gen 4 x16 max
clocksgfx 1965/2100 MHz · mem 9501 MHz
temp43°C idle · 83°C peak
peak draw437 W
backendllama.cpp cuda-4f13cb7 (cuda)
serverlemonade unknown
osUbuntu 24.04 LTS
kernel6.17.13-7-pve
driverNVIDIA 590.48.01 + CUDA 13.1
libc2.39
python3.12.3
containerizedtrue
llama.cppversion: 18 (4f13cb7) built with GNU 13.3.0 for Linux x86_64
build flagsGGML_CUDA=ON CMAKE_BUILD_TYPE=Release
runs/cell5
warmups2
endpoint/v1/chat/completions
streamingtrue
GeForce RTX 3090 · 24 GiB
cpuAMD EPYC 7302P 16-Core Processor
gpuNVIDIA GeForce RTX 3090
archNVIDIA
vram24 GiB (system 64.0 GiB)
power200 W / 450 W max(44% cap)
backendllama.cpp 59778f0 (cuda)
serverlemonade unknown
osUbuntu 24.04 LTS
kernel6.17.13-7-pve
driver590.48.01
python3.12.3
containerizedtrue
runs/cell5
warmups2
endpoint/v1/chat/completions
streamingtrue
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)
cpuAMD RYZEN AI MAX+ 395 w/ Radeon 8060S
gpuAMD Radeon 8060S
archStrix Halo (gfx1151)
vram96 GiB (system 31.1 GiB, unified)
pcieGen 4 x16 / Gen 4 x16 max
clocksgfx 1007 MHz · mem 1000 MHz
temp46°C idle · 76°C peak
peak draw99 W
backendllama.cpp rocm-4f13cb7 (rocm)
serverlemonade unknown
osUbuntu 24.04 LTS
kernel7.0.2-2-pve
driverROCm 7.2.3
libc2.39
python3.12.3
containerizedtrue
llama.cppversion: 1 (4f13cb7) built with Clang 22.0.0 for Linux x86_64
build flagsGGML_HIP=ON AMDGPU_TARGETS=gfx1151 CMAKE_BUILD_TYPE=Release
runs/cell5
warmups2
endpoint/v1/chat/completions
streamingtrue