Gemma-4 26B-A4B-it

Q4_K_M·26B params·256K ctx·GGUF
visiontool-callinghottool-callingvisionllamacpp
checkpoint: unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q4_K_M
commit: b68961b3c96e
weights 16.82 GiB · on-disk 16.90 GiB

All runs (20)

HardwareBackendModeShapeConc.Gen tok/sPrefill tok/sTTFTTPOT (ms)Prompt tokOut tokTotalVRAM Δ
GeForce RTX 3090 · 24 GiB300 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baselinecodegen1
109.6
313.2235ms8.57110009.12s0.000 GiB
GeForce RTX 3090 · 24 GiB300 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baselinechat1
104.1
370.0103ms8.436100960ms0.000 GiB
GeForce RTX 3090 · 24 GiB300 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baselineagent1
100.1
1229.9426ms8.66185005.00s0.000 GiB
GeForce RTX 3090 · 24 GiB300 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baselinerag1
84.9
1004.8634ms8.68532002.35s0.000 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unified
llama.cpp b8940 (vulkan)baselinecodegen1
47.9
277.2296ms20.771100020.88s0.000 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unified
llama.cpp b8940 (vulkan)baselinechat1
47.7
151.7244ms19.2371002.10s0.001 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unified
llama.cpp b8940 (vulkan)baselineagent1
44.8
744.6712ms21.161850011.15s0.000 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unified
llama.cpp b8940 (vulkan)baselinerag1
43.2
1212.2590ms21.210122004.63s0.000 GiB
GeForce RTX 3090 · 24 GiB300 Wdrv 590
llama.cpp cuda-4f13cb7 (cuda)baselineagent4
42.4
85.07.71s8.661850012.21s0.000 GiB
Strix Halo · Radeon 8060S · 128 GiB unified (96 GiB VRAM)unified
llama.cpp b8940 (vulkan)baselineagent4
18.3
84.67.29s40.861850027.38s0.000 GiB

Environment

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)
backendllama.cpp b8940 (cpu)
serverlemonade 10.4.0
osUbuntu 24.04.4 LTS
kernel7.0.2-2-pve
python3.12.3
containerizedtrue
runs/cell3
warmups1
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)
power300 W / 450 W max(67% cap)
pcieGen 4 x16 / Gen 4 x16 max
clocksgfx 1920/2100 MHz · mem 9501 MHz
temp52°C idle · 67°C peak
peak draw295 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
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)
backendllama.cpp b1203 (rocm)
serverlemonade 10.4.0
osUbuntu 24.04.4 LTS
kernel7.0.2-2-pve
python3.12.3
containerizedtrue
runs/cell3
warmups1
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)
backendllama.cpp b8940 (vulkan)
serverlemonade 10.4.0
osUbuntu 24.04.4 LTS
kernel7.0.2-2-pve
python3.12.3
containerizedtrue
runs/cell3
warmups1
endpoint/v1/chat/completions
streamingtrue