With Power Limits 95/110, Ubuntu 24.04, Intel Core Ultra 9 185H,

Crucial 128GB (2x64GB) 5600MT/s DDR5 SODIMM, WD_BLACK SN850x 8TB

GPU Version: intel-ollama-0.6.2 GPU SYCL0 (Intel(R) Arc(TM) Graphics) - 120187 MiB

CPU Version: ollama version is 0.7.0 CPU 123.5 GiB available

Modelprompt eval ratesecond prompteval rate
NameParams, BSIZE, GBCPUGPUCPUGPUCPUGPU
deepseek-r1:70b70422.630.8912.3914.131.241.38
deepseek-v2:16b168.959.4715.83361.51175.0224.3912.00
qwen3:32b 32205.672.8438.8841.602.532.73
llama4:scout176711.14-1683.33-4.81-
llama3.3:70b70422.601.5421.3523.371.251.37
tinyllama:1.1b1.10.6198.18112.982595.122211.2162.9957.53
gemma3:27b27176.60-49.38-3.04-
gemma3:12b128.115.73-109.35-6.55-
gemma3:4b43.348.55-284.47-17.88-
gemma3:1b10.8140.40-177.24-45.26-
orca-mini:70b
70383.28-62.05-1.24-
orca-mini:13b
137.417.27-307.94-6.23-
orca-mini:7b
73.833.20-576.83-11.76-
orca-mini:3b
3
2.0
68.21-1049.24-21.91-
phi4:14b-fp16142912.4511.5335.5340.162.082.09
phi4:14b-q4_K_M149.112.989.87100.02113.455.786.77
phi4-mini:3.8b-q4_K_M3.82.549.26-186.63-18.83-
dolphin-phi:2.7b2.71.685.6786.81744.07649.4325.4221.73
dolphin3:8b3.84.926.0430.97325.85373.3010.7612.58
openthinker:32b-v2-fp1632654.457.3120.2422.310.810.90
openthinker:32b32195.934.5966.3169.742.602.78
mistral-small3.1:24b24157.71-1321.32-3.64-
openchat:7b74.130.4727.15273.39361.0411.1014.81