Hardware · CPU · Intel

Intel Core Ultra 7 155H

Runs models from system RAM. Slower than a GPU, but not limited by graphics memory.

What it runs

Largest models without offloading on the Windows laptop 16 GB (no graphics card), 8K context.

Variant
Qwen2.5 14B Instruct
14.8B · Q4_K_M via llama.cpp
Runs well
~6 tok/s estimate
Variant
Phi-4
14.7B · Q4_K_M via llama.cpp
Runs well
~6 tok/s estimate
Variant
Gemma 2 9B IT
9.24B · Q4_K_M via llama.cpp
Runs well
~9 tok/s estimate
Variant
Hermes 3 Llama 3.1 8B
8.03B · Q8_0 via llama.cpp
Runs well
~6 tok/s estimate
Variant
Llama 3.1 8B Instruct
8.03B · Q8_0 via llama.cpp
Runs well
~6 tok/s estimate
Variant
Qwen2.5 7B Instruct
7.62B · Q8_0 via llama.cpp
Runs well
~7 tok/s estimate
Variant
DeepSeek-R1-Distill-Qwen-7B
7.62B · Q4_K_M via llama.cpp
Runs well
~11 tok/s estimate
Variant
Mistral 7B Instruct v0.3
7.25B · Q8_0 via llama.cpp
Runs well
~7 tok/s estimate

System Windows laptop 16 GB (no graphics card)
By estimate, runs models up to ~18B parameters at 4-bit in system RAM (CPU only). What runs

Throughput on systems containing this device.

No performance measurements recorded yet.

Contributions are not open yet, so there is nothing here from members.

Contributions are not open yet, so there is nothing here from members.

No source records linked.