Intel Arc Pro B70

GPU

1

Run records

1

Models covered

1

Frameworks covered

L1 Reproduced × 1

Evidence mix

Typical measured performance

Published benchmark averages per model × framework × quantization on this hardware.

  1. llama.cpp · Q4_K_M Decode 35.6 · Prefill 562 ·

Fit check

Physics-based estimate from VRAM and memory bandwidth: pick a model to see which quantizations fit and the expected speed

Formula: tok/s = (bandwidth ÷ model size) × 0.55, adapted from llmfit; see run records for measured numbers

Key specs

Vendor
Intel
VRAM
32 GB
Memory bandwidth
552 GB/s

Run records

1 configurations

ModelFrameworkQuantizationDecodeGen timeSamples
Qwen3.8-27Bllama.cppQ4_K_M35.6 tok/s—1