RigBench data 2026-10-22

Home › Models › Qwen2.5 1.5B › Intel Arc Pro B70

Qwen2.5 1.5B on the Intel Arc Pro B70

Qwen2.5 1.5B runs at about 216 tokens/s on the Intel Arc Pro B70 (measured, Q4_K_M, SYCL, llama-bench).

Best measured generation216tok/s

Q4_K_M · SYCL · FA on

Prompt processing8,048tok/s

same run

For comparison, our size-scaling estimate for this combination is 381 tok/s.

Every run

QuantBackendGenerate tok/sPrompt tok/sTestBuild / by / notesSource
Q4_K_MSYCL FA on2168,048tg128ec6f7a6a5c PMZFX SYCL on one Arc Pro B70, context 4096, KV f16/f16; build b8840-12-gec6f7a6a5-dirty: the author's notes say the local change is a Q2_K DMMV work-in-progress patchPMZFX intel-arc-pro-b70-benchmarks (MIT)

tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128. "decode, llama-server" = single-stream generation speed measured against a running llama-server (no speculative decoding).

Check prices

Amazon Newegg

As an Amazon Associate I earn from qualifying purchases.

No card yet?

Try it on a rented GPU (RunPod)

Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.

All results for the Intel Arc Pro B70 · Qwen2.5 1.5B on other hardware

Embed this result

Qwen2.5 1.5B · Intel Arc Pro B70: 216 tok/s (Q4_K_M, measured)

HTML

<a href="https://rigbench.pages.dev/model/qwen2-5-1-5b/on/intel-arc-pro-b70/"><img src="https://rigbench.pages.dev/badge/qwen2-5-1-5b--intel-arc-pro-b70.svg" alt="Qwen2.5 1.5B · Intel Arc Pro B70: 216 tok/s (Q4_K_M, measured)"></a>

Markdown (GitHub README, forums)

[![Qwen2.5 1.5B · Intel Arc Pro B70: 216 tok/s (Q4_K_M, measured)](https://rigbench.pages.dev/badge/qwen2-5-1-5b--intel-arc-pro-b70.svg)](https://rigbench.pages.dev/model/qwen2-5-1-5b/on/intel-arc-pro-b70/)