Intel Arc Pro B70 Benchmarks: 52 tok/s with Qwen3.8-27B via vLLM XPU
No-Dot-Not · reddit · 2026-08-22
The author benchmarks the Intel Arc Pro B70 (32GB) running Qwen3.8-27B INT4 on Linux. Using vLLM XPU with MTP2 speculative decoding, it achieves 52.2 tok/s median decode speed at 64K context, outperforming llama.cpp SYCL by 1.8x. The test covers 128K context validation, vision capabilities, tool calling, and a real-world coding agent task. The post details current XPU stack bugs (e.g., concurrency limits with MTP) and concludes the B70 is a viable 32GB inference card with the right configuration.
More from Infra
- Muse Spark Open runs locally on a Mac M3 Ultra via llama.cpp: frontier model on desktop — pcuenq · 2026-10-16
- Muse-Spark open-weights drop: 816B-A42B MoE model gets 359GB MLX 3-bit build — ns123abc · 2026-10-16
- vLLM ships Day-0 LoRA RL for Muse Spark Open, lifting AIME from 0.56 to 0.85 — vllm_project · 2026-10-16
- DSpark draft model for Muse Spark Open proposes 7 tokens per pass with 23-42% acceptance — vllm_project · 2026-10-16
- Cloudflare Stream's WebRTC ultra-low-latency live streaming is now Generally Available — ritakozlov · 2026-10-16
- YC-backed Overshoot launches inference optimization, claims $5M saved for AI teams — ycombinator · 2026-10-16