XeMaya alpha: open-source Intel Arc inference engine runs DeepSeek V4.1 on 88GB of Arc GPUs
Jumbomuffin1312 · reddit · 2026-11-11
First public alpha of XeMaya, an open-source inference engine built on XeStrata using native SYCL, Level Zero and Intel XMX acceleration, bringing DeepSeek-V4.1-Flash, GLM-5.3-Flash and Qwen3.8-Flash-Next support to Intel Arc GPUs. On a 3-GPU Arc Pro rig (88GB VRAM): DeepSeek V4.1 Flash generation improved from 10.9 to 14.56 tok/s (two GPUs beat three due to a slow chipset link), Flash Coder prefill jumped from 48.5 to 224.8 tok/s via chunked prefill and XMX attention, and GLM 5.3 Flash hit 152–161 tok/s prefill (20 tok/s with MTP), though Strata still leads on Qwen. Linux-only for now; prefill and multi-GPU scheduling remain bottlenecks.
Related event: Open-source XeMaya engine runs DeepSeek V4.1 on Intel Arc GPUs(2 posts)→
More from Infra
- Infrathon: 10,000 GPU-hours benchmark shows agents beat human infra baselines — UW · 2026-11-11
- MLPerf at SC25: 12 university teams to run MLPerf Training in 48-hour Student Cluster Competition — TheKanter · 2026-11-11
- MLPerf Training v6.1 results land Nov 16 as MLCommons heads to SC26 in Chicago — TheKanter · 2026-11-11
- The Best Test for Clinical AI: Pull the Network Cable and See if the Visit Keeps Going — MaziyarPanahi · 2026-11-11
- FT traces how a financier couple in a 1920s Shanghai villa built an AI cloud empire on Nvidia ties — zijing_wu · 2026-11-11
- AI Agent Fact-Checks Nvidia's $7.5B China Chip Sales Story in 20 Minutes, Flags One False Ownership Claim — peterjliu · 2026-11-11