ROCm Beats Vulkan on Strix Halo: Up to 47% Faster Decoding at 41.9 tok/s
QuixiAI · x · 2026-09-17
The Lucebox team reports pushing ROCm past Vulkan on AMD Strix Halo. Benchmarked same-day on the same machine with identical prompts: with the DSpark drafter, Lucebox (ROCm) decodes at 41.9 tok/s at 8K and 38.0 tok/s at 123K prompt tokens — 28-47% faster than llama.cpp/Vulkan v0.7.5, with 17-47% faster prefill. A notable win for local inference on AMD's AI Max hardware.
More from Infra
- Palmer Luckey asks: can you use datacenter RAM for things prohibited by the operator's terms of service? — avaitopiper · 2026-11-02
- Disable MTP for 20-30% faster local LLM inference on limited RAM — InternalMode8159 · 2026-11-02
- Kioxia up 216% in a year but down 68% from its June 2026 peak — zephyr_z9 · 2026-11-02
- Intel plunges 29%, worst day since the 1970s, wiping $12B off US government stake — emmanuelvivier · 2026-11-02
- Researcher hoards nearly 1TB of DRAM, refuses to sell a stick to data centers — kevinnbass · 2026-11-02
- Self-Hosting LLMs at a Law Firm: Infra and Data Governance Harder Than the Model — EducationalGroup7245 · 2026-11-02