@pupposandro: 2.5x faster than llama.cpp on Strix Halo. We just shipped DFlash + PFlash for the AMD Ryzen AI MAX+ 395 iGPU (gfx1151, …
Summary
A new toolset (DFlash + PFlash) achieves 2.5x faster inference than llama.cpp on AMD Ryzen AI MAX+ 395 iGPU, demonstrating significant speedups for Qwen3.6-27B with 128 GiB unified memory.
Similar Articles
@pupposandro: https://x.com/pupposandro/status/2054241934164492328
The article announces support for DFlash and PFlash speculative decoding in llama.cpp for AMD Strix Halo iGPUs, demonstrating significant speedups in inference performance using ROCm.
Luce DFlash + PFlash on AMD Strix Halo: Qwen3.6-27B at 2.23x decode and 3.05x prefill vs llama.cpp HIP
Luce releases DFlash and PFlash support for AMD Strix Halo APUs, achieving 2.23x decode and 3.05x prefill speedups over llama.cpp HIP on Qwen3.6-27B.
If you are running Qwen 3.8 Flash Next on Strix Halo, use this software for inference. It's so much faster than llama.cpp especially at high context.
A user recommends a GitHub project for faster inference of Qwen 3.8 Flash Next on Strix Halo, showing benchmark speeds that significantly outperform llama.cpp.
I tested freshly merged DFlash in llama.cpp on Qwen 3.6 27B Local AI win. 4.44x faster at 36K context. Here are my findings RTX 6000 PRO.
A user benchmarks the newly merged DFlash speculative decoding method in llama.cpp on Qwen 3.6 27B, achieving up to 4.44x speedup at 36K context compared to baseline, with detailed leaderboard and quality tests.
@pupposandro: Excited to announce that Lucebox engine now serves Qwen3.8-27B on a single AMD Radeon AI PRO R9700 (32 GB, RDNA4) with …
The Lucebox engine now supports running the Qwen3.8-27B model on a single AMD Radeon AI PRO R9700 GPU, achieving up to 227 tok/s on code tasks using the DFlash2 drafter with lossless verification.