AMD Users: Have you tried the llamma.cpp AMD-Ecosystem branch? Up to 2x PP Speed
Summary
The article discusses an AMD-specific branch of llama.cpp that significantly boosts prompt processing speed for AMD users, with up to 2x faster performance on dense models using ROCm/Hip, though with some trade-offs in other metrics.
Similar Articles
GLM and I created a llama.cpp fork optimized for AMD GFX906 (Mi50, Mi60, Radeon VII, GCN HIP) - Machine Learning, LLMs, & AI
A fork of llama.cpp has been created and optimized for AMD GFX906 GPUs, improving performance on Mi50, Mi60, Radeon VII, and GCN HIP for machine learning and LLM applications.
llama.cpp B9387 Significant AMD/ROCm PP Update
llama.cpp version b9387 introduces MFMA support for AMD CDNA architecture (MI100, MI200, MI300 series), improving processing pipeline performance on datacenter AMD GPUs.
Tip: use this llama.cpp PR to improve PP on Intel ARC
A llama.cpp PR significantly improves prompt processing speed on Intel ARC GPUs, with benchmark showing speed increase from 245t/s to 462t/s on a B580. The improvement currently works for F16 KV quantization, with plans to support other quants.
@pupposandro: 2.5x faster than llama.cpp on Strix Halo. We just shipped DFlash + PFlash for the AMD Ryzen AI MAX+ 395 iGPU (gfx1151, …
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.
@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.