Popular repositories Loading

1. q38rocm q38rocm Public

   Qwen 3.8 27B ROCmFP4 on AMD Strix Halo (Ryzen AI Max+ 395). Up to 36 tok/s via MTP Speculation, TurboQuant & Mesa RADV Wave64.

   Python 63 4

2. ds4fa ds4fa Public

   DeepSeek V4 Flash 284B on AMD Strix Halo (gfx1151) — up to 32 tok/s decode & ~250 tok/s prefill via ROCmFPX, DSpark & ROCm 7.2

   C 10 1

3. nemotron-3.5-30b-a3b-rocmfp4 nemotron-3.5-30b-a3b-rocmfp4 Public

   Converter fixes, NVFP4 findings & Strix Halo ROCmFP4 benchmarks for NVIDIA Nemotron 3.5 Lightning 30B-A3B

   Shell 1

4. rocmfpx-server rocmfpx-server Public

   Unified Model Server & Model Zoo for ROCmFPX / ROCmFP4 on AMD Strix Halo

   Python 1

5. BlockFlash BlockFlash Public

   C#

6. HFTMEX HFTMEX Public

   C# 1