gfx1151

The fastest way to run Qwen3.8-Flash-Next on Strix Halo (gfx1151)

Strix Halo inference engine. Qwen Flash Next Q4_K_XL: 1,628.52pp, 59.41tg single user, 157.22 tok/s 8 users; Qwen27B Q4_K_XL: 656.33pp, 70.56tg tok/s single user with DFlash2

Evidence-backed AMD Strix Halo local-AI setup and benchmarks: Qwen3.8, Ollama, llama.cpp, Vulkan/ROCm, large GGUFs, and cross-OEM results.

Performance-tuned llama.cpp for AMD Strix Halo (gfx1151): FA + MoE-prefill fixes with a bundled current Mesa driver. Vulkan and HIP; portable dir, Docker, and distrobox.

The fastest way to run Qwen3.8 27B on Strix Halo (gfx1151)

vLLM Qwen 3.6-27B (AWQ-INT4) + DFlash speculative decoding on AMD Strix Halo (gfx1151 iGPU, 128 GB UMA, ROCm 7.13). 24.8 t/s single-stream, vision, tool calling, 256K context, OpenAI-compatible, Docker. Matches DGX Spark FP8+DFlash+MTP at a third of the cost. No CUDA.

Local text to textured GLB on an AMD Strix Halo iGPU (gfx1151): FLUX.2 klein, a Vulkan-only TRELLIS.2 engine, and humanoid auto-rigging with SkinTokens on ROCm. No Blender, no CUDA.

SGLang Docker image for AMD Strix Halo (gfx1151 / Radeon 8060S)

Docker Compose for llama.cpp GGUF servers on AMD Strix Halo: Qwen, Gemma, and Laguna packages (abliterated and quantized), stock Vulkan plus ROCmFP4/MTP and ROCmFPX, parallel slots, with prefill/decode and quality metrics measured on this rig.

Rulith Inference, formerly Strix Llama. Serving a 125B MoE model fast on one AMD Strix Halo machine, on Windows: a patched llama.cpp, a manager, and a desktop app.

Tensor-parallel DeepSeek V4 Flash inference on dual AMD Strix Halo over OdinLink USB4/TB5 RDMA or Mellanox RoCE v2.

ComfyUI with Flash Attention for AI+ MAX 395 (gfx1151)

vLLM + Qwen3.6-27B (BF16) OpenAI-compatible inference server on AMD Strix Halo (Ryzen AI Max+ 395, gfx1151). Vision input, 256K context, /v1/responses with separated reasoning, via TheRock ROCm.

Clean-room AGPL-3.0 reimplementation of the halogen 0.1.3 inference engine for Qwen3.8-27B on AMD Strix Halo (gfx1151) — same wire protocol and OpenAI-compatible API. Not affiliated with Peonist.

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

ROCmFPX llama.cpp fork for Windows 🏆 — native build, headless OpenAI-compatible server & benchmarks. Tested on AMD Strix Halo (gfx1151), runs on other GPUs too.

llama.cpp fork for GSQ-quantized Qwen3.8-Flash-Next on AMD Strix Halo (gfx1151): faster ROCm prefill, MTP that fits 2×256K, persistent KV cache on SSD. Fork of halo-box/strix-llama.cpp.

llama.cpp + Qwen3.6-27B (Q8_0 GGUF) OpenAI-compatible inference server on AMD Strix Halo (Ryzen AI Max+ 395, gfx1151). 256K context, ~7.5 t/s decode via TheRock ROCm Docker.

Claude Code skill for AMD Strix Halo (Ryzen AI MAX+ 395) ML setup. Handles PyTorch installation (official wheels don't work with gfx1151), GTT memory config, and environment setup. Enables 30B parameter models.

Reproducible local-LLM benchmark harness: llama.cpp on AMD Strix Halo (gfx1151, Ryzen AI Max+ 395) and NVIDIA DGX Spark — frozen corpora, quality gates with unit tests, sealed run bundles. Apache-2.0