radeon-8060s 开源项目推荐（19 个）

hogeheer499-commits /strix-halo-guide

AMD Strix Halo本地LLM指南：为Ryzen AI MAX+ 395 / Radeon 8060S、Ollama、llama.cpp Vulkan/RADV、ROCm、原始证据、直接100 t/s 30B Qwen、140 t/s CHADROCK MTP、120B GGUF。

tsaipifong /whirl-llm

WHIRL (Windows HIP Inference for RDNA LLMs): native Windows C++/HIP LLM inference engine for the AMD Ryzen AI Max+ 395 (Radeon 8060S) and Radeon AI PRO R9700. Unzip and run, only the AMD driver.

jaedync /crucible-strix-halo

Run Qwen3.8-Flash-Next on one AMD Strix Halo (Ryzen AI Max+ 395, gfx1151). EXL3 4.05 and 3.05 bpw packs, own ROCm/HIP kernels, OpenAI-compatible server, 4 sessions of 262K context, setup guide included.

kaka86mm /TitanVault

A turnkey, fully-local AI workstation engineered for the AMD Ryzen AI Max+ 395. LLM inference, voice, document parsing, browser automation, agents — all on-device.

1bit-MONSTER /engine

One binary to rule them all: model-agnostic local LLM inference for AMD Strix Halo (Ryzen AI MAX+ 395) on Linux. Radeon 8060S iGPU (gfx1151) via HRX + our HIP kernels, XDNA 2 NPU via our NPU runtime, inside Lemonade Server with an OpenAI-compatible API. Qwen3.8, ZAYA1, BitNet b1.58, any GGUF. Apache-2.0.

kingjones30 /strix-halo-quant-lab

Run large LLMs locally on AMD Ryzen AI Max+ 395 (Strix Halo, gfx1151) with ROCmFP4 4-bit quantization. Measured benchmarks, build + serving recipes, and 118 ready-to-run GGUF models.

botAGI /strix-halo-llm-benchmarks

Measured LLM benchmarks for AMD Strix Halo / Ryzen AI Max+ 395 (Radeon 8060S, 128 GB unified): llama.cpp Vulkan & ROCm — decode pace, TTFA, prompt cache, quants, sustained load. Every number links to raw runs.

aic0d3r /neon-ladder

playtest-graded benchmark for local LLM stacks: a 10-file game build, 19 static checks, a runtime soak, and a human playing it

Heretek-AI /ROCmFPX-BUILDER

🔥 Automated nightly & release builds of ROCmFPX, Ciru ROCmFPX (DualView Q7/Q8), and q38rocm with AMD ROCm 7 for Windows & Linux. Portable standalone binaries with bundled runtime libraries for Strix Halo (gfx1151), RDNA4, RDNA3, and Homebrew tap.

julianmb /Ornith-1.0-35B-ROCmFPX-StrixHalo

High-performance Vulkan runner & deployment package for Ornith-1.0-35B on AMD Strix Halo (Radeon 8060S). Delivering 115+ t/s aggregate decode, 256k context, and high-accuracy 16-agent burst tool calling.

GEEKOMOfficial /geekom-a9-mega-local-llm-benchmarks

Measured local-LLM benchmark records for the GEEKOM A9 Mega (AMD Ryzen AI Max+ 395 / 388): 72 reproducible llama.cpp test records — single-node, dual-node and triple-node.

MaxusAI /ryzen-ai-max-rocm-ollama-testbench

Docker stack: Ollama v0.21.0 built from source against ROCm 7.2.2 with native gfx1151 (Strix Halo) — serves Gemma 4 up to 256K context on AMD Ryzen AI MAX+ 395 / Radeon 8060S. Includes a 9-layer make validate ladder for the host firmware, ROCm runtime, container, and long-context inference.

YauhenBichel /strix-halo-jax

Run JAX and MuJoCo MJX on AMD Ryzen AI MAX (Strix Halo, gfx1151) from pip wheels - no system ROCm - with a hang-safe check that tells you which wheel set works

xCloudinfo-EnterpriseTeam /ai395-strix-halo-llm

Measured throughput and version traps for running large MoE models (DeepSeek V4 Flash, Qwen3.8 Flash-Next) on an AMD Ryzen AI Max+ 395 / Strix Halo / gfx1151 desktop

jimliddle /qwen38-dflash2-strix-halo-windows

Windows Qwen3.8-27B Q8 benchmarks with DFlash2 on Radeon 8060S (timings, limitations and synthetic replay scripts).

tahircengiz /strix-halo-llm-benchmarks

Measured LLM inference on Ryzen AI MAX+ 395 (Strix Halo): why Vulkan ships instead of ROCm, what the 96 GiB memory split really costs, and why prefill is the real constraint.

Heretek-AI /ember-BUILDER

⚡ Automated nightly builds & portable ROCm 7 releases of Ember for DeepSeek-V4-Flash on AMD Strix Halo (gfx1151 / Radeon 8060S). Standalone C inference server with bundled ROCm runtime, continuous batching, and Homebrew support.

Heretek-AI /CachyLLama-BUILDER

🚀 Automated nightly builds & multi-backend releases (ROCm 7 TheRock, Vulkan RADV, CUDA, Metal) for CachyLLama & llama-ai on AMD Strix Halo (8060S), Strix Point (890M), Phoenix (780M), and Steam Deck with Homebrew support.

1bit-MONSTER /site

Source of https://1bit.gg/, docs and blog of the 1bit engine: model-agnostic local LLM inference on AMD Strix Halo (Ryzen AI MAX+ 395) inside Lemonade Server, on the Radeon 8060S iGPU via HRX and the XDNA 2 NPU. Apache-2.0.