gfx1151 repositories
Trending repositories tagged gfx1151, ranked by trending score. Updated daily.
-
1
peonist-ai/halogen-flash-server
Local inference that behaves like a frontier API: Qwen3.8-Flash-Next on AMD Strix Halo plus small NPU models, over the OpenAI and Anthropic APIs on one port.
GitHub · Shell · 1,407 stars · 102 forks · +17 stars in 24 hours · Trending score 2.66
-
2
gufo-org/gufo
The open-source AI engine for AMD Strix Halo. Text, embeddings, speech, image and video models in one binary, with OpenAI-compatible APIs, caching for coding agents and model extensions. Linux and Windows.
GitHub · C++ · 1,233 stars · 211 forks · +17 stars in 24 hours · Trending score 2.65
-
3
hec-ovi/vllm-awq4-qwen
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.
GitHub · Python · 53 stars · 5 forks · +8 stars in 24 hours · Trending score 1.60
-
4
hyukudan/strix-halo-engine-lab
Exact LLM inference engine written from scratch for AMD Strix Halo (gfx1151, ROCm/HIP). Qwen3.8-Flash-Next on one node or two replicas; DeepSeek V4.1 Flash across two nodes over USB4 with an optional RTX PRO. Speculative decoding bit-identical to greedy, up to 255 tok/s on coding chat. MiniMax and GLM next.
GitHub · C++ · 38 stars · 6 forks · +4 stars in 24 hours · Trending score 1.48
-
5
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.
GitHub · Python · 35 stars · 7 forks · +2 stars in 24 hours · Trending score 1.45
-
6
IIIIIllllIIIIIlllll/Qwenox
An inference framework customized for AI MAX 395, compatible with both Linux and Windows, and supports deploying only qwen3.8-flash-next. This project is for reference and learning purposes only. 为AI MAX 395定制的推理框架,Linux和Windows通用,只能部署qwen3.8-flash-next。本项目仅供参考学习使用。
GitHub · C++ · 150 stars · 31 forks · +5 stars in 24 hours · Trending score 1.45
-
7
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.
GitHub · C++ · 10 stars · 2 forks · +1 stars in 24 hours · Trending score 0.69
-
8
pepuscz/strix-halo-llm
DeepSeek-V4.1-Flash (1M context) and GLM-5.3-Flash on a BOSGAME M5 (Ryzen AI Max+ 395, Radeon 8060S, 128 GiB) with two RX 7900 XTX over OCuLink: pinned Ansible recipes and measured results. Also the M5-only Qwen and DeepSeek setups.
GitHub · Python · 31 stars · 2 forks · 0 stars in 24 hours · Trending score 0.43
-
9
franzmoca/ds41-gufo-extension
DeepSeek V4.1 Flash for Gufo on AMD Strix Halo: DSpark drafting, vision, tool calls and SSD expert streaming
GitHub · C++ · 7 stars · 2 forks · +1 stars in 24 hours · Trending score 0.42
-
10
hogeheer499-commits/strix-halo-guide
Evidence-backed AMD Strix Halo local-AI setup and benchmarks: Qwen3.8, Ollama, llama.cpp, Vulkan/ROCm, large GGUFs, and cross-OEM results.
GitHub · Python · 371 stars · 27 forks · 0 stars in 24 hours · Trending score 0.03
-
11
outerreaches/moonshine
Run the official Kimi K3 MoE checkpoint on one 128 GB AMD Strix Halo box. ROCm-resident static weights, MXFP4 experts streamed from NVMe via io_uring. C engine, chat client, OpenAI-compatible server. Very experimental.
GitHub · C · 7 stars · 0 forks · 0 stars in 24 hours · Trending score 0.01
-
12
julianmb/ds4fa
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
GitHub · C · 21 stars · 2 forks · 0 stars in 24 hours · Trending score 0.00
-
13
Nathanw1014/strix-halo-llamacpp
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.
GitHub · Python · 227 stars · 21 forks · 0 stars in 24 hours · Trending score 0.00
-
14
peonist-ai/halogen-server
The fastest way to run Qwen3.8 27B on Strix Halo (gfx1151)
GitHub · Python · 137 stars · 7 forks · 0 stars in 24 hours · Trending score 0.00
-
15
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.
GitHub · 6 stars · 0 forks · 0 stars in 24 hours · Trending score 0.00
-
16
rulith-dev/rulith-inference
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.
GitHub · Python · 34 stars · 3 forks · 0 stars in 24 hours · Trending score 0.00
-
17
Aristo94/GSQHalo.cpp
llama.cpp fork for GSQ-quantized Qwen3.8-Flash-Next on AMD Strix Halo (gfx1151): faster ROCm prefill, MTP and 4×256K chats with Q8_0 KV and one KV stream per slot, persistent KV cache on SSD. Fork of halo-box/strix-llama.cpp.
GitHub · C++ · 11 stars · 0 forks · 0 stars in 24 hours · Trending score 0.00
-
18
asf0/libstrix
HIP kernels for AMD Strix Halo (gfx1151) and a vLLM plugin that runs them on stock upstream vLLM: up to 31x faster prefill and ~2x decode at long context.
GitHub · Python · 19 stars · 1 forks · 0 stars in 24 hours · Trending score 0.00