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.
☆52Nov 9, 2026Updated this week
Alternatives and similar repositories for rulith-inference
Users that are interested in rulith-inference are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- 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, 7…☆38Nov 2, 2026Updated last week
- Halowin Bubo: gufo for native Windows with AMD HIP, tuned for Strix Halo laptops☆23Updated this week
- LLM inference in C/C++☆43Updated this week
- 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 a…☆11Oct 11, 2026Updated 3 weeks ago
- A fast diffusion model engine for image generation and editing.☆39Updated this week
- Virtual machines for every use case on DigitalOcean • AdGet dependable uptime with 99.99% SLA, simple security tools, and predictable monthly pricing with DigitalOcean's virtual machines, called Droplets.
- Portable optimized ROCm and Qwen3.8-27B setup for AMD Strix Halo☆85Sep 16, 2026Updated last month
- The Open Source Strix Halo AI engine. Qwen 3.8 Flash Next Q4_K_XL: 1,801.01 pp, 73.17 tg single user, 194.37 tok/s 8 users; Qwen 3.8 27B…☆1,123Updated this week
- ☆18Updated this week
- LLM inference engine written from scratch in HIP for AMD Strix Halo (gfx1151) and Qwen3.8-Flash-Next - ~50 tokens/s out to 492k context☆100Updated this week
- GPU + XDNA2 NPU inference for AMD Strix Halo (gfx1151): Qwen3.6-35B-A3B, Qwen3.8-27B, Qwen3.8 Flash-Next, Flint and GLM-5.3 Flash (321B M…☆93Updated this week
- ☆12Updated this week
- Gufo OCI Images for AMD Strix Halo☆24Updated this week
- Local inference that behaves like a frontier API: Qwen3.8-Flash-Next on AMD Strix Halo plus small NPU models, over the OpenAI and Anthrop…☆1,261Updated this week
- The fastest way to run Qwen3.8 27B on Strix Halo (gfx1151)☆157Sep 14, 2026Updated last month
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- llama.cpp fork tuned for Qwen 3.8 Flash-Next on AMD Strix Halo — RDNA3.5 kernel patches, true QSA sparse gather, working MTP, 27 GB engra…☆50Oct 14, 2026Updated 3 weeks ago
- High-performance Lemonade alternative for AMD Strix Halo & Radeon — latest MoE models (Ling-3.0-Flash, Qwen 3.8 Flash Next, Ornith 1.5), …☆95Sep 24, 2026Updated last month
- Performance-tuned llama.cpp for AMD Strix Halo (gfx1151): FA + MoE-prefill fixes with a bundled current Mesa driver. Vulkan and HIP; port…☆224Oct 25, 2026Updated 2 weeks ago
- An integrated toolkit for viral phylogeographic analysis☆13Jun 22, 2026Updated 4 months ago
- DeepSeek-V4-Flash inference for AMD Strix Halo☆28Sep 11, 2026Updated last month
- Kyojin: open-source engine that runs 100 GB-class open models (GLM-5.3, Qwen3.8, DeepSeek V4.1) on one desktop machine. AMD Strix Halo, N…☆234Updated this week
- LLM inference in C/C++☆260Updated this week
- Docker Compose for llama.cpp GGUF servers on AMD Strix Halo: Qwen, Gemma, and Laguna packages (abliterated and quantized), stock Vulkan p…☆28Aug 17, 2026Updated 2 months ago
- Tensor-parallel DeepSeek V4 Flash inference on dual AMD Strix Halo over OdinLink USB4/TB5 RDMA or Mellanox RoCE v2.☆23Sep 20, 2026Updated last month
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- A wide range of importing tools for converting third party reports into our internal format.☆14Updated this week
- RDNA-native LLM inference engine in Rust.☆806Updated this week
- Lightweight Python tool using Optuna for tuning llama.cpp flags: towards optimal tok/s for your machine☆61Aug 17, 2026Updated 2 months ago
- qwen3.8-flash-next on strix halo — 91g quant, 56 tok/s, 262k context, receipts included☆28Sep 24, 2026Updated last month
- A beautifully crafted desktop client for running and managing local LLMs via llama.cpp.☆24May 25, 2026Updated 5 months ago
- Pinned one-DGX-Spark Docker recipe for DeepSeek V4 Flash with EXL3, SparkInfer, and 262K NVFP4 MLA KV cache☆31Sep 6, 2026Updated 2 months ago
- Enhances Kokoro TTS output by merging segments with dynamic, programmable pauses for meditative or narrative flow.☆17Oct 16, 2025Updated last year
- Simple Tool Caller for llama.cpp☆11Aug 12, 2024Updated 2 years ago
- unofficial version of eratoho TWKR☆25Aug 19, 2025Updated last year
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Open-source EFI bootloader that unlocks compute + render of NVIDIA CMP 90HX (GA102) before Windows boots. Compute/AI focused: this mining…☆40Oct 14, 2026Updated 3 weeks ago
- Evidence-backed AMD Strix Halo local-AI setup and benchmarks: Qwen3.8, Ollama, llama.cpp, Vulkan/ROCm, large GGUFs, and cross-OEM results…☆371Updated this week
- Open-source platform for creating, distributing and running sovereign EU-compliant LLMs. Verticalize any model for your domain, language …☆61Updated this week
- Technical docs to help you make you Halo Strix WORK!☆70Sep 3, 2026Updated 2 months ago
- 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-com…☆19Updated this week
- This will be the official ROCmFPX stack and studio☆50Oct 17, 2026Updated 3 weeks ago
- Kotai is a fully local, zero-cost voice assistant that combines the power of Kyutai TTS/STT, LiveKit, and local LLMs to create natural co…☆10Jul 14, 2025Updated last year
Are these results useful?