Benchmarking dots.tts on Strix Halo
RedNote's 2B continuous autoregressive TTS hits RTF 0.35 on the NVIDIA 4070 Super with the MeanFlow-distilled checkpoint, putting it among the fastest voice-cloning-capable models I've tested locally.
read more →BeeLlama.cpp DFlash on Strix Halo: 2.7x Gemma 31B, But MTP Is Still Faster
BeeLlama.cpp's DFlash speculative decoding nearly triples dense model throughput on AMD Strix Halo, but in a strict head-to-head against my existing MTP setups, MTP still wins by 23-67% depending on the configuration.
read more →Running the MOSS-TTS Family on Strix Halo
Benchmarking OpenMOSS's 8B llama.cpp GGUF backend and 100M ONNX Nano model on AMD's Ryzen AI MAX+ 395, with thread scaling analysis and a surprising Nano result that beats everything else I've tested.
read more →Meet People Where They Are
The AI discourse is stuck between two poles that refuse to talk to each other. If we actually care about this technology being used well, we need to stop evangelizing and start listening.
read more →Testing llama.cpp PR #21344: Faster MoE Prefill, but MTP Fights Back
A community PR optimizing CUDA kernels for GFX1151 delivers +24% prefill throughput on MoE models, but combining those same kernel changes with MTP speculative decoding makes inference slower. Not every optimization stacks.
read more →ROCm 7 on Strix Halo: Benchmarking the New Toolbox Images
AMD released ROCm 7.13 with Strix Halo optimizations. I benchmarked kyuz0's latest toolbox images against my current ROCm 6.4.4 production baseline to see if upgrading my llama-swap stack is worth it. The answer is complicated.
read more →Running DramaBox on Strix Halo
Getting Resemble AI's expressive TTS model running on AMD Strix Halo with no NVIDIA hardware. TheRock gfx1151 nightlies, bitsandbytes preview for ROCm, reduced step counts, and torch.compile bringing the 3.3B DiT from RTF 4.0 down to 1.75.
read more →Unsloth Studio on Strix Halo: Full GPU Training Without System ROCm
Getting Unsloth Studio's full training pipeline running on AMD Strix Halo (gfx1151) using pip-packaged ROCm nightlies, no /opt/rocm required. Chat, training, data recipes, and model export all working on Fedora 43.
read more →Gemma 4 MTP Assistant: 3.7x Faster 31B and +45% Faster 26B-A4B on Strix Halo
Google's official Gemma 4 MTP assistant heads bring speculative decoding to MoE models that couldn't benefit before, and nearly quadruple dense model throughput on AMD Strix Halo's bandwidth-limited unified memory.
read more →Optimizing Echo-TTS: CPU Beats GPU
Eight optimization attempts on Echo-TTS CPU inference, the five that worked, quality evaluation with voice cloning, and how the optimized CPU path ended up faster than the GPU hybrid.
read more →MTP Speculative Decoding: 4.8x Faster Qwen 3.6 27B on Strix Halo
Multi-Token Prediction turns Qwen 3.6 27B from 6 t/s to 30 t/s on AMD Strix Halo, succeeding where draft models and ngram decoding failed, by using prediction heads baked into the model itself.
read more →Benchmarking Echo-TTS on Strix Halo
Running a diffusion-based TTS model on AMD's Strix Halo, patching CUDA-only code for CPU, discovering a bf16 GPU hang on gfx1151, and a hybrid GPU/CPU trick that beats every other TTS model I've tested.
read more →Running HY-World 2.0 on Strix Halo: 3D World Reconstruction on an AMD iGPU
Porting Tencent's CUDA-only 3D world model to AMD's Radeon 8060S via ROCm Docker, flash-attention CK kernels, a fully compiled gsplat with wave32 patches, and complete 3D reconstruction output including Gaussian splats.
read more →Friends Don't Let Friends Use Ollama
Ollama gained traction by being the first easy llama.cpp wrapper, then spent years dodging attribution, misleading users, and pivoting to cloud, all while riding VC money earned on someone else's engine. Here's the full history, and why the alternatives are better.
read more →Speculative Decoding on Strix Halo: 2x Faster Gemma 4 31B Token Generation
Benchmarking speculative decoding with Gemma 4 E2B as a draft model for Gemma 4 31B on AMD Strix Halo, a bandwidth-bound setup where the optimal draft-max differs from discrete GPUs.
read more →Pi Web UI: A Browser Interface for the Pi Coding Agent
A full-stack web interface that puts the Pi coding agent in the browser, with system-level access, session history, and model switching through a local LiteLLM proxy.
read more →Local LLM Infrastructure on Strix Halo
How LiteLLM, llama-swap, and Lemonade Server compose into a unified local inference platform, routing dozens of models across GPU and NPU through a single API endpoint, accessible anywhere via Tailscale and a local reverse proxy.
read more →Running LLMs on the AMD NPU with Lemonade Server
Setting up AMD's Lemonade Server on Strix Halo to run LLM and Whisper inference on the XDNA 2 NPU, driver builds, architecture decisions, and benchmarks against the integrated GPU.
read more →Benchmarking OmniVoice on Strix Halo
Running a 600+ language zero-shot TTS model on an AMD integrated GPU, voice cloning benchmarks, ROCm compatibility adventures, and the container workaround that actually worked.
read more →Benchmarking VoxCPM2 on Strix Halo
Running a 2B parameter tokenizer-free TTS model in both Python and C++ on AMD's integrated GPU, near-real-time speech synthesis on CPU, and the Vulkan crash that stopped GPU acceleration in its tracks.
read more →Self-Hosting Fish Audio on Strix Halo
Running Fish Audio's 4B parameter S2-Pro text-to-speech model locally on an AMD Strix Halo integrated GPU via ROCm and Podman.
read more →Medium-Claw: A Persistent AI Companion on Telegram
A Telegram bot backed by the Pi coding agent with autonomous scheduling, persistent memory, cross-session search, and a web dashboard.
read more →LoopMaker Web
A browser-based AI music generation tool powered by ACE-Step, ported to Linux for local generation on AMD Strix Halo hardware.
read more →QuizForge: Self-Learning Quiz Maker
A full-stack quiz platform that turns markdown files and YouTube transcripts into mixed-format quizzes with AI grading, contextual chat, and performance analytics.
read more →Oneiros: A Personal AI Agent Platform
A modular collection of services for building a personal AI agent, tool use, memory, browser automation, TTS, and multi-platform chat interfaces.
read more →OCR List Maker
Snap a photo of a handwritten list, OCR it with a local vision model, and print a formatted checklist on a thermal receipt printer.
read more →llama-cpp-python in Docker
A Dockerfile and docker-compose setup for running llama.cpp with its Python bindings in a container, because finding a working one shouldn't be this hard.
read more →Lavabo: The Kitchen Sink of Local AI
An all-in-one Docker container that bundles LLMs, embeddings, vision, and TTS into a single unified inference server.
read more →Web Browser Wrapped
Generating weekly Spotify Wrapped-style reports from browser history using local models and Browser Recall data.
read more →Speech To Text Typing for Wayland Users
Building a custom speech-to-text solution for Linux Wayland users using NVIDIA's Canary model, Silero VAD, and ydotool.
read more →Total (Browser) Recall
Building a personal browser history search engine with full-text recall, inspired by Microsoft's Recall and rewind.ai.
read more →A Practical, Fully Local Desktop Voice Agent
Building a natural language voice controller for the Linux desktop using Qt6, a tiny 1.7B LLM, and a clever vector embedding trick for tool calling.
read more →A Fully Local, In-Browser Voice Assistant
Building a private, browser-based voice assistant using WebAssembly, Moonshine STT, Piper TTS, and local LLMs.
read more →You Are John
A text-driven simulation where you interact with a guy named John through natural language, and an LLM determines how his world responds.
read more →