Tom Turney

turboquant_plus. Python

llama-cpp-turboquant. LLM inference in C/C++

vllm-swift. vLLM Metal plugin powered by mlx-swift — high-performance LLM inference on Apple Silicon

elm327_obd_for_mac. I'm crazy and trying to make a ForScan OBD reader work on my mac.

longctx. Open long-context inference stack: retrieval + open weights, no closed parts. pip install longctx.

momento. Deterministic State Recovery for AI Coding Agents

mlx. MLX: An array framework for Apple silicon

BookTies. The new Book Ties!!

pascal-egpu. Driverless NVIDIA Pascal (GTX 1060) compute from macOS Apple Silicon over Thunderbolt eGPU

BookTiesOld. Book Ties old source code.

pastors-pocket-spurgeon. Offline Spurgeon study companion — pastoral counsel, sermon prep, and Sword & Trowel sermon grading. Gemma-4-12B fine-tune served on TurboQuant llama.cpp.

cs161. StemVille real-time data plotting experiment

LearnPerl. just perl code

main_web.

turboquant-tinygrad-bridge. Compressed KV cache as cross-backend wire format for Metal + CUDA split inference over Thunderbolt 5

tqkit. Unified toolkit for benchmarking and integrating TurboQuant+ KV-cache compression across llama.cpp, vLLM, MLX, and vllm-swift.

ds4-turboquant_plus. Private backup of TQ+ Metal port work on antirez/ds4. Branch tom/turbo3-kv-cuda (turbo2/3/4 + Wave M3 inline-dequant + half-tile + CUDA stubs + h8 Flash WIP).

ffai. FFAI — F*cking Fast AI. Dual-engine inference: Swift (Apple/iPhone, native Metal) + Rust (cross-platform: CUDA/Vulkan/ROCm). Shared kernels via metaltile.