Lists (3)
Stars
Experimental MLIR-to-systolic-array compiler/runtime flow: lowers `linalg.matmul` to a custom accelerator call and validates execution against cocotb/Verilator RTL simulation, with plans to evolve …
Speech synthesis, speech recognition and voice activity detection for voice conversation, in C++ on ggml: Qwen3-TTS, Irodori-TTS, Qwen3-ASR, ReazonSpeech, parakeet and Silero VAD on Metal, Vulkan a…
LLM-authored task-specific Jev harnesses with optional full-trajectory reward reflection and GEPA evolution.
A nano replica of Jev: parallel decisions, dynamic candidates, and an end-to-end training pipeline.
A Difficulty-Calibrated Benchmark for Building Terminal Agents
Conformance Tests for MCP
DeepSeek Harness: Everything is a Plugin.
早耳 - Real-time multilingual speech-to-text on CPU only. Live subtitles, browser dashboard, speaker labels, translation. No GPU, no cloud.
[EMNLP Main '25] LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation
A high-performance, universal serving framework for any-to-any models.
FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.
Open source inference engine that tunes its kernels on your hardware to find the fastest way to run each model. 2x faster than Ollama. Works on Apple Silicon, NVIDIA, AMD, or just a CPU.
A specification that python filesystems should adhere to.
FineEnvs — RL Environments 101: building and scaling RL environments in the age of LLMs
A slide framework built for agents.
A fancy NVIDIA GPU monitor for the terminal. One Python file, no dependencies.
Lightweight Benchmark Server for Agentic GPU Programming
An Open Compiler Harness for Agentic GPU Programming
Everything you need to know about LLM inference
Elasticsearch gui - desktop app, browser extension, docker, self hosted
LLM inference server with continuous batching & SSD caching for Apple Silicon — managed from the macOS menu bar