Lists (14)
Starred repositories
The headless browser for AI agents and web scraping
Arbitrary verified fast precision floating point arithmetic in Lean, with proofs, optimized backends, and support for IEEE binary and decimal, posits, and custom formats.
A training-free multi-agent framework for recursive self-improvement in new environments through broad-then-deep autonomous exploration and reusable memory.
SoL-Pi: Scaling Auto-Research Loops for Efficient Agent Harnesses
A lightweight library for xPU kernel JIT compilation
DeepSelect: TopK kernels for DeepSeek Sparse Attention (DSA) and Samplers
DeepSeek Harness: Everything is a Plugin.
Production-ready MoE load balancing via real-time expert replication
MoonEP: A Perfectly Balanced Expert Parallelism Library via Dynamic Redundant Experts
Programmatic memory for long-horizon LLM agents: the harness appends everything to one log, and the agent searches it with code. 97.4% on ARC-AGI-3 (arXiv:2607.20064)
Typst Open-Source Server for collaborative editing
Atmospheric Transport Model in pure Julia - GPU-CPU capable, ERA or GEOS met sources, LatLon, Reduced Gaussian and CubeSphere Topologies
AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI
Engine-agnostic LLM gateway in Rust. Full OpenAI & Anthropic API compatibility across vLLM, TRT-LLM, TokenSpeed, SGLang, OpenAI, Gemini & more. Industry-first gRPC pipeline, KV cache-aware routing,…
A tutorial on modern GPU programming for machine learning systems
An open source design system that's fully customizable and agent ready
The open-source AI workbench for scientific research
利用 Zotero 中文 CSL 开发组测试文献，测试十种支持 GB/T 7714 的参考文献引擎。
DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms
An open-source, PyTorch-like runtime for dynamic multi-agent and multi-session workflows.
Humming is a high-performance, lightweight, and highly flexible JIT (Just-In-Time) compiled GEMM kernel library specifically designed for quantized inference.
RTX 6000 Pro Wiki — Running Large LLMs (Qwen3.5-397B, Kimi-K2.5, GLM-5) on PCIe GPUs without NVLink
AIPerf is a comprehensive benchmarking tool that measures the performance of generative AI models served by your preferred inference solution.
ISEEKYAN / Megatron-LM
Forked from NVIDIA/Megatron-LMOngoing research training transformer models at scale
Design the LSP(language server) but for Agent. native parser and ast understanding first.