Lists (12)
Stars
Unity 商用级别开发框架，原生内置 AI 工作流支持，集成 HybridCLR 高性能热更、Obfuz 代码混淆加固、YooAssets 企业级资源管理方案，构建高效、安全、可扩展的工业化开发底座。
Hindsight: Agent Memory That Learns
Multi-instance split-pane manager for Claude Code — a cross-platform desktop app built with Tauri 2
Universal Rust multiplexer with a typed SDK — drive any CLI or TUI app from code. Native on Linux, macOS, and Windows.
为纯文本 DeepSeek 模型提供按需识图的 Codex Skill（VEP/1 视觉证据包 + 多 Provider 降级）
A native CLIProxyAPI vision bridge that keeps long-context text models as the final answerer.
Rust-first financial market data core and TDX provider
Geometry-aware LLM quantization, per-direction bit allocation priced by measured KL divergence given injected perturbation.
tdxrs 是通达信 (TDX) 行情数据解析库的 Rust 高性能实现，通过 PyO3/maturin 提供原生 Python 接口。它无缝兼容 [tdxpy] 的 API，并将核心解析引擎以 Rust 重构，从而实现数量级的本地解析性能提升，尤其在海量历史数据处理场景下优势显著。
tdx通达信实时数据查询及api接口，docker本地部署，支持32个接口- ✅ **实时行情** - 五档买卖盘口、最新价、涨跌幅 - ✅ **K线数据** - 支持10种周期（1分钟~年K） - ✅ **分时数据** - 当日分时走势、逐笔成交 - ✅ **股票搜索** - 代码/名称模糊搜索 - ✅ **批量查询** - 同时获取多只股票数据
a community oriented 1:1, vLLM-alike (Continuous batching, paged KV) engine in C++ with additional features
LvLLM is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, su…
📦 🚀 Blazing fast, simple and complete solution for micro frontends.
A framework-agnostic embedded file viewer for vanilla JavaScript, React, Vue and Svelte. Put PDF, Office, images, media, archives, email, drawings, 3D, GIS and engineering files inside one stable c…
Native HIP/C++ Z-Image-Turbo inference optimized for AMD Radeon RX 7900 XTX gfx1100 WMMA.
General plug-and-play inference library for Recursive Language Models (RLMs), supporting various sandboxes.
Patching for XiaoAi Speakers (小爱音箱), add custom binaries and open source software. Tested on LX06, LX01, LX05, L09A
Deep Code 是专为 deepseek-v4 模型优化的终端 AI 编码助手，支持深度思考、推理强度控制以及 Agent Skills。
面向长篇小说创作的 AI Native 开源系统，用 Agent、世界观、写法引擎、RAG 和整本生产工作流，帮助新手从一句灵感走到完整小说。AI-native engine for end-to-end novel creation — from idea to full chapters, with structured planning, worldbuilding, and agen…
AirLLM 70B inference with single 4GB GPU
A lightweight, lightning-fast, in-process vector database