Stars
Lightweight SQL-based stream processing engine for IoT edge.
A high-performance, Go-based streaming media server that supports JT/T 1078 and GB28181 protocols.
lalserver with srt/webrtc/whip/whep/http-fmp4/hls(fmp4)/low latency hls(llhls)/gb28181. 卍解
jt808服务端 单机[2核4G]并发10w+ 100%测试覆盖率 例子丰富(每日保存亿+经纬度[2核4G] 分布式集群 主动安全协议 JT1078流媒体播放 兼容任意808服务 808模拟客户端 真实设备对接案例等)
Dolphin is a multilingual, multitask ASR model jointly trained by DataoceanAI and Tsinghua University.
The easiest tool for fine-tuning LLM models, synthetic data generation, and collaborating on datasets.
A lightweight dependency-free workflow automation platform. Supports iPaaS, stream computing, MCP, and AI capabilities.
A powerful tool for creating fine-tuning datasets for LLM
A MULTI-GENERATOR ENSEMBLE FRAMEWORK FOR NATURAL LANGUAGE TO SQL
vproxy-tools / ktransformers
Forked from kvcache-ai/ktransformersA Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations
fastllm是后端无依赖的高性能大模型推理库。同时支持张量并行推理稠密模型和混合模式推理MOE模型，任意10G以上显卡即可推理满血DeepSeek。双路9004/9005服务器+单显卡部署DeepSeek满血满精度原版模型，单并发20tps；INT4量化模型单并发30tps，多并发可达60+。
一个基于✨HOOK机制的微信机器人，支持🌱安全新闻定时推送【FreeBuf，先知，安全客，奇安信攻防社区】，👯Kfc文案，⚡漏洞查询，⚡手机号归属地查询，⚡知识库查询，🎉星座查询，⚡天气查询，🌱摸鱼日历，⚡微步威胁情报查询， 🐛视频，⚡图片，👯帮助菜单。📫 支持积分功能，⚡支持自动拉人，，🌱自动群发，👯Ai回复（国内主流AI模型，扣子，FastGpt，Dify全面支持！），⚡视频号解析，😄自…
D^2-MoE: Delta Decompression for MoE-based LLMs Compression
微信、支付宝、通联支付、拉卡拉、PayPal、Apple 的Go版本SDK。【极简、易用的聚合支付SDK】
Official repository of "Investigating Tradeoffs in Real-World Video Super-Resolution"
Simplifying SSL/TLS traffic analysis for researchers by making SSL decryption effortless.
VPTQ, A Flexible and Extreme low-bit quantization algorithm
Official code implementation of General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model
Accelerate local LLM inference and finetuning (LLaMA, Mistral, ChatGLM, Qwen, DeepSeek, Mixtral, Gemma, Phi, MiniCPM, Qwen-VL, MiniCPM-V, etc.) on Intel XPU (e.g., local PC with iGPU and NPU, discr…
A prize winning solution for Multimedia Deepfake Detection competition.
This is a repository used by individuals to experiment and reproduce the pre-training process of LLM.
🚀 「大模型」1小时从0训练26M参数的视觉多模态VLM！🌏 Train a 26M-parameter VLM from scratch in just 1 hours!
🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!
Provide the same packet processing capabilities as wireshark for Go.[wireshark4.4.5 supported] Due to delays in github LFS upgrade, the latest dll is pushed to google；
Anti Virtulization, Anti Debugging, AntiVM, Anti Virtual Machine, Anti Debug, Anti Sandboxie, Anti Sandbox, VM Detect package. Windows ONLY.
A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations
AI Native Data App Development framework with AWEL(Agentic Workflow Expression Language) and Agents
real time face swap and one-click video deepfake with only a single image
UDE (USB Device Emulation) Hardware-less sample, with matching Host-Side drivers. Used as a USB study test bed.