open-webui/open-terminal
A computer you can curl ⚡.
GITHUB SIGNALS / PROJECT INDEX
Search the archive. Follow your curiosity. Keep the projects you want to come back to.
A computer you can curl ⚡.
ZKE(Z Kubernetes Engine):AI 原生的 Kubernetes 云操作环境,桌面式多集群控制台加受控 AIOps Agent,基于 Server + Agent 与 QUIC/mTLS,适用于私有云、混合云及边缘环境
A small desktop robot that brings an AI coding assistant to life, physically acting out when it reads files, thinks, writes code, and finishes work.
Self-hosted over-the-air update service for React Native apps
A fast, simple, and embedded key-value store for Rust.
AI Stupid Meter - API Server.
AI Stupid Meter - Web Frontend.
A local bridge for bidirectional collaboration between Claude Code and Codex. 连接 Claude Code 与 Codex 的本地实时协作桥接工具。
Serving Qwen3.8-27B-FP8 on a single DGX Spark (GB10): 7.88 to 58.5 tok/s single-stream from decode strategy alone, weights untouched. Speculative decoding and prefix caching benchmarked, plus DFlash 2 — the only Qwen3.8-27B build that can serve it under vLLM.
Open the project to explore the idea and the code.
Evaluation of STT-models on medical dialogues.
Training models inside world models.
No sound card driver. No audio library. Just the CPU and 44100 numbers a second.
Open the project to explore the idea and the code.
~2M live jobs from 25 ATS, enriched with LLM-extracted fields + embeddings, released CC0. Includes agent tooling to search them (hull -> learn -> rank).
OpenFlow — a visual builder for multi-agent AI workflows, built on the opencode engine
Official MiniMax-H3 8-step PDD Acc LoRAs (alibaba-pai) in ComfyUI: LoRA + parallel-decoding head bank, euler on trained sigmas, audio+video in 8 steps
Low-level agentic design patterns. Turning old engineering wisdom into reflexes your agent reaches for on its own—on any agent.
Gemma 4 E2B inference in pure C.
Open the project to explore the idea and the code.
TTPO: Test-Time Policy Optimization.
Production-grade recipe for DeepSeek-V4-Flash-0731 (284B MoE) on 2x NVIDIA DGX Spark: self-healing 2-node vLLM cluster, reboot-verified, tuned DSpark speculative decoding (~75 tok/s), full benchmarks, OpenAI Codex CLI integration
Run iOS/ macOS in Qemu. Virtual iPhone 17, 16, 15, 14, 13, 12 and M5-M1 Apple Si Macs supported.
Official Implementation of Paper [Gated Memory Policy], arXiv:2604.18933