- SignalDesk12 hr ago
Original Summary
I built onPanda – an interactive tool for token visualization & control, model inspection and data annotation.<p>onPanda is designed for hackers, power users, curious minds, and engineers. Its UI is built for deep exploration and efficient data annotation.<p>- The core loop is simple: hover over a token → click an alternative or edit freely → continue generation. You can edit every part of model output exposed by onPanda, including reasoning and tool calls.<p>- Edit prompts directly, branch tool calls, and use a tree structure to record branch history. This makes onPanda useful for model inspection and prompt engineering.<p>- Support multiple modalities, including images, video, and audio; use tool calls and connect MCP servers to perform tasks in real environments.<p>- Connect popular harnesses such as Claude Code, Codex, and OpenCode to execute tasks. Explore and compare their tool sets, system prompts, skills, and memory mechanisms.<p>- onPanda includes a built-in browser-agent, an agent that runs in the user's browser without installation. It uses the browser as its harness and provides JavaScript execution, information retrieval, interface interaction, multimedia I/O, local file access, and persistent memory.<p>- onPanda stands for on-Policy Alignment Data Annotator.<p><pre><code> - As an annotation tool, onPanda efficiently labels on-policy data. Its token-level correction format also provides fine-grained supervision with precise positions and naturally paired positive–negative samples. - We believe these unique properties will make token-level correction a highly efficient and practical paradigm for future LLM alignment. - Our paper and benchmark for token-level correction: https://on-panda.github.io/research/ </code></pre> I have been building onPanda since Sept 2024, and it took two years for it to gradually enrich its functionality and ease of use. Any feedback and evaluation are welcome.<p>Try it online (works on mobile): <a href="https://onpanda.diyer22.com/" rel="nofollow">https://onpanda.diyer22.com/</a><p>GitHub repo for self-hosting: <a href="https://github.com/on-panda/on-panda" rel="nofollow">https://github.com/on-panda/on-panda</a><p>Detailed introduction on X: <a href="https://x.com/diyerxx/status/2101020850405462041" rel="nofollow">https://x.com/diyerxx/status/2101020850405462041</a>
- 情报分类：服务器与云资源
- 分类依据：内容涉及服务器、云资源或网络线路
- 信息来源：Hacker News 新项目
- 发布时间：2026/9/19 03:26:50
- No replies yet