feat: hy3-research — 深度研究助手，由 Hy3 驱动#143
Open
ilovecplusplus230 wants to merge 5 commits into
Open
Conversation
Update README files to include dataset preparation instructions and clarify key parameters for Hy3 reinforcement learning training. Enhance README files with acknowledgements for support from Tencent Hunyuan team and relevant communities. Fix typo in README by changing "Posttraining" to "Post-training" for consistency and clarity. Fix RL Post-training TOC anchor in README
docs: add RL post-training guide
- quickstart.md: API basics, curl/Python examples, parameter guide, troubleshooting - 6 runnable examples under issue1/examples/: basic_chat, streaming, latency_compare, tool_calling, reasoning_mode, retry - each example includes .py (runnable) and .md (documentation) - uses TokenHub OpenAI-compatible API (hy3-preview) - .gitignore updated to exclude .env files - api_demo.py: standalone API demo under project root
- Plan → Search → Fetch → Synthesize → Report full pipeline - Dual search engine (Tavily + DuckDuckGo) - Interactive CLI + auto mode + Web report viewer - Mock offline mode for demo without API key - Hy3-driven: planning, synthesis, report generation via chat/completions API
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
hy3-research
零外部依赖（仅 Python 标准库 + 少量 pip 包），支持 OpenAI 兼容的 Hy3 API，内置 mock 离线模式，开箱即可演示。
✨ 它能做什么
--auto一键到底。🧩 Hy3 在系统中承担的角色
即使在无 key 的 mock 模式下，整体交互链路也完全可跑通。
🎬 两个端到端 Demo
Demo 1 — 浅度研究
python -m hy3research --auto "Kubernetes vs Serverless 2026年企业选型对比"
→ Plan: 3 subtopics → Search: ~20 sources → Synthesize: 3 concurrent Hy3 calls → Report
Demo 2 — 深度研究
python -m hy3research --auto "Transformer架构的替代方案最新进展"
→ Plan: 5 subtopics → Search: ~30 sources → Synthesize: 5 concurrent Hy3 calls → 深度综述
🤝 与 CodeBuddy / WorkBuddy 的协作记录
hy3research/client.py：Hy3 OpenAI 兼容客户端、重试逻辑与 mock 模式（CodeBuddy 骨架 + WorkBuddy 推理模型兼容）hy3research/planner.py：研究计划生成、JSON 解析与容错（WorkBuddy 设计 prompt 结构）hy3research/searcher.py：Tavily + DDG 双引擎搜索（复用 issue3 逻辑，WorkBuddy 适配）hy3research/reporter.py：长文报告生成 prompt 设计（WorkBuddy 撰写）hy3research/__main__.py：CLI 编排逻辑（WorkBuddy 搭建骨架）templates/report.html：Web 报告模板（WorkBuddy 设计暗色/亮色主题）README.md与demo/目录（WorkBuddy 撰写）其余配置、测试与调试由作者完成。
📦 项目结构
issue4/
├── hy3research/
│ ├── init.py # 版本信息
│ ├── main.py # CLI 入口 (argparse)
│ ├── config.py # 配置 / .env 加载
│ ├── client.py # Hy3 客户端 (OpenAI 兼容 + mock)
│ ├── planner.py # 研究计划生成 (Hy3)
│ ├── searcher.py # Tavily + DDG 后备搜索
│ ├── fetcher.py # URL 内容抓取
│ ├── synthesizer.py # 源材料 → Hy3 综合
│ ├── reporter.py # 长文报告 + 引用生成
│ ├── server.py # HTTP 静态文件服务
│ └── ui.py # 终端渲染
├── templates/report.html # 报告 Web 模板
├── tests/ # 单元测试
├── demo/ # 演示脚本 + 视频
└── README.md
🔒 安全说明
.env提供，.env已被 git 忽略。