MiroThinker is a deep research agent optimized for complex research and prediction tasks. Our latest models, MiroThinker-1.7, achieves 74.0 and 75.3 on the BrowseComp and BrowseComp Zh, respectively.
Democratizing Reinforcement Learning for LLMs
[EMNLP'25] s3 - ⚡ Efficient & Effective Search Agent Training via RL for RAG (RLVR for Search with Minimal Data)
[ACL 2026 KnowFM] Awesome Agentic Deep Research Resources
[COLM’25] DeepRetrieval — 🔥 Training Search Agent by RLVR with Retrieval Outcome
An open-source LLM based automatically daily news collecting workflow showcase powered by Agently AI application development framework.
🔍 The hardest search benchmark in the wild — vague, multi-turn, proactive. 200 long-horizon tasks with persona-driven progressive disclosure, scored by verifiable schema-free knowledge-graph evaluation. No vibes, just triplet F1.
🔍 OpenSearch-VL provides a fully open recipe for training strong multimodal deep search agents through high-quality data curation, diverse visual/search tools, and fatal-aware agentic reinforcement learning.
GISA: A Benchmark for General Information-Seeking Assistant
🎯Awesome-AgenticRAG_DeepResearch: A curated list of resources on Agentic RAG & DeepResearch. 学习参考关于AgenticRAG、DeepResearch的发展相关论文
a lightning-fast, crash-proof search engine aggregator built in Rust. Focused on memory efficiency, safe character slicing, and high-performance result fetching.
An MCP server providing real-time web search capabilities to any AI model.
8 Puzzle solver using uninformed and informed search algorithms as DFS, BFS and A*.
An enterprise-ready Google Search Agent demonstrating secure Agent-to-Agent (A2A) communication using Google ADK on GCP.
A RAG-based AI agent for breast cancer that uses Qdrant to retrieve answers, filter relevance, and grow its knowledge base.
Official code implementation for my ready tensor publication, a religiously aligned agent
8 Puzzle Game Solver using A*, BFS and DFS
🧠 Transform ideas into structured insights with MiroThinker, a powerful tool for enhancing creativity and collaboration in your projects.
A self-contained AI project that runs a quantized Large Language Model (Qwen2.5-0.5B) entirely on your local machine. Built with FastAPI and llama-cpp-python, this agent intelligently switches between standard chat and "Search Mode" to fetch real-time data from the internet. The project features a responsive HTML/CSS/JS frontend and is fully Docker
Airweave lets agents search any app
Add a description, image, and links to the search-agent topic page so that developers can more easily learn about it.
To associate your repository with the search-agent topic, visit your repo's landing page and select "manage topics."