InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery

Autonomous Discovery Across All Sciences

• Papers: InternAgent 1.0 | InternAgent 1.5
• Links: Website | HuggingFace

🔥 News

• 2026.5.07: 🔥🔥 We have open-sourced InternAgent-1.5 with advanced features, including enhanced algorithm discovery tasks, autonomous scientific paper reproduction, memory module, and deep research for tackling complex research challenges.

• 2026.3.17: 🚀🚀 We provide public access to InternAgent's Deep Research capabilities, enabling developers and researchers to seamlessly integrate its advanced deep research functionality into their own workflows.

• 2026.2.14: ❤️‍🔥❤️‍🔥 We open-source MLEvolve, the core implementation of InternAgent's solution optimization subsystem for algorithm design tasks. As the open-source method to achieve #1 on MLEBench, MLEvolve demonstrates powerful capabilities in solution optimization within bounded hypothesis spaces.

• 2026.2.10: 🔥 Official release of the InternAgent 1.5 Technical Report. InternAgent 1.5 achieves leading performance on scientific reasoning benchmarks including GAIA, HLE, GPQA, and FrontierScience, and supports end-to-end autonomous scientific discovery tasks across Physical, Biology, Earth, and Life Science domains, enabling both algorithm discovery and empirical discovery (dry/wet-lab experiments).

• 2025.10.13: InternAgent-1.0 code has been fully open-sourced, supporting end-to-end automation and autonomous evolution across 12 scientific research tasks.

more...

• 2025.07.17: The source code of InternAgent has been partially open-sourced. The complete version of InternAgent (covering 12 types of tasks for autonomous scientific research) will be open-sourced soon. This code repository can be used for full-cycle autonomous scientific research, ranging from hypothesis generation to automated experimental execution.

• 2025.07.10: NovelSeek has been renamed to InternAgent. This change embodies our hopeful vision for autonomous scientific research framework, and we hope it will empower all researchers to achieve great scientific discoveries.

---

🚀 Getting Started

Installation

conda create -n InternAgent python=3.11
conda activate InternAgent
pip install -r requirements.txt

Configure API Keys

Rename .env.example to .env and fill in your API keys:

mv .env.example .env

Key fields in .env:

OPENAI_API_KEY=        # OpenAI or compatible API key (used for embeddings and memory)
OPENAI_API_BASE_URL=   # Base URL for OpenAI-compatible endpoints
ANTHROPIC_API_KEY=     # Anthropic API key (for Claude-based experiment backends)

Run a Discovery Experiment

AutoDebug is a self-contained toy task (no dataset or model downloads required) and is the recommended first run to verify your setup.

python launch_discovery.py \
    --config ./config/default_config.yaml \
    --task AutoDebug \
    --exp_backend claudecode

Run a QA Query

QA mode uses InternAgent's deep research pipeline to answer a research question directly — no experiment loop, just a synthesized answer grounded in literature.

python launch_qa.py --question "What are recent advances in memory-augmented LLMs?"

# Optionally save the answer to a file
python launch_qa.py -q "What are recent advances in memory-augmented LLMs?" -o answer.md

Master Launcher

launch.py is a unified entry point for both modes:

python launch.py --mode discovery --task AutoDebug --exp_backend claudecode
python launch.py --mode qa --question "What are recent advances in memory-augmented LLMs?"

Configuration tips:

• Configuration lives in config/ — default_config.yaml is the main starting point.
• Results are saved under results/, logs under logs/.
• To skip idea generation and run experiments from existing ideas: add --skip_idea_generation --idea_path <path/to/ideas.json>.
• See scripts/ for ready-to-use example scripts.

Research Tasks

Algorithm discovery tasks live under tasks/. Each contains a prompt.json (task description), baseline code/, and a launcher.sh. Datasets and environment-specific setup vary per task — refer to the code in each task folder.

Scientific paper reproduction tasks (sci_tasks) are a distinct mode where InternAgent is given a published paper and its data, and asked to reproduce the key findings autonomously. These tasks live under sci_tasks/tasks/ (from the ResearchClawBench benchmark). See docs/sci_tasks.md for a full guide.

Quick-Start Scripts

Ready-to-use example scripts are provided under scripts/:

Script	Description
run_discovery.sh	Run a full discovery experiment (idea generation → experiments)
run_skip-idea.sh	Run experiments from an existing idea file, skipping generation
run_sci.sh	Run a scientific paper reproduction task (defaults to Astronomy_000)
run_qa.sh	Answer a research question via deep research

---

🔧 Advanced Features

Memory Module

InternAgent 1.5 includes a persistent memory module that records experiment outcomes across sessions, helping the agent avoid previously failed directions and build on successful ones. See docs/memory_module.md for configuration and setup.

Deep Research

The Deep Research (DR) module decomposes a research question into subtasks, gathers information from academic databases and the web in parallel, and synthesizes findings into a direct answer or structured report. See docs/deep_research.md for configuration and setup.

---

📝 Citation

@article{feng2026internagent,
  title={InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery},
  author={Shiyang Feng and Runmin Ma and Xiangchao Yan and Yue Fan and Yusong Hu and Songtao Huang and Shuaiyu Zhang and Zongsheng Cao and Tianshuo Peng and Jiakang Yuan and Zijie Guo and Zhijie Zhong and Shangheng Du and Weida Wang and Jinxin Shi and Yuhao Zhou and Xiaohan He and Zhiyin Yu and Fangchen Yu and Bihao Zhan and Qihao Zheng and Jiamin Wu and Mianxin Liu and Chi Zhang and Shaowei Hou and Shuya Li and Yankai Jiang and Wenjie Lou and Lilong Wang and Zifu Wang and Jiong Wang and Wanghan Xu and Yue Deng and Dongrui Liu and Yiheng Wang and Wenlong Zhang and Fenghua Ling and Shufei Zhang and Xiaosong Wang and Shuangjia Zheng and Xun Huang and Siqi Sun and Shuyue Hu and Peng Ye and Chunfeng Song and Bin Wang and Conghui He and Yihao Liu and Xin Li and Qibin Hou and Tao Chen and Xiangyu Yue and Bin Wang and Liang He and Dahua Lin and Bowen Zhou and Bo Zhang and Lei Bai},
  journal={arXiv preprint arXiv:2602.08990},
  year={2026}
}

@article{team2025internagent,
  title={InternAgent: When Agent Becomes the Scientist--Building Closed-Loop System from Hypothesis to Verification},
  author={Team, InternAgent and Zhang, Bo and Feng, Shiyang and Yan, Xiangchao and Yuan, Jiakang and Ma, Runmin and Hu, Yusong and Yu, Zhiyin and He, Xiaohan and Huang, Songtao and others},
  journal={arXiv e-prints},
  pages={arXiv--2505},
  year={2025}
}

@article{hu2025flowsearch,
  title={FlowSearch: Advancing deep research with dynamic structured knowledge flow},
  author={Yusong Hu and Runmin Ma and Yue Fan and Jinxin Shi and Zongsheng Cao and Yuhao Zhou and Jiakang Yuan and Xiangchao Yan and Wenlong Zhang and Lei Bai and Bo Zhang},
  journal={arXiv preprint arXiv:2510.08521},
  year={2025}
}

@article{du2025automlgen,
  title={AutoMLGen: Navigating Fine-Grained Optimization for Coding Agents},
  author={Shangheng Du and Xiangchao Yan and Dengyang Jiang and Jiakang Yuan and Yusong Hu and Xin Li and Liang He and Bo Zhang and Lei Bai},
  journal={arXiv preprint arXiv:2510.08521},
  year={2025}
}