pinned
Running
Agents
10
SGI-Bench Leaderboard
🥇
Scientific General Intelligence of LLMs/vLLMs
None defined yet.
Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition
MLExpert: Towards a Generalist Agentic Model for Autonomous Machine Learning Engineering
Scientific General Intelligence of LLMs/vLLMs
Open, science-focus leaderboards benchmarking LLMs and VLMs
Lightweight harness for tool-using LLM agents.
Validate and submit a ResearchClawBench task ZIP to Hugging Face