facebookresearch/llm-speedrunner
Benchmark for agents to reproduce scientific findings
View on index · View in 3D Map
Related repos
- Benchmark for LLMs on software engineering tasks
- Schedules speed tests and tracks internet performance over time
- Generates research reports using LLMs and web search
- Library for benchmarking multi-agent reinforcement learning algorithms
- Video models like SlowFast help computers spot actions and understand what’s happening in videos by combining slow and fast motion cues
- Benchmark for evaluating LLMs on real-world GitHub issues
- Agent for research tasks across LLMs
- Benchmark evaluates LLMs as agents
- Pipeline turns social posts into LLM training datasets
