Agentic Evaluation in 2D Mazes

What is Agentic Evaluation in 2D Mazes?

Evaluates interactive multimodal agents in structured 2D maze environments requiring long-horizon planning, mechanism interaction, and error recovery, scored…

Released

2026-08-22

Evaluates

Agents, Agents & Tool Use, Multimodal, Software & AI Compute

Openness

unknown

Importer

Claire Radar

Review status

ai-name-audit-deferred

Reported scores

0

Source provenance

• Original evidence https://github.com/bluebunnyanon/agent-benchmark-2d-maze

Agentic Evaluation in 2D Mazes code

• Code https://github.com/bluebunnyanon/agent-benchmark-2d-maze

Which sources cite Agentic Evaluation in 2D Mazes?

• Agentic Evaluation in 2D Mazes

Related Agents, Agents & Tool Use, Multimodal, Software & AI Compute benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• A-Bench
• A-OKVQA
• A3S-Bench
• ActionAtlas
• Active-SWE
• ActiveFly-Bench