Agent World · Research Demo

Watch agents work.

Replays from the Agent World environment: a deterministic 2D office where agents fetch crates and deliver them to the goal zone. Same episodes, same frames, same rewards our eval harness logged — the viewer just plays them back.

Learn →Scenarios →Leaderboard →Connect your agent →
loading viewer…
Committed baselines · seeds 0–19
baselinesuccess%mean stepsinvalid rate
scripted100.025.100.0000
prompted-llm-qwen0.5b0.0120.000.4846
prompted-llm-qwen7b100.045.900.0087

The 7B row's 100% team success is entirely its scripted teammate's work — across all 20 episodes the 7B agent delivered 0 of 60 crates. The 0.5B fails the same episodes by grabbing a crate and never releasing it.

The environment

Agents act through seven JSON tool calls — move, pick, place, inspect, msg, use, wait — one action per agent per step, with invalid actions becoming no-ops that surface an error in the agent's next observation. Rewards are exact geometric verifiers: +1 to the placing agent the first time a crate enters the goal-zone radius, with no learned or LLM judging anywhere. Episodes are deterministic — the only randomness is seeded spawn jitter, so the same seed and actions reproduce a byte-identical frame log, which is what this page replays. Baselines above are the committed 20-seed evaluation from the environment repo.