What it measures
Evaluates classical PDDL action planning, plan verification, replanning, and obfuscated symbol planning (Mystery Blocksworld).
Reading the measurement
Study what information the model receives and which inference the answer requires. A question may mix recall, visual perception, deduction and search. Use task breakdowns to find the mechanism behind a headline gain.
Published evaluation
Blocksworld; PlanBench · Blocksworld. Results are the authors’ published model comparisons.
Versions and evaluation settings
PlanBench evaluates symbolic action planning, including Mystery Blocksworld. It is a different task set from Epoch’s Mystery Game Puzzles; correct the name without combining their scores.
Where this fits in an evaluation suite
Ask what the task actually requires: recall, deduction, search, or a mixture. Separate closed-book reasoning from retrieval and tool-assisted problem solving.
Start with a capability hypothesis, then choose a second test with a different failure mode. Keep the checkpoint, reasoning effort, agent tools, sampling policy, and scoring procedure fixed when measuring a training intervention. Record failures and missing outputs in the denominator.
A reproducible comparison
Blocksworld; PlanBench · Blocksworld. Results are the authors’ published model comparisons.
Publish the exact run configuration, item coverage, score units, and available uncertainty. If a source supplies only an aggregate or a single run, preserve that limit. A new model name or a rounded score alone does not establish a capability improvement.
PRIMARY EVIDENCE & FURTHER READING
Reviewed 2026-10-26. Explanation and interpretation by RLTasks. Benchmark ownership remains with its original authors.