Datasets › SMAC

SMAC (The StarCraft Multi-Agent Challenge)

Introduced by Mikayel Samvelyan et al. in The StarCraft Multi-Agent Challenge11 Feb 2019 archive 2025-07-28

The StarCraft Multi-Agent Challenge (SMAC) is a benchmark that provides elements of partial observability, challenging dynamics, and high-dimensional observation spaces. SMAC is built using the StarCraft II game engine, creating a testbed for research in cooperative MARL where each game unit is an independent RL agent.

Benchmarks archive 2025-07-28

All 5 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
SMAC SMAC 6h_vs_8z ACE Median Win Rate 93.75 ACE: Cooperative Multi-agent Q-learning with... opendilab/ace 14 Compare
SMAC SMAC MMM2 ACE Median Win Rate 100 ACE: Cooperative Multi-agent Q-learning with... opendilab/ace 14 Compare
SMAC SMAC 3s5z_vs_3s6z ACE Median Win Rate 100 ACE: Cooperative Multi-agent Q-learning with... opendilab/ace 13 Compare
SMAC SMAC corridor ACE Median Win Rate 100 ACE: Cooperative Multi-agent Q-learning with... opendilab/ace 13 Compare
SMAC SMAC 27m_vs_30m DDN Median Win Rate 91.48 DFAC Framework: Factorizing the Value Function via... j3soon/dfac 11 Compare

Papers archive 2025-07-28

5 shown of 5 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 324. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
A Unified Framework for Factorizing Distributional Value Functions for Multi-Agent Reinforcement Learning 1 10 4 Jun 2023 not harvested
ACE: Cooperative Multi-agent Q-learning with Bidirectional Action-Dependency 1 4 29 Nov 2022 ran 0 of 4 samples (4 unverified)
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning 1 30 16 Feb 2021 ran 1 of 1 samples (0 unverified)
Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning 1 5 19 Mar 2020 ran 1 of 1 samples (0 unverified)
The StarCraft Multi-Agent Challenge 23 16 11 Feb 2019 ran 6 of 15 samples (9 unverified; 13 pointer-only for licence)

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

License archive 2025-07-28

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

Modalities archive 2025-07-28

No modality tagged.

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • SMAC MMM2
  • SMAC 6h_vs_8z
  • SMAC 3s5z_vs_3s6z
  • SMAC corridor
  • SMAC 27m_vs_30m
  • SMAC

6 variant names, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections