Browse › Playing Games › Atari Games › Atari 2600 Breakout

Atari Games archive 2025-07-28

Atari 2600 Breakout Benchmark (Atari Games)

58 rows 50 with code listed 1 metric Dataset page

The Atari 2600 Games task (and dataset) involves training an agent to achieve high game scores.

The archive carries no text for this table; the description above is the archive's text for the task Atari Games. archive 2025-07-28

Over time archive 2025-07-28

The chart needs JavaScript; the table below carries every value.

Direction inferred from the metric name, not from the archive: Score (higher is better). Points are placed at the row's paper date; 57 of 58 rows carry one.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 GDI-H3(200M frames) 864.00 – Paper – 2022 no code linked report
2 GDI-I3(200M frames) 864.00 – Paper – 2022 no code linked report
3 GDI-I3 864.00 – Paper – 2022 no code linked report
4 GDI-H3 864 – Paper – 2022 no code linked report
5 Bootstrapped DQN 855 – Paper Code 2016 1 of 2 ran · 1 unverified report
6 FQF 854.2 – Paper Code 2019 1 of 10 ran · 9 unverified report
7 R2D2 837.7 – Paper Code 2019 linked, not harvested report
8 Ape-X 800.9 – Paper Code 2018 0 of 15 ran · 15 unverified report
9 Agent57 790.4 – Paper Code 2020 0 of 9 ran · 9 unverified report
10 IMPALA (deep) 787.34 – Paper Code 2018 16 of 34 ran · 18 unverified report
11 A3C LSTM hs 766.8 – Paper Code 2016 38 of 95 ran · 57 unverified report
12 MuZero (Res2 Adam) 758.04 – Paper Code 2021 linked, not harvested report
13 C51 noop 748.0 – Paper Code 2017 0 of 1 ran · 1 unverified report
14 QR-DQN-1 742 – Paper Code 2017 2 of 2 ran · 0 unverified report
15 IQN 734 – Paper Code 2018 0 of 2 ran · 2 unverified report
16 A3C FF hs 681.9 – Paper Code 2016 38 of 95 ran · 57 unverified report
17 DNA 626 – Paper Code 2022 linked, not harvested report
18 ASL DDQN 621.7 – Paper Code 2023 linked, not harvested report
19 A3C FF (1 day) hs 551.6 – Paper Code 2016 38 of 95 ran · 57 unverified report
20 Reactor 500M 514.8 – Paper – 2017 no code linked report
21 POP3D 458.41 – Paper Code 2018 linked, not harvested report
22 A2C + SIL 452 – Paper Code 2018 1 of 6 ran · 5 unverified report
23 Persistent AL 431.89 – Paper Code 2015 linked, not harvested report
24 Advantage Learning 425.32 – Paper Code 2015 linked, not harvested report
25 DDQN (tuned) noop 418.5 – Paper Code 2015 5 of 11 ran · 6 unverified report
26 Duel hs 411.6 – Paper Code 2015 5 of 11 ran · 6 unverified report
27 Nature DQN 401.2 – Paper Code 2015 linked, not harvested report
28 DQN noop 385.5 – Paper Code 2015 55 of 106 ran · 51 unverified report
29 Prior noop 373.9 – Paper Code 2015 78 of 111 ran · 33 unverified report
30 MAC 372.7 – Paper Code 2017 linked, not harvested report
31 DDQN (tuned) hs 368.9 – Paper Code 2015 55 of 106 ran · 51 unverified report
32 Prior+Duel noop 366.0 – Paper Code 2015 5 of 11 ran · 6 unverified report
33 UCT 364.4 – Paper Code 2012 2 of 2 ran · 0 unverified report
34 Prior+Duel hs 354.6 – Paper Code 2015 55 of 106 ran · 51 unverified report
35 DQN hs 354.5 – Paper Code 2015 55 of 106 ran · 51 unverified report
36 DDRL A3C 350 – Paper Code 2018 linked, not harvested report
37 Duel noop 345.3 – Paper Code 2015 5 of 11 ran · 6 unverified report
38 DDQN+Pop-Art noop 344.1 – Paper – 2016 no code linked report
39 Prior hs 343.0 – Paper Code 2015 78 of 111 ran · 33 unverified report
40 Recurrent Rational DQN Average 336 – Paper Code 2021 6 of 6 ran · 0 unverified report
41 Rational DQN Average 316 – Paper Code 2021 6 of 6 ran · 0 unverified report
42 Gorila 313.0 – Paper Code 2015 linked, not harvested report
43 DreamerV2 312 – Paper Code 2020 3 of 15 ran · 12 unverified report
44 DT 267.5 – Paper Code 2021 17 of 26 ran · 9 unverified report
45 NoisyNet-Dueling 263 – Paper Code 2017 1 of 3 ran · 2 unverified report
46 DQN Best 225 – Paper Code 2013 56 of 117 ran · 61 unverified report
47 SPOS 180.6 – Paper Code 2020 linked, not harvested report
48 ENAS Search space 1 161.1 – Paper Code 2020 linked, not harvested report
49 SPOS Search space 1 144.4 – Paper Code 2020 linked, not harvested report
50 ENAS 91.4 – Paper Code 2020 linked, not harvested report
51 DARQN hard 20 – Paper Code 2015 linked, not harvested report
52 CURL 18.2 – Paper Code 2020 6 of 8 ran · 2 unverified report
53 CGP 13.2 – Paper Code 2018 0 of 9 ran · 9 unverified report
54 Discrete Latent Space World Model (VQ-VAE) 11.6 – Paper – 2020 no code linked report
55 ES FF (1 hour) noop 9.5 – Paper Code 2017 7 of 29 ran · 22 unverified report
56 SARSA 6.1 – – – not matched report
57 Best Learner 5.2 – Paper Code 2012 2 of 2 ran · 0 unverified report
58 SAC 0.7 – Paper Code 2019 4 of 18 ran · 14 unverified report

All 58 rows shown. 57 link to a paper page on this site; 0 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 34 rows have a graph line, from 23 distinct papers; 29 rows (18 papers) have at least one sample that ran. Counting each paper once: Syntology ran 299 of 637 samples; 338 unverified. Separately, 202 of those 637 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.

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