RLTasksby ulam.ai
Models / OpenAI

OpenAI / MODEL EVIDENCE

GPT-4o / Plan+Execute

1 benchmarks · 9 observations · 1 sources

Scores at a glance

1 benchmarks · large dot: GPT-4o / Plan+Execute; small dots: closest models where reported

DiscoveryWorld ↗56%
5659

GPT-4o / Plan+Execute · DiscoveryWorld · Average (Easy) · Procedure · GPT-4o agent scaffolds ↗

8 more reported results

18% · GPT-4o / Plan+Execute · DiscoveryWorld · Average (Easy) · Task completion · GPT-4o agent scaffolds · DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents ↗

11% · GPT-4o / Plan+Execute · DiscoveryWorld · Average (Easy) · Explanatory knowledge · GPT-4o agent scaffolds · DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents ↗

64% · GPT-4o / Plan+Execute · DiscoveryWorld · Average (Normal) · Procedure · GPT-4o agent scaffolds · DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents ↗

18% · GPT-4o / Plan+Execute · DiscoveryWorld · Average (Normal) · Task completion · GPT-4o agent scaffolds · DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents ↗

2% · GPT-4o / Plan+Execute · DiscoveryWorld · Average (Normal) · Explanatory knowledge · GPT-4o agent scaffolds · DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents ↗

50% · GPT-4o / Plan+Execute · DiscoveryWorld · Average (Challenge) · Procedure · GPT-4o agent scaffolds · DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents ↗

15% · GPT-4o / Plan+Execute · DiscoveryWorld · Average (Challenge) · Task completion · GPT-4o agent scaffolds · DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents ↗

1% · GPT-4o / Plan+Execute · DiscoveryWorld · Average (Challenge) · Explanatory knowledge · GPT-4o agent scaffolds · DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents — published result table ↗

    Jansen, Peter, Côté, Marc-Alexandre, Khot, Tushar, Bransom, Erin, Mishra, Bhavana Dalvi, Majumder, Bodhisattwa Prasad et al. · 2024

    All 8 authors

    Jansen, Peter, Côté, Marc-Alexandre, Khot, Tushar, Bransom, Erin, Mishra, Bhavana Dalvi, Majumder, Bodhisattwa Prasad, Tafjord, Oyvind, Clark, Peter

Names used in source tables

GPT-4o / Plan+Execute