RLTasksby ulam.ai
Models / Meta

Meta / MODEL EVIDENCE

Llama 3.1-Instruct-405B

1 benchmarks · 6 observations · 1 sources

Scores at a glance

1 benchmarks · large dot: Llama 3.1-Instruct-405B; small dots: closest models where reported

ScienceAgentBench ↗3.6%
3.623.9

Llama-3.1-Instruct-405B · ScienceAgentBench · Task success · Without expert knowledge ↗

5 more reported results

2.6% · Llama-3.1-Instruct-405B · ScienceAgentBench · Task success · Without expert knowledge · ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery ↗

8.8% · Llama-3.1-Instruct-405B · ScienceAgentBench · Task success · Without expert knowledge · ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery ↗

2.9% · Llama-3.1-Instruct-405B · ScienceAgentBench · Task success · With expert knowledge · ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery ↗

4.3% · Llama-3.1-Instruct-405B · ScienceAgentBench · Task success · With expert knowledge · ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery ↗

8.2% · Llama-3.1-Instruct-405B · ScienceAgentBench · Task success · With expert knowledge · ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery ↗

Closest models by score

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

30.0 percentile-point gap

1 of 1 shared benchmark comparisons

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. ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery — published result table ↗

    Ziru Chen, Shijie Chen, Yuting Ning, Qianheng Zhang, Boshi Wang, Botao Yu et al. · 2024

    All 20 authors

    Ziru Chen, Shijie Chen, Yuting Ning, Qianheng Zhang, Boshi Wang, Botao Yu, Yifei Li, Zeyi Liao, Chen Wei, Zitong Lu, Vishal Dey, Mingyi Xue, Frazier N. Baker, Benjamin Burns, Daniel Adu-Ampratwum, Xuhui Huang, Xia Ning, Song Gao, Yu Su, Huan Sun

Names used in source tables

Llama-3.1-Instruct-405B