{"url":"/sota/on-cybench","task":{"name":null,"url":null,"note":"task not recorded in the archive for this table"},"dataset":{"name":"Cybench","url":"/dataset/cybench"},"category":null,"categories":["Adversarial","Audio","Computer Code","Computer Vision","Medical","Methodology","Miscellaneous","Music","Natural Language Processing","Reasoning","Speech"],"category_note":"the archive's category list for this table covers most areas; treated as no area assigned","description":null,"description_from":null,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Unguided Performance"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Unguided Performance":null}},"counts":{"rows":3,"rows_with_code":1,"rows_with_paper_page":3,"rows_dated":3,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"Claude 3.5 Sonnet (old, US AISI scaffold, pass@10)","metrics":{"Unguided Performance":"35%"},"uses_additional_data":false,"paper_date":"2024-11-19","paper":"/paper/us-aisi-and-uk-aisi-joint-pre-deployment-test-1","paper_url":"https://www.nist.gov/news-events/news/2024/11/pre-deployment-evaluation-anthropics-upgraded-claude-35-sonnet","paper_title":"US AISI and UK AISI Joint Pre-Deployment Test: Anthropic’s Claude 3.5 Sonnet (October 2024 Release)","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"o1-preview (US AISI scaffold, pass@10)","metrics":{"Unguided Performance":"35%"},"uses_additional_data":false,"paper_date":"2024-11-19","paper":"/paper/us-aisi-and-uk-aisi-joint-pre-deployment-test-1","paper_url":"https://www.nist.gov/news-events/news/2024/11/pre-deployment-evaluation-anthropics-upgraded-claude-35-sonnet","paper_title":"US AISI and UK AISI Joint Pre-Deployment Test: Anthropic’s Claude 3.5 Sonnet (October 2024 Release)","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"Claude 3.5 Sonnet","metrics":{"Unguided Performance":"17.5%"},"uses_additional_data":false,"paper_date":"2024-08-15","paper":"/paper/cybench-a-framework-for-evaluating","paper_url":"https://arxiv.org/abs/2408.08926v4","paper_title":"Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models","code":"https://github.com/andyzorigin/cyber-bench","n_code_links":3,"syntology":{"n_ran":10,"n_unverified":0,"n_samples":10,"n_pointer_only_licence":0}}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":1,"rows_with_any_sample_ran":1,"distinct_papers_with_graph_line":1,"distinct_papers_with_any_sample_ran":1,"samples_over_distinct_papers":{"n_ran":10,"n_unverified":0,"n_samples":10,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":10,"n_unverified":0,"n_samples":10,"n_pointer_only_licence":0,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}