AI Progress Tracker

    Claude Opus 4.6 · Claude Opus 4.6

    Anthropic · 5 Feb 2026

    Ask the Advisor about this model
    74from 8
    benchmarks

    Claude Opus 4.6 is identified as a model in Anthropic's Claude family. Not yet confirmed by public reporting.

    context: 1Mtextimagefile

    Benchmark results

    Every figure links to where it was published. Compare like with like — harnesses and prompt setups differ between vendors.

    BenchmarkScoreSource
    GPQA DiamondPhD-level science questions90.5%6 Feb 2026View
    SWE-bench VerifiedResolving real GitHub issues78.7%18 Feb 2026View
    OTIS Mock AIME 2024-25Competition mathematics94.4%6 Feb 2026View
    Terminal-BenchLong-horizon command-line agent tasks79.8%12 Mar 2026View
    FrontierMath40.7%12 Feb 2026View
    SimpleQA VerifiedShort factual questions (hallucination resistance)47%27 Aug 2026View
    LMArena TextHuman preference Elo from blind chat votes (not a percentage)1501 Elo22 Oct 2026View
    LMArena WebDevHuman preference Elo for building web apps (not a percentage)1546 Elo23 Oct 2026View

    Weighted mean of 6 verified benchmark results: GPQA Diamond 90.5%, SWE-bench Verified 78.7%, OTIS Mock AIME 2024-25 94.4%, Terminal-Bench 79.8%, FrontierMath 40.7%, SimpleQA Verified 47%.

    Capabilities

    • Not yet confirmed by public reporting.

    What improved

    • Not yet confirmed by public reporting.

    Known flaws & limitations

    • Not yet confirmed by public reporting.

    Risks, safety & security issues

    • Not yet confirmed by public reporting.

    Good fit for

    • Not yet confirmed by public reporting.

    Poor fit for

    • Not yet confirmed by public reporting.
    Security & risk dashboardLatest vulnerabilities, advisories and real-world incidents for Claude Opus 4.6

    References

    Profile last reviewed 10/27/2026 and re-checked automatically every few hours against the vendor's published documentation, papers and public leaderboards. Scores come from the cited benchmark results above — confirm limits and pricing against the provider's own documentation before building on them.