Vibeleaderboard
← Benchmark guide

Model results

SciCode

Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.

Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.

Official source
← All evaluation groups

SciCode

Source
SciCode via Epoch AI
Version
Not reported
Measurement date
Not reported
Metric
% · Higher is better
Evaluation scope
effort max · result date not supplied

Positions are withheld because at least one result lacks explicit independent provenance or a documented evaluation scope. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

  1. Claude Opus 4.7

    anthropic/claude-opus-4-7

    54.5%

    View modelEvidence aggregated

  2. Claude Sonnet 5

    anthropic/claude-sonnet-5

    53.6%

    View modelEvidence aggregated

  3. Claude Opus 4.8

    anthropic/claude-opus-4-8

    53.5%

    View modelEvidence aggregated

  4. GLM 5.2

    z-ai/glm-5.2-20260616

    51.2%

    View modelEvidence aggregated

  5. DeepSeek V4 Pro 0423

    deepseek/deepseek-v4-pro-20260423

    50%

    View modelEvidence aggregated

  6. DeepSeek V4 Flash 0731

    deepseek/deepseek-v4-flash-20260731

    49.9%

    View modelEvidence aggregated

  7. Claude Sonnet 4.6

    anthropic/claude-sonnet-4-6

    46.8%

    View modelEvidence aggregated

  8. DeepSeek V4 Flash 0423

    deepseek/deepseek-v4-flash-20260423

    44.9%

    View modelEvidence aggregated
SciCode model results | VibeLeaderboard