Vibeleaderboard
← Benchmark guide

Model results

Aider Polyglot Benchmark

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

Aider Polyglot Benchmark

Source
Aider via Epoch AI
Version
polyglot-225
Measurement date
Not reported
Metric
% · Higher is better
Evaluation scope
Edit format diff · 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. R1 0528

    deepseek/deepseek-r1-0528

    71.4%

    View modelEvidence aggregated

  2. DeepSeek V3.2 Exp

    deepseek/deepseek-v3.2-exp

    70.2%

    View modelEvidence aggregated

  3. Qwen3 235B A22B

    qwen/qwen3-235b-a22b-04-28

    59.6%

    View modelEvidence aggregated

  4. Kimi K2 0711

    moonshotai/kimi-k2

    59.1%

    View modelEvidence aggregated

  5. R1

    deepseek/deepseek-r1

    56.9%

    View modelEvidence aggregated

  6. DeepSeek V3 0324

    deepseek/deepseek-chat-v3-0324

    55.1%

    View modelEvidence aggregated

  7. DeepSeek V3

    deepseek/deepseek-chat-v3

    48.4%

    View modelEvidence aggregated

  8. Qwen3 32B

    qwen/qwen3-32b-04-28

    40%

    View modelEvidence aggregated

  9. Qwen2.5 Coder 32B Instruct

    qwen/qwen-2.5-coder-32b-instruct

    8%

    View modelEvidence aggregated
Aider Polyglot Benchmark model results | VibeLeaderboard