+1 (415) 347-6981Get started
Fixed prices · You own the code · Proposal in 24h hello@startrise.io →
Engineering discipline — maintenance

Brownfield Change Request

30 models built this brief single-shot. Claude Opus 5 leads at 81.1. No blind human ratings yet on this brief; scores are gates and panel only. 6 split the judge panel.

GitHub repo ↗
↗
  1. Claude Opus 5 Anthropic · Jul 2026
    81.1
    79 gates83 panel- human
    Open live ↗
  2. GPT-6 Astra OpenAI · Oct 2026
    79.7
    79 gates80 panel- human
    Open live ↗
  3. Kimi K3 Moonshot · Jul 2026 · split panel
    79.6
    79 gates80 panel- human
    Open live ↗
  4. Grok 4.6 xAI · Aug 2026
    79.6
    79 gates80 panel- human
    Open live ↗
  5. Claude Sonnet 5.5 (high) Anthropic · Oct 2026
    79.6
    79 gates80 panel- human
    Open live ↗
  6. GPT-5.6 Sol OpenAI · Jul 2026
    78.1
    79 gates78 panel- human
    Open live ↗
  7. Muse Spark 1.3 Meta · Oct 2026
    78.1
    79 gates78 panel- human
    Open live ↗
  8. GPT-6 Luna OpenAI · Oct 2026
    78.0
    79 gates78 panel- human
    Open live ↗
  9. Claude Fable 5.1 Anthropic · Oct 2026
    78.0
    79 gates78 panel- human
    Open live ↗
  10. Claude Sonnet 5.5 Anthropic · Oct 2026
    78.0
    79 gates78 panel- human
    Open live ↗
  11. GLM 5.3 Z.ai · Oct 2026
    78.0
    79 gates78 panel- human
    Open live ↗
  12. Gemini 3.8 Flash Google · Oct 2026
    78.0
    79 gates78 panel- human
    Open live ↗
  13. DeepSeek V4 Pro 0813 DeepSeek · Oct 2026
    78.0
    79 gates78 panel- human
    Open live ↗
  14. Claude Opus 5.5 Anthropic · Oct 2026 · split panel
    77.9
    79 gates78 panel- human
    Open live ↗
  15. Grok 4.7 xAI · Oct 2026
    77.9
    79 gates78 panel- human
    Open live ↗
  16. GPT-5.6 Terra OpenAI · Jul 2026 · split panel
    76.5
    79 gates75 panel- human
    Open live ↗
  17. Claude Fable 5 Anthropic · Jul 2026
    76.4
    79 gates75 panel- human
    Open live ↗
  18. GLM 5.2 Z.ai · Jul 2026
    76.4
    79 gates75 panel- human
    Open live ↗
  19. Qwen 3.8 Max Alibaba · Aug 2026
    76.4
    79 gates75 panel- human
    Open live ↗
  20. Qwen 3.8 Max 0902 Alibaba · Oct 2026
    76.4
    79 gates75 panel- human
    Open live ↗
  21. Claude Opus 5.5 (high) Anthropic · Oct 2026 · split panel
    76.3
    79 gates75 panel- human
    Open live ↗
  22. Qwen 3.7 Max Alibaba · Jul 2026 · split panel
    74.9
    79 gates73 panel- human
    Open live ↗
  23. Muse Spark 1.2 Meta · Aug 2026
    74.9
    79 gates73 panel- human
    Open live ↗
  24. GPT-6.1 Sol OpenAI · Oct 2026
    74.9
    79 gates73 panel- human
    Open live ↗
  25. DeepSeek V4.1 Flash DeepSeek · Oct 2026 · split panel
    74.8
    79 gates73 panel- human
    Open live ↗
  26. Claude Sonnet 5 Anthropic · Jul 2026
    74.7
    79 gates73 panel- human
    Open live ↗
  27. DeepSeek V4 Pro DeepSeek · Jul 2026
    71.6
    79 gates68 panel- human
    Open live ↗
  28. Grok 4.5 xAI · Jul 2026
    71.5
    79 gates68 panel- human
    Open live ↗
  29. GPT-5.6 Luna OpenAI · Jul 2026
    60.8
    80 gates50 panel- human
    Open live ↗
  30. Claude Haiku 4.5 Anthropic · Jul 2026
    60.4
    79 gates50 panel- human
    Open live ↗

Scores are the blended final: 25% browser gates, 45% blind three-lab panel, 30% blind human rating; unrated cells redistribute the human weight. Screenshots are the gate run at 1440×900. "Run it live" loads the untouched deliverable in a sandboxed frame - bugs included. Back to the leaderboard · Compare the top two