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

Open Creative Brief

29 models built this brief single-shot. Claude Fable 5 leads at 85.7. 29 of 29 builds carry a blind human rating. 4 split the judge panel.

GitHub repo ↗
↗
  1. Claude Fable 5 Anthropic · Jul 2026
    85.7
    92 gates83 panel85 human
    Open live ↗
  2. GPT-5.6 Sol OpenAI · Jul 2026
    83.8
    85 gates83 panel85 human
    Open live ↗
  3. Claude Opus 5 Anthropic · Jul 2026
    83.5
    100 gates80 panel75 human
    Open live ↗
  4. GPT-6 Astra OpenAI · Oct 2026
    83.2
    85 gates88 panel75 human
    Open live ↗
  5. Kimi K3 Moonshot · Jul 2026
    82.6
    85 gates80 panel85 human
    Open live ↗
  6. Claude Sonnet 5 Anthropic · Jul 2026
    82.1
    93 gates78 panel80 human
    Open live ↗
  7. Claude Fable 5.1 Anthropic · Oct 2026
    81.3
    100 gates85 panel60 human
    Open live ↗
  8. GLM 5.3 Z.ai · Oct 2026
    80.9
    100 gates78 panel70 human
    Open live ↗
  9. Claude Opus 5.5 (high) Anthropic · Oct 2026
    80.3
    90 gates85 panel65 human
    Open live ↗
  10. Claude Opus 5.5 Anthropic · Oct 2026
    80.1
    89 gates85 panel65 human
    Open live ↗
  11. Grok 4.5 xAI · Jul 2026
    78.6
    100 gates73 panel70 human
    Open live ↗
  12. Grok 4.7 xAI · Oct 2026
    78.3
    81 gates83 panel70 human
    Open live ↗
  13. Grok 4.6 xAI · Aug 2026
    77.0
    77 gates75 panel80 human
    Open live ↗
  14. Qwen 3.8 Max Alibaba · Aug 2026
    75.4
    86 gates80 panel60 human
    Open live ↗
  15. GLM 5.2 Z.ai · Jul 2026
    75.3
    97 gates70 panel65 human
    Open live ↗
  16. GPT-5.6 Luna OpenAI · Jul 2026
    73.9
    84 gates68 panel75 human
    Open live ↗
  17. GPT-6.1 Sol OpenAI · Oct 2026
    73.9
    87 gates83 panel50 human
    Open live ↗
  18. Muse Spark 1.3 Meta · Oct 2026
    73.0
    100 gates70 panel55 human
    Open live ↗
  19. DeepSeek V4.1 Flash DeepSeek · Oct 2026
    72.7
    81 gates80 panel55 human
    Open live ↗
  20. Muse Spark 1.2 Meta · Aug 2026
    70.7
    83 gates68 panel65 human
    Open live ↗
  21. Gemini 3.8 Flash Google · Oct 2026
    69.9
    77 gates73 panel60 human
    Open live ↗
  22. Qwen 3.8 Max 0902 Alibaba · Oct 2026
    68.9
    84 gates80 panel40 human
    Open live ↗
  23. GPT-6 Luna OpenAI · Oct 2026
    68.2
    84 gates75 panel45 human
    Open live ↗
  24. Qwen 3.7 Max Alibaba · Jul 2026
    67.6
    66 gates70 panel65 human
    Open live ↗
  25. DeepSeek V4 Pro 0813 DeepSeek · Oct 2026 · split panel
    56.1
    83 gates45 panel50 human
    Open live ↗
  26. Claude Haiku 4.5 Anthropic · Jul 2026
    54.4
    96 gates38 panel45 human
    Open live ↗
  27. DeepSeek V4 Pro DeepSeek · Jul 2026 · split panel
    50.2
    79 gates58 panel15 human
    Open live ↗
  28. GPT-5.6 Terra OpenAI · Jul 2026 · split panel
    32.3
    74 gates28 panel5 human
    Open live ↗
  29. Claude Sonnet 5.5 (high) Anthropic · Oct 2026 · split panel
    28.0
    81 gates18 panel0 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