View overall rankings across AI models on their ability to generate websites from images and screenshots, alongside agentic coding workflows that involve multi-step reasoning and tool use.
Lab Rank | Model Score | Rank Spread | ||
|---|---|---|---|---|
| 1 | Anthropic claude-opus-5-max · Proprietary | 1664+16/-16 | 1 | 11 |
| 2 | Alibaba qwen3.8-max · Proprietary | 1618+15/-15 | 3 | 24 |
| 3 | OpenAI gpt-5.6-sol-xhigh (codex-harness) · Proprietary | 1606+15/-15 | 4 | 24 |
| 4 | SpaceXAI grok-4.5 · Proprietary | 1574+15/-15 | 6 | 510 |
| 5 | Moonshot kimi-k3-max · Kimi K3 license | 1573+17/-17 | 8 | 510 |
| 6 | Google gemini-3.6-flash-high · Proprietary | 1544+16/-16 | 10 | 619 |
| 7 | Meta muse-spark-1.1 · Proprietary | 1537+13/-13 | 14 | 1020 |
| 8 | Bytedance seed-2.1-pro-preview · Proprietary | 1508+14/-14 | 21 | 1525 |
| 9 | MiniMax minimax-m3 · MiniMax Community License | 1472+12/-12 | 27 | 2431 |
| 10 | Z.ai glm-5v-turbo · Proprietary | 1423+17/-17 | 36 | 3039 |
| 11 | Thinky Inkling Small · Apache 2.0 | 1411+23/-23 | 39 | 3039 |
| 12 | Mistral mistral-large-3 · Apache 2.0 | 1356+20/-20 | 41 | 4043 |