View overall rankings across AI models on front-end web development tasks, including agentic coding workflows that require multi-step reasoning and tool use.
Lab Rank | Model Score | Rank Spread | ||
|---|---|---|---|---|
| 1 | Moonshot kimi-k3 · Proprietary | 1682+13/-13 | 1 | 12 |
| 2 | Anthropic claude-opus-5-high · Proprietary | 1673+14/-14 | 2 | 12 |
| 3 | OpenAI gpt-5.6-sol-xhigh (codex-harness) · Proprietary | 1625+10/-10 | 4 | 34 |
| 4 | Z.ai glm-5.2 (max) · MIT | 1587+9/-9 | 5 | 55 |
| 5 | SpaceXAI grok-4.5 · Proprietary | 1549+11/-11 | 9 | 615 |
| 6 | Meta muse-spark-1.1 · Proprietary | 1536+11/-11 | 14 | 920 |
| 7 | Bytedance seed-2.1-pro-preview · Proprietary | 1529+10/-10 | 15 | 1020 |
| 8 | Google gemini-3.6-flash · Proprietary | 1524+13/-13 | 16 | 1122 |
| 9 | Tencent hy3 · Apache 2.0 | 1518+17/-17 | 19 | 1125 |
| 10 | Alibaba qwen3.7-max-20260517 · Proprietary | 1517+8/-8 | 20 | 1422 |
| 11 | MiniMax minimax-m3 · MiniMax Community License | 1493+8/-8 | 24 | 2129 |
| 12 | Xiaomi mimo-v2.5-pro · MIT | 1474+7/-7 | 29 | 2534 |
| 13 | DeepSeek deepseek-v4-pro-thinking · MIT | 1464+7/-7 | 32 | 2837 |
| 14 | Thinky inkling · Apache 2.0 | 1419+10/-10 | 49 | 4153 |
| 15 | Poolside laguna-m.1 · Apache 2.0 | 1349+10/-10 | 73 | 6578 |
| 16 | Mistral mistral-medium-3.5 · Modified MIT | 1267+16/-16 | 87 | 8494 |
| 17 | KAT-Coder-Pro-V1 · Proprietary | 1255+20/-20 | 89 | 8497 |
| 18 | IBM granite-4.1-8b · Apache 2.0 | 1194+19/-19 | 98 | 95102 |
| 19 | Inception AI mercury-2 · Proprietary | 1166+26/-26 | 100 | 97102 |