Benchmark · coding
Code Arena (Elo)
arena.ai's crowdsourced pairwise Elo for web-dev code generation (WebDev / Frontend / Fullstack boards, formerly WebDev Arena). Not the overall Text board (lmsys-arena) and not WebArena's task-success %.
| Date | Model | Score | Source |
|---|---|---|---|
| 2026-08-14 | Gemini 3.7 Flash | 1588.0Elo | Google releases Gemini 3.7 Flash; DeepSeek open-sources Harness |
| 2026-08-13 | Gemini 3.7 Flash | 1588.0Elo | Google introduces Gemini 3.7 Flash with improved coding and lower cost |
| 2026-08-07 | DeepSeek-V4-Flash-0731 | 1577.0Elo | DeepSeek releases V4-Flash-0731, outperforming V4-Pro at lower cost |