Next.js Agent Evals (Codex) - Success Rate: leaderboard
Vercel's Next.js Agent Evals run coding agents on 31 real Next.js tasks; success is pass@4, infrastructure failures are rerun and evals an agent never ran count as failures. This board is the percentage of evals passed by models in the Codex harness without bundled docs; higher is better.
Metric: Evals passed, pass@4 (%). Source: nextjs.org. Saturation forecast: Around February 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Terra (High) | 100 | #15 (GPT-6.1 Terra) |
| 2 | GPT-6 Sol (High) | 100 | #19 (GPT-6 Sol) |
| 3 | GPT-6 Astra (High) | 93 | #6 (GPT-6 Astra) |
| 4 | GPT-6 Luna (High) | 86 | #61 (GPT-6 Luna) |
| 5 | GPT-5.4 (xHigh) | 68 | #83 (GPT-5.4) |
| 6 | GPT-5.3 Codex (xHigh) | 68 | #72 (GPT-5.3 Codex) |
| 7 | GPT-5.5 Pro | 64 | #23 |
| 8 | GPT-5.2 Codex (xHigh) | 46 | #99 (GPT-5.2 Codex) |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=next-js-agent-evals-codex-success-rate · How It Works · Data refreshed daily, snapshot 2026-11-05.