Next.js Evals

Vercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better.

Claude Fable 5.1 by Anthropic holds the best Next.js Evals result among tracked models, at 100% (released Sep 1 2026). Higher scores are better on this benchmark.

1. Claude Fable 5.1 (Anthropic) — 100%; 1. GPT-6 Sol (OpenAI) — 100%; 1. Claude Opus 5.5 (Anthropic) — 100%; 1. Gemini 4 Argon (Google) — 100%; 1. GPT-6.1 Terra (OpenAI) — 100%.

36 tracked models have a published Next.js Evals score. Scores are gathered from Next.js Evals, which runs the benchmark and publishes the full field, and are kept in step with its board.

Muse Spark Open by Meta is the highest-ranked model with downloadable weights on Next.js Evals, scoring 93% at rank 8 overall.