Long context
MRCR256k-512k
Tests whether the AI can find specific details buried inside a very long document, here across inputs of roughly 256k to 512k tokens — several books' worth of text. Higher is better.
Rankings
Higher is betterMRCR — frequently asked questions
- What is MRCR?
- Tests whether the AI can find specific details buried inside a very long document, here across inputs of roughly 256k to 512k tokens — several books' worth of text. Higher is better.
- Which AI model scores highest on MRCR?
- Muse Spark 1.3 by Meta holds the best MRCR (256k-512k) result among tracked models, at 98.5% (released Sep 2 2026). Higher scores are better on this benchmark.
- What are the top 1 models on MRCR?
- 1. Muse Spark 1.3 (Meta) — 98.5%.
- How many models have a published MRCR score?
- 1 tracked model has a published MRCR (256k-512k) score. Scores are the figures reported by each lab at that model's release, so this page is a record of results over time rather than a re-run leaderboard.