New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
OpenAI MRCR (1M) benchmark
AI model leaderboard for the OpenAI MRCR (1M) benchmark. Compare how large language models score on OpenAI MRCR (1M), see the full ranking, and understand what this AI benchmark measures. MiniMax-M1-40k currently leads with 58.6%. Multi-round coreference / retrieval benchmark at 1M context length.
Leaderboard
| # | Model | Organization | Score | Variant | Source |
|---|---|---|---|---|---|
| #1 | MiniMax-M1-40k | MiniMax | 58.6% | official ↗ | |
| #2 | MiniMax-M1-80k | MiniMax | 56.2% | official ↗ |
Frequently asked questions about OpenAI MRCR (1M)
What is the OpenAI MRCR (1M) benchmark?
Multi-round coreference / retrieval benchmark at 1M context length.
How is the OpenAI MRCR (1M) benchmark scored?
OpenAI MRCR (1M) is scored using the accuracy (%) metric, where a higher score is better. The maximum achievable score is 100.000. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on OpenAI MRCR (1M)?
As of the latest reported scores on GenAIList, MiniMax-M1-40k achieves the highest result on OpenAI MRCR (1M) with a score of 58.6%.
Is a higher OpenAI MRCR (1M) score better?
Yes. On OpenAI MRCR (1M) a higher score indicates better performance, so models near the top of the leaderboard are the strongest.