# MRCR v2 (8-needle) (512K–1M) — AI model rankings

mean match ratio for eight needles within the indicated context band

6 tracked models have a published MRCR v2 (8-needle) (512K–1M) score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | GPT-6 Astra | OpenAI | 96.3% | Lab | Sep 3 2026 |
| 2 | GPT-5.5 | OpenAI | 74% | Lab | Apr 23 2026 |
| 3 | GPT-5.6 Sol | OpenAI | 73.8% | Lab | Jun 26 2026 |
| 4 | GPT-5.6 Terra | OpenAI | 72.5% | Lab | Jun 26 2026 |
| 5 | GPT-5.6 Luna | OpenAI | 41.3% | Lab | Jun 26 2026 |
| 6 | GPT-5.4 | OpenAI | 36.6% | Lab | Mar 5 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/mrcr-v2-8needle-512k-1m
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.