# MRCR v2 256K-512K leaderboard

Reasoning benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelscale.dev/leaderboards/benchmarks/reasoning-mrcrv2-256-512

OpenAI MRCR v2 8-needle 256K-512K. Every model the catalog carries a published MRCR v2 256K-512K value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Reasoning |
| Measure | Mean sequence-matcher ratio |
| Tasks | 100 eight-needle retrieval examples |
| Difficulty | Very long-context retrieval |
| Published by | [Muse Spark 1.3 Evaluation Methodology](https://research.meta.ai/static/muse-spark-1-3-multimodal-evaluation-methodology) |

MRCR v2 slice focused on retrieval across 256K-512K-token contexts.

## MRCR v2 256K-512K ranking

3 models with a published MRCR v2 256K-512K value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Mean sequence-matcher ratio | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GPT-6 Astra](https://modelscale.dev/models/gpt-6-astra) | OpenAI | 100.0 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Muse Spark 1.3](https://modelscale.dev/models/muse-spark-1-3) | Meta | 98.5 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Inkling 2 Mini](https://modelscale.dev/models/inkling-2-mini) | Thinking Machines Lab | 46.4 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
