# MRCR v2 128K-256K 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://modelchoice.dev/leaderboards/benchmarks/reasoning-mrcrv2-128-256

OpenAI MRCR v2 8-needle 128K-256K. Every model the catalog carries a published MRCR v2 128K-256K value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Reasoning |
| Measure | Very-long-context retrieval |
| Tasks | 8-needle retrieval tasks |
| Difficulty | Very long-context reasoning |
| Published by | [Introducing GPT-5.4 mini and nano](https://openai.com/index/introducing-gpt-5-4-mini-and-nano/) |

MRCR v2 slice focused on very long contexts at 128K-256K lengths.

## MRCR v2 128K-256K ranking

3 models with a published MRCR v2 128K-256K 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 | Very-long-context retrieval | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GPT-5.5](https://modelchoice.dev/models/gpt-5-5) | OpenAI | 87.5 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Claude Opus 4.7](https://modelchoice.dev/models/claude-opus-4-7) | Anthropic | 59.2 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Claude Opus 4.7 (Adaptive)](https://modelchoice.dev/models/claude-opus-4-7-max) | Anthropic | 59.2 | Last good 2026-11-18 · 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.
