# MuirBench leaderboard

external 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/external-muir-bench

MuirBench: Robust Multi-Image Understanding. Every model the catalog carries a published MuirBench value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | external |
| Measure | Accuracy (%) |
| Tasks | 2,600 pairwise multiple-choice questions across 11,264 images and 12 multi-image tasks |
| Difficulty | Multi-image visual reasoning and robustness |
| Published by | [MuirBench: A Comprehensive Benchmark for Robust Multi-image Understanding](https://arxiv.org/abs/2406.09411) |

A multi-image understanding benchmark of 2,600 multiple-choice questions across 11,264 images and 12 tasks, pairing every answerable question with a minimally altered unanswerable variant.

## MuirBench ranking

2 models with a published MuirBench 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 | Accuracy (%) | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GPT-4o](https://modelscale.dev/models/gpt-4o) | OpenAI | 68.0 | Observed 2026-11-21 · source benchlm:models |
| 2 | [GPT-4 Turbo](https://modelscale.dev/models/gpt-4-turbo) | OpenAI | 62.3 | 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.
