MuirBench
A comprehensive benchmark for robust multi-image understanding capabilities of multimodal LLMs. Consists of 12 diverse multi-image tasks involving 10 categories of multi-image relations (e.g., multiview, temporal relations, narrative, complementary). Comprises 11,264 images and 2,600 multiple-choice questions created in a pairwise manner, where each standard instance is paired with an unanswerable variant for reliable assessment.
Qwen3 VL 32B Thinking from Alibaba Cloud / Qwen Team currently leads the MuirBench leaderboard with a score of 0.803 across 11 evaluated AI models.
Qwen3 VL 32B Thinking leads with 80.3%, followed by Qwen3 VL 235B A22B Thinking at 80.1% and Qwen3 VL 30B A3B Thinking at 77.6%.
Progress Over Time
Interactive timeline showing model performance evolution on MuirBench
MuirBench Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 |  Alibaba Cloud / Qwen Team | 33B | — | — | ||
| 2 | Alibaba Cloud / Qwen Team | 236B | 262K | $0.45 / $3.49 | ||
| 3 | Alibaba Cloud / Qwen Team | 31B | — | — | ||
| 4 | Alibaba Cloud / Qwen Team | 9B | 262K | $0.18 / $2.09 | ||
| 5 | Alibaba Cloud / Qwen Team | 4B | 262K | $0.10 / $1.00 | ||
| 6 | Alibaba Cloud / Qwen Team | 33B | — | — | ||
| 6 | Alibaba Cloud / Qwen Team | 236B | 262K | $0.30 / $1.49 | ||
| 8 | Alibaba Cloud / Qwen Team | 9B | 262K | $0.08 / $0.50 | ||
| 9 | Alibaba Cloud / Qwen Team | 4B | 262K | $0.10 / $0.60 | ||
| 10 | Alibaba Cloud / Qwen Team | 31B | — | — | ||
| 11 | Alibaba Cloud / Qwen Team | 7B | — | — |
FAQ
Common questions about MuirBench.
More evaluations to explore
Related benchmarks in the same category
A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.
A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.
Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions