MAVERIX

MAVERIX (Multimodal Audio-Visual Evaluation Reasoning Index) evaluates multimodal models on tasks that demand tight integration of video and audio information. It features challenges like situational awareness and social sentiment analysis where the answer cannot be reliably determined from a single modality, rigorously testing joint audio-visual understanding.

Nova 2 Omni from Amazon currently leads the MAVERIX leaderboard with a score of 0.666 across 1 evaluated AI models.

Paper
About this benchmark

What MAVERIX measures

MAVERIX is a multimodal benchmark that evaluates large language models on video, vision, multimodal, reasoning, and audio tasks. LLM Stats tracks 1 model on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.7, with the leader reaching 0.7.

Compare leaders on the best AI for video, best AI for vision, best AI for multimodal, best AI for reasoning and best AI for audio leaderboards.

Publication

Paper
MAVERIX: Multimodal Audio-Visual Evaluation and Recognition IndeX
Authors
Liuyue Xie, Avik Kuthiala, George Z. Wei, Ce Zheng, and 11 others
Published

Abstract

We introduce MAVERIX (Multimodal audiovisual Evaluation and Recognition IndeX), a unified benchmark to probe the video understanding in multimodal LLMs, encompassing video, audio, text inputs with human performance baselines. Although recent advancements in models with vision and audio understanding capabilities have shown substantial progress, the field lacks a standardized evaluation framework to thoroughly assess their cross-modality comprehension performance. MAVERIX curates 2,556 questions from 700 videos, in the form of both multiple-choice and open-ended formats, explicitly designed to evaluate multimodal models through questions that necessitate tight integration of video and audio information, spanning a broad spectrum of agentic scenarios. MAVERIX uniquely provides models with audiovisual questions, closely mimicking the multimodal perceptual experiences available to humans during inference and decision-making processes. To our knowledge, MAVERIX is the first benchmark aimed explicitly at assessing comprehensive audiovisual integration in such granularity. Experiments with state-of-the-art models, including Qwen 2.5 Omni and Gemini 2.5 Flash-Lite, show performance around 64% accuracy, while human experts reach near-ceiling performance of 92.8%, exposing a substantial gap to human-level comprehension. With standardized evaluation protocols, a rigorously annotated pipeline, and a public toolkit, MAVERIX establishes a challenging testbed for advancing audiovisual multimodal intelligence.

AmazonNova 2 Omni leads with 66.6%.

Progress Over Time

Interactive timeline showing model performance evolution on MAVERIX

State-of-the-art frontier
Open
Proprietary

MAVERIX Leaderboard

1 models
ContextCostLicense
1
Notice missing or incorrect data?

FAQ

Common questions about MAVERIX.

What is the MAVERIX benchmark?

MAVERIX (Multimodal Audio-Visual Evaluation Reasoning Index) evaluates multimodal models on tasks that demand tight integration of video and audio information. It features challenges like situational awareness and social sentiment analysis where the answer cannot be reliably determined from a single modality, rigorously testing joint audio-visual understanding.

What is the MAVERIX leaderboard?

The MAVERIX leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, Nova 2 Omni by Amazon leads with a score of 0.666. The average score across all models is 0.666.

What is the highest MAVERIX score?

The highest MAVERIX score is 0.666, achieved by Nova 2 Omni from Amazon.

How many models are evaluated on MAVERIX?

1 models have been evaluated on the MAVERIX benchmark, with 0 verified results and 1 self-reported results.

Where can I find the MAVERIX paper?

The MAVERIX paper is available at https://arxiv.org/abs/2503.21699. The paper details the methodology, dataset construction, and evaluation criteria.

What categories does MAVERIX cover?

MAVERIX is categorized under video, vision, multimodal, reasoning, and audio. The benchmark evaluates multimodal models.

How recent are the MAVERIX leaderboard results?

The MAVERIX leaderboard was last updated in June 2026 and currently includes 1 evaluated models.

More evaluations to explore

Related benchmarks in the same category

View all video
GPQA

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.

reasoning
220 models
MMLU-Pro

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.

reasoning
124 models
AIME 2025

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.

reasoning
113 models
MMLU

Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains

reasoning
99 models
SWE-Bench Verified

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.

reasoning
97 models
Humanity's Last Exam

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

reasoningmultimodal
78 models