LVBench
LVBench is an extreme long video understanding benchmark designed to evaluate multimodal models on videos up to two hours in duration. It contains 6 major categories and 21 subcategories, with videos averaging five times longer than existing datasets. The benchmark addresses applications requiring comprehension of extremely long videos.
Kimi K2.5 from Moonshot AI currently leads the LVBench leaderboard with a score of 0.759 across 20 evaluated AI models.
What LVBench measures
LVBench is a multimodal benchmark that evaluates large language models on long context, multimodal, and vision tasks. LLM Stats tracks 20 models on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.6, with the leader reaching 0.8.
Compare leaders on the best AI for long context, best AI for multimodal and best AI for vision leaderboards.
Publication
- Paper
- LVBench: An Extreme Long Video Understanding Benchmark
- Authors
- Weihan Wang, Zehai He, Wenyi Hong, Yean Cheng, and 8 others
- Published
- arXiv
- 2406.08035
Abstract
Recent progress in multimodal large language models has markedly enhanced the understanding of short videos (typically under one minute), and several evaluation datasets have emerged accordingly. However, these advancements fall short of meeting the demands of real-world applications such as embodied intelligence for long-term decision-making, in-depth movie reviews and discussions, and live sports commentary, all of which require comprehension of long videos spanning several hours. To address this gap, we introduce LVBench, a benchmark specifically designed for long video understanding. Our dataset comprises publicly sourced videos and encompasses a diverse set of tasks aimed at long video comprehension and information extraction. LVBench is designed to challenge multimodal models to demonstrate long-term memory and extended comprehension capabilities. Our extensive evaluations reveal that current multimodal models still underperform on these demanding long video understanding tasks. Through LVBench, we aim to spur the development of more advanced models capable of tackling the complexities of long video comprehension. Our data and code are publicly available at: https://lvbench.github.io.
Kimi K2.5 leads with 75.9%, followed by 
Qwen3.5-122B-A10B at 74.4% and Qwen3.5-27B at 73.6%.
Progress Over Time
Interactive timeline showing model performance evolution on LVBench
LVBench Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 | Moonshot AI | 1.0T | — | — | ||
| 2 |  Alibaba Cloud / Qwen Team | 122B | 262K | $0.40 / $3.20 | ||
| 3 | Alibaba Cloud / Qwen Team | 27B | 262K | $0.30 / $2.40 | ||
| 4 | Alibaba Cloud / Qwen Team | 35B | — | — | ||
| 4 | Alibaba Cloud / Qwen Team | 35B | 262K | $0.25 / $2.00 | ||
| 6 | Alibaba Cloud / Qwen Team | 236B | 262K | $0.30 / $1.50 | ||
| 7 | Alibaba Cloud / Qwen Team | 33B | — | — | ||
| 8 | Alibaba Cloud / Qwen Team | 236B | — | — | ||
| 9 | Alibaba Cloud / Qwen Team | 33B | — | — | ||
| 10 | Alibaba Cloud / Qwen Team | 31B | — | — | ||
| 11 | Alibaba Cloud / Qwen Team | 31B | — | — | ||
| 12 | Alibaba Cloud / Qwen Team | 9B | 262K | $0.08 / $0.50 | ||
| 13 | Alibaba Cloud / Qwen Team | 4B | 262K | $0.10 / $0.60 | ||
| 14 | Alibaba Cloud / Qwen Team | 9B | 262K | $0.18 / $2.09 | ||
| 15 | Alibaba Cloud / Qwen Team | 4B | 262K | $0.10 / $1.00 | ||
| 16 | Alibaba Cloud / Qwen Team | 34B | — | — | ||
| 17 | Alibaba Cloud / Qwen Team | 72B | — | — | ||
| 18 | Alibaba Cloud / Qwen Team | 8B | — | — | ||
| 19 |  Amazon | — | — | — | ||
| 20 | Amazon | — | — | — |
FAQ
Common questions about LVBench.
More evaluations to explore
Related benchmarks in the same category
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
MMMU (Massive Multi-discipline Multimodal Understanding) is a benchmark designed to evaluate multimodal models on college-level subject knowledge and deliberate reasoning. Contains 11.5K meticulously collected multimodal questions from college exams, quizzes, and textbooks, covering six core disciplines: Art & Design, Business, Science, Health & Medicine, Humanities & Social Science, and Tech & Engineering across 30 subjects and 183 subfields.
A more robust multi-discipline multimodal understanding benchmark that enhances MMMU through a three-step process: filtering text-only answerable questions, augmenting candidate options, and introducing vision-only input settings. Achieves significantly lower model performance (16.8-26.9%) compared to original MMMU, providing more rigorous evaluation that closely mimics real-world scenarios.
CharXiv-R is the reasoning component of the CharXiv benchmark, focusing on complex reasoning questions that require synthesizing information across visual chart elements. It evaluates multimodal large language models on their ability to understand and reason about scientific charts from arXiv papers through various reasoning tasks.
MathVista evaluates mathematical reasoning of foundation models in visual contexts. It consists of 6,141 examples derived from 28 existing multimodal datasets and 3 newly created datasets (IQTest, FunctionQA, and PaperQA), combining challenges from diverse mathematical and visual tasks to assess models' ability to understand complex figures and perform rigorous reasoning.