OpenCompass Multi-modal Leaderboard

Real-time updated leaderboard of multi-modal model performance

CommonProductProductivityMulti-modalPerformance Evaluation

The OpenCompass multi-modal leaderboard is a real-time platform for evaluating and ranking different multi-modal models (VLMs). It calculates the average score of models based on 8 multi-modal benchmarks and provides detailed performance data. The platform only includes open-source VLMs or publicly available APIs, aiming to help researchers and developers understand the latest advancements and performance of current multi-modal models.

Provide real-time ranking and performance data for multi-modal models
Support adding new models to VLMEvalKit by creating PRs
Performance evaluation using 8 multi-modal benchmarks
Provide detailed model performance reports including scores for each benchmark
Allow users to obtain help and support by contacting the email address opencompass@pjlab.org.cn

This leaderboard is suitable for researchers
developers
and enterprise decision-makers in the field of artificial intelligence because it provides a transparent and impartial platform to evaluate and compare the performance of different multi-modal models. This helps them choose the most suitable model for their project requirements or understand industry trends and development directions.

Researchers use the leaderboard to compare the performance of different models on specific tasks
Developers leverage leaderboard data to select models suitable for their application scenarios
Enterprise decision-makers refer to the leaderboard to formulate technology procurement or R&D strategies

Visit the OpenCompass multi-modal leaderboard website
View the current ranking and performance data for multi-modal models
Click on a model of interest to obtain a more detailed performance report
Visit VLMEvalKit and create a PR to add new models if needed
Get further assistance and support through the contact information provided on the leaderboard

Visit

OpenCompass Multi-modal Leaderboard Visit Over Time

Monthly Visits

24771

Bounce Rate

40.62%

Page per Visit

3.4

Visit Duration

00:02:44

OpenCompass Multi-modal Leaderboard Visit Trend

OpenCompass Multi-modal Leaderboard Visit Geography

OpenCompass Multi-modal Leaderboard Traffic Sources

OpenCompass Multi-modal Leaderboard Alternatives

SEED-Story — Multi-modal Long-form Story Generation Model

Productivity•Artificial Intelligence•Multi-modal

258

Video-MME — The first comprehensive benchmark for evaluating the performance of Multi-Modal Large Language Models (MLLMs) in video analysis.

Video•Multi-modal•Video Analysis

456

Astris AI — Astris AI is a division launched by Lockheed Martin that focuses on secure artificial intelligence solutions.

Others•Artificial Intelligence•Machine Learning

Recursal AI — Making artificial intelligence accessible

Productivity•Artificial Intelligence•Machine Learning

Apollo-LMMs — Exploration of Video Understanding in Large Multimodal Models

Video•Video Understanding•Multimodal Models

CosyVoice Speech Generation Model 2.0-0.5B — Efficient, multilingual speech synthesis model

Music•Speech Synthesis•Artificial Intelligence

180

CausVid — A fast causal video generator that enables instant video generation.

Video•Video Generation•Artificial Intelligence

150

InternVL 2.5 — Open-source multimodal large language model series

Productivity•multimodal•large language model

OLMo-2-1124-7B-RM — A large language model for text generation and classification.

Productivity•Artificial Intelligence•Natural Language Processing

Amazon Nova — Amazon Nova is Amazon's next-generation foundational model, offering cutting-edge intelligence and industry-leading cost-effectiveness.

InternationalSelection•AWS•Artificial Intelligence

138

HunyuanVideo — A large-scale video generation model training framework open-sourced by Tencent.

Video•Video Generation•Artificial Intelligence

300

OLMo-2-1124-13B-DPO — High-performance English language model suitable for diverse tasks.

Programming•Language Model•Natural Language Processing

ProactiveAgent — A proactive agent based on large language models that predicts user needs and offers assistance.

Programming•Artificial Intelligence•Machine Learning

OLMo 2 — State-of-the-art fully open language model

Programming•Language Model•Natural Language Processing

SoraVids — Archive library for the video generation model Sora.

Video•Video Generation•Machine Learning

LTX-Video — A video generation model based on DiT, capable of real-time high-quality video generation.

Video•Video Generation•Artificial Intelligence

258

AlphaQubit — AI-based decoder for quantum computing error correction

Programming•Quantum Computing•Artificial Intelligence

Qwen Turbo 1M Demo — The Qwen Turbo 1M Demo is a Hugging Face space provided by Qwen.

Productivity•Natural Language Processing•Machine Learning

DataChain — A modern Python data frame library designed specifically for artificial intelligence.

Productivity•Machine Learning•Artificial Intelligence

132

O1-Journey — O1 Journey: Strategic Progress Report Part One

Programming•Artificial Intelligence•Machine Learning

Aya Expanse 32B — A multilingual large language model supporting 23 languages

Productivity•multilingual•text generation

LongVU — Spatiotemporal Adaptation Compression Model for Long Video Language Understanding

Video•Video Understanding•Spatiotemporal Compression

genmoai — Open-source video generation model

Video•Video Generation•Artificial Intelligence

144

Act-One — Generate expressive character performances using simple video input

Video•Artificial Intelligence•Machine Learning

444

RoboticsDiffusionTransformer — A diffusion-based model designed for bimanual manipulation.

Programming•Robotics•Artificial Intelligence

102

Best AI Websites & Tools

OpenCompass Multi-modal Leaderboard

OpenCompass Multi-modal Leaderboard Visit Over Time

OpenCompass Multi-modal Leaderboard Visit Trend

OpenCompass Multi-modal Leaderboard Visit Geography

OpenCompass Multi-modal Leaderboard Traffic Sources

OpenCompass Multi-modal Leaderboard Alternatives

SEED-Story — Multi-modal Long-form Story Generation Model

Video-MME — The first comprehensive benchmark for evaluating the performance of Multi-Modal Large Language Models (MLLMs) in video analysis.

Astris AI — Astris AI is a division launched by Lockheed Martin that focuses on secure artificial intelligence solutions.

Recursal AI — Making artificial intelligence accessible

Apollo-LMMs — Exploration of Video Understanding in Large Multimodal Models

CosyVoice Speech Generation Model 2.0-0.5B — Efficient, multilingual speech synthesis model

CausVid — A fast causal video generator that enables instant video generation.

InternVL 2.5 — Open-source multimodal large language model series

OLMo-2-1124-7B-RM — A large language model for text generation and classification.

Amazon Nova — Amazon Nova is Amazon's next-generation foundational model, offering cutting-edge intelligence and industry-leading cost-effectiveness.

HunyuanVideo — A large-scale video generation model training framework open-sourced by Tencent.

OLMo-2-1124-13B-DPO — High-performance English language model suitable for diverse tasks.

ProactiveAgent — A proactive agent based on large language models that predicts user needs and offers assistance.

OLMo 2 — State-of-the-art fully open language model

SoraVids — Archive library for the video generation model Sora.

LTX-Video — A video generation model based on DiT, capable of real-time high-quality video generation.

AlphaQubit — AI-based decoder for quantum computing error correction

Qwen Turbo 1M Demo — The Qwen Turbo 1M Demo is a Hugging Face space provided by Qwen.

DataChain — A modern Python data frame library designed specifically for artificial intelligence.

O1-Journey — O1 Journey: Strategic Progress Report Part One

Aya Expanse 32B — A multilingual large language model supporting 23 languages

LongVU — Spatiotemporal Adaptation Compression Model for Long Video Language Understanding

genmoai — Open-source video generation model

Act-One — Generate expressive character performances using simple video input

RoboticsDiffusionTransformer — A diffusion-based model designed for bimanual manipulation.

Movie Gen Bench — Video Generation Evaluation Benchmark

ml-retreat — A personal repository of learning notes for advanced AI studies

GR-2 — Advanced General-purpose Robotic Agent

UniMuMo — Unified model for text, music, and motion generation.