// MODEL CATEGORY

Multimodal AI models

Explore multimodal AI models that understand images, text, audio and video together. Compare vision-language models (VLMs) by multimodal benchmark scores and capabilities.

271 models
NY

Cambrian-1-34B

๐Ÿ‡บ๐Ÿ‡ธ New York University (NYU)

Jun 2024 34B
Multimodal Open (Restricted)
NY

Cambrian-1-8B

๐Ÿ‡บ๐Ÿ‡ธ New York University (NYU)

Jun 2024 8B
Multimodal Open (Restricted)
Anthropic

Claude 3.5 Sonnet

๐Ÿ‡บ๐Ÿ‡ธ Anthropic

Jun 2024
Multimodal Proprietary
AL

Ovis-7B

๐Ÿ‡จ๐Ÿ‡ณ Alibaba

Jun 2024 7B
Multimodal Open (Restricted)
IA

Megrez-3B-Omni

๐Ÿ‡จ๐Ÿ‡ณ Infinigence AI

Jun 2024 3B
Multimodal Open (Restricted)
UO

Diamond

๐Ÿ‡บ๐Ÿ‡ธ University of Geneva

May 2024
Multimodal Open (Restricted)
FA

Chameleon-34B

๐Ÿ‡บ๐Ÿ‡ธ Facebook AI Research

May 2024 34B
Multimodal Open Weights
FA

FragLlama: Next-fragment prediction for molecular design

๐Ÿ‡บ๐Ÿ‡ธ Facebook AI Research

May 2024 7B
Multimodal Open Weights
NVIDIA

VILA-7B

๐Ÿ‡บ๐Ÿ‡ธ NVIDIA

May 2024 7B
Multimodal Proprietary
NVIDIA

VILA1.5-40B

๐Ÿ‡บ๐Ÿ‡ธ NVIDIA

May 2024 40B
Multimodal Open Weights
OpenAI

GPT-4o (May 2024)

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

May 2024
Multimodal Proprietary
Google DeepMind

Gemini 1.5 Flash (May 2024)

๐Ÿ‡บ๐Ÿ‡ธ Google DeepMind

May 2024
Multimodal Proprietary
Google DeepMind

Gemini 1.5 Flash 8B

๐Ÿ‡บ๐Ÿ‡ธ Google DeepMind

May 2024 8B
Multimodal Proprietary
BA

Emu2

๐Ÿ‡จ๐Ÿ‡ณ Beijing Academy of Artificial Intelligence / BAAI

May 2024 37B
Multimodal Open Weights
HF

Idefics2

๐Ÿ‡บ๐Ÿ‡ธ Hugging Face

May 2024 8B
Multimodal Open (Restricted)
NVIDIA

VILA1.5-13B

๐Ÿ‡บ๐Ÿ‡ธ NVIDIA

May 2024 13.5B
Multimodal Open Weights
SA

InternVL1.5

๐Ÿ‡จ๐Ÿ‡ณ Shanghai AI Lab

Apr 2024 25.5B
Multimodal Open (Restricted)
"L

Insights into Human Harmony (ๆดž่งไบบๅ’Œ)

๐Ÿ‡จ๐Ÿ‡ณ "Zhejiang Lianxin Digital Co.

Apr 2024
Multimodal Proprietary
RA

Reka Edge

๐Ÿ‡บ๐Ÿ‡ธ Reka AI

Apr 2024 7B
Multimodal Proprietary
Google DeepMind

SIMA

๐Ÿ‡บ๐Ÿ‡ธ Google DeepMind

Apr 2024
Multimodal Proprietary
RA

Reka Core

๐Ÿ‡บ๐Ÿ‡ธ Reka AI

Apr 2024 67B
Multimodal Proprietary
RA

Reka Flash

๐Ÿ‡บ๐Ÿ‡ธ Reka AI

Apr 2024 21B
Multimodal Proprietary
NI

NOMI GPT

๐Ÿ‡จ๐Ÿ‡ณ NIO

Apr 2024
Multimodal Proprietary
OpenAI

GPT-4 Turbo (Apr 2024)

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Apr 2024
Multimodal Proprietary
WA

WeituAI 1.0

๐Ÿ‡จ๐Ÿ‡ณ Weitu AI

Apr 2024 15B
Multimodal Proprietary
"H

APUS-xDAN-4.0(MoE)

๐Ÿ‡จ๐Ÿ‡ณ "Qilin Hesheng Network Technology Co.

Apr 2024 136B
Multimodal Open (Restricted)
GI

POKEยดLLMON

๐Ÿ‡บ๐Ÿ‡ธ Georgia Institute of Technology

Apr 2024
Multimodal Proprietary
xAI

Grok-1.5V

๐Ÿ‡บ๐Ÿ‡ธ xAI

Mar 2024
Multimodal Proprietary
HI

Xinghai (ๆ˜Ÿๆตท)

๐Ÿ‡จ๐Ÿ‡ณ Hisense

Mar 2024
Multimodal Proprietary
Apple

MM1-30B

๐Ÿ‡บ๐Ÿ‡ธ Apple

Mar 2024 30B
Multimodal Proprietary
DE

DeepSeek-VL-1.3B

๐Ÿ‡จ๐Ÿ‡ณ DeepSeek

Mar 2024 1.3B
Multimodal Open (Restricted)
DE

DeepSeek-VL-7B

๐Ÿ‡จ๐Ÿ‡ณ DeepSeek

Mar 2024 7B
Multimodal Open (Restricted)
ByteDance

GroundingGPT

๐Ÿ‡จ๐Ÿ‡ณ ByteDance

Mar 2024 7B
Multimodal Open (Restricted)
Anthropic

Claude 3 Opus

๐Ÿ‡บ๐Ÿ‡ธ Anthropic

Mar 2024
Multimodal Proprietary
Anthropic

Claude 3 Sonnet

๐Ÿ‡บ๐Ÿ‡ธ Anthropic

Mar 2024
Multimodal Proprietary
AL

Ovis2 16B

๐Ÿ‡จ๐Ÿ‡ณ Alibaba

Mar 2024 16B
Multimodal Proprietary
AL

Ovis2 34B

๐Ÿ‡จ๐Ÿ‡ณ Alibaba

Mar 2024 34B
Multimodal Proprietary
DE

Distilled Grandmaster

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Feb 2024 270M
Multimodal Open (Restricted)
AL

RecGPT / Xingchen LLM (ๆท˜ๅฎ๏ผˆไธญๅ›ฝ๏ผ‰่ฝฏไปถๆœ‰้™ๅ…ฌๅธ)

๐Ÿ‡จ๐Ÿ‡ณ Alibaba

Feb 2024 50B
Multimodal Proprietary
"I

Hanhai (็€šๆตท)

๐Ÿ‡จ๐Ÿ‡ณ "Shuchi Information Technology ( Shanghai ) Co.

Jan 2024
Multimodal Proprietary
UO

LLaVA-NeXT-34B (LLaVA-1.6)

๐Ÿ‡บ๐Ÿ‡ธ University of Wisconsin Madison

Jan 2024 34.8B
Multimodal Open (Restricted)
Qwen

Qwen-VL-Max

๐Ÿ‡จ๐Ÿ‡ณ Qwen

Jan 2024 7B
Multimodal Proprietary
AD

Fuyu-Heavy

๐Ÿ‡บ๐Ÿ‡ธ Adept

Jan 2024 100B
Multimodal Proprietary
E(

Yiye Qingzhou-0.7B

๐Ÿ‡จ๐Ÿ‡ณ EFFYIC (่ฏ†ๅ› ๆ™บ่ƒฝ)

Jan 2024 700M
Multimodal Proprietary
E(

Yiye Qingzhou-45B

๐Ÿ‡จ๐Ÿ‡ณ EFFYIC (่ฏ†ๅ› ๆ™บ่ƒฝ)

Jan 2024 45B
Multimodal Proprietary
Google DeepMind

Gemini Nano-1

๐Ÿ‡บ๐Ÿ‡ธ Google DeepMind

Dec 2023 1.8B
Multimodal Proprietary
Google DeepMind

Gemini Nano-2

๐Ÿ‡บ๐Ÿ‡ธ Google DeepMind

Dec 2023 3.3B
Multimodal Proprietary
NVIDIA

VILA-13B

๐Ÿ‡บ๐Ÿ‡ธ NVIDIA

Dec 2023 13.4B
Multimodal Open Weights
Google DeepMind

Gemini 1.0 Pro

๐Ÿ‡บ๐Ÿ‡ธ Google DeepMind

Dec 2023
Multimodal Proprietary
Google DeepMind

Gemini 1.0 Ultra

๐Ÿ‡บ๐Ÿ‡ธ Google DeepMind

Dec 2023
Multimodal Proprietary
CU

OneLLM

๐Ÿ‡บ๐Ÿ‡ธ Chinese University of Hong Kong (CUHK)

Dec 2023 7B
Multimodal Open Weights
AA

OmniFusion-7B (InternViT-6B-448px V1-2)

๐Ÿ‡ท๐Ÿ‡บ AIRI Artificial Intelligence Research Institute

Nov 2023 12.5B
Multimodal Open (Restricted)
FU

LLaVA + LVIS-INSTRUCT4V

๐Ÿ‡จ๐Ÿ‡ณ Fudan University

Nov 2023 13B
Multimodal Open (Restricted)
TE

OmniVec

๐Ÿ‡บ๐Ÿ‡ธ TensorTour

Nov 2023
Multimodal Proprietary
NT

OtterHD-8B

๐Ÿ‡จ๐Ÿ‡ณ Nanyang Technological University

Nov 2023 8B
Multimodal Proprietary
TU

CogVLM-17B

๐Ÿ‡จ๐Ÿ‡ณ Tsinghua University

Nov 2023 17B
Multimodal Open (Restricted)
OpenAI

GPT-4 Turbo (Nov 2023)

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Nov 2023
Multimodal Proprietary
UO

LLaVA 1.5

๐Ÿ‡บ๐Ÿ‡ธ University of Wisconsin Madison

Nov 2023 13B
Multimodal Open (Restricted)
Zhipu AI

ChatGLM3-6B

๐Ÿ‡จ๐Ÿ‡ณ Zhipu AI

Oct 2023 6B
Multimodal Open (Restricted)
DT

Xinghan Foundation Model

๐Ÿ‡จ๐Ÿ‡ณ Dahua Technology

Oct 2023
Multimodal Proprietary

About Multimodal AI models

Multimodal AI models โ€” often called vision-language models (VLMs) โ€” understand more than text: they reason over images, documents, charts, audio and sometimes video alongside language. The best multimodal model depends on your inputs: document understanding rewards strong OCR and layout reasoning, while visual question answering rewards fine-grained image understanding. When comparing VLMs, look at multimodal benchmark scores, supported input types, resolution limits and context window, then weigh quality against cost and latency. Browse the list below to compare multimodal models by provider and availability, check rankings on our benchmarks page, and use compare to put two vision-language models side by side.

Frequently asked questions

How do I choose the right model?

Weigh raw capability against practical constraints like context window, latency, licensing and price. Use the benchmarks page to compare rankings and the compare tool to evaluate two candidates side by side.

Where can I see benchmark scores?

Visit the benchmarks page to compare these models on standardised tests, then use the compare tool for a detailed side-by-side of any two models.