// 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
BA

ERNIE 4.0

๐Ÿ‡จ๐Ÿ‡ณ Baidu

Oct 2023
Multimodal Proprietary
AD

Fuyu-8B

๐Ÿ‡บ๐Ÿ‡ธ Adept

Oct 2023 8B
Multimodal Open Weights
Google DeepMind

PaLI-3

๐Ÿ‡บ๐Ÿ‡ธ Google DeepMind

Oct 2023 5B
Multimodal Proprietary
CU

Ferret (13B)

๐Ÿ‡บ๐Ÿ‡ธ Columbia University

Oct 2023 13B
Multimodal Open Weights
SA

InternLM-XComposer

๐Ÿ‡จ๐Ÿ‡ณ Shanghai AI Lab

Sep 2023 7B
Multimodal Open (Restricted)
OpenAI

GPT-4V

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Sep 2023
Multimodal Proprietary
XJ

DreamLLM

๐Ÿ‡จ๐Ÿ‡ณ Xiโ€™an Jiaotong University

Sep 2023 7B
Multimodal Proprietary
3S

360 Smart Brain

๐Ÿ‡จ๐Ÿ‡ณ 360 Security Technology

Sep 2023
Multimodal Proprietary
Qwen

Qwen-VL

๐Ÿ‡จ๐Ÿ‡ณ Qwen

Aug 2023 9.6B
Multimodal Open (Restricted)
HF

IDEFICS-80B

๐Ÿ‡บ๐Ÿ‡ธ Hugging Face

Aug 2023 80B
Multimodal Open Weights
HF

IDEFICS-9B

๐Ÿ‡บ๐Ÿ‡ธ Hugging Face

Aug 2023 9B
Multimodal Open Weights
LA

EXAONE 2.0

๐Ÿ‡ฐ๐Ÿ‡ท LG AI Research

Jul 2023 300B
Multimodal Proprietary
WL

West Lake (โ€œXฤซhรบ / ่ฅฟๆน–ๅคงๆจกๅž‹โ€)

๐Ÿ‡จ๐Ÿ‡ณ West Lake Xinchen / Xinchen AI / ่ฅฟๆน–ๅฟƒ่พฐ๏ผˆๆญๅทž๏ผ‰็ง‘ๆŠ€ๆœ‰้™ๅ…ฌๅธ

Jul 2023
Multimodal Open Weights
HU

Pangu 3.0

๐Ÿ‡จ๐Ÿ‡ณ Huawei

Jul 2023 100B
Multimodal Proprietary
GR

PaLI-X

๐Ÿ‡บ๐Ÿ‡ธ Google Research

May 2023 55B
Multimodal Proprietary
AL

ONE-PEACE

๐Ÿ‡จ๐Ÿ‡ณ Alibaba

May 2023 4B
Multimodal Open (Restricted)
SR

InstructBLIP

๐Ÿ‡บ๐Ÿ‡ธ Salesforce Research

May 2023 13B
Multimodal Open Weights
MA

ImageBind

๐Ÿ‡บ๐Ÿ‡ธ Meta AI

May 2023 932M
Multimodal Open Weights
NT

Otter

๐Ÿ‡จ๐Ÿ‡ณ Nanyang Technological University

May 2023 1.3B
Multimodal Open (Restricted)
"H

AiLMe-100B v3

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

Apr 2023 100B
Multimodal Proprietary
UO

LLaVA

๐Ÿ‡บ๐Ÿ‡ธ University of Wisconsin Madison

Apr 2023 13B
Multimodal Open (Restricted)
E(

Yiye Qingzhou

๐Ÿ‡จ๐Ÿ‡ณ EFFYIC (่ฏ†ๅ› ๆ™บ่ƒฝ)

Apr 2023 540B
Multimodal Proprietary
OpenAI

GPT-4 (Jun 2023)

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Mar 2023 1.8T
Multimodal Proprietary
OpenAI

GPT-4 (Mar 2023)

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Mar 2023 1.8T
Multimodal Proprietary
Microsoft

Kosmos-1

๐Ÿ‡บ๐Ÿ‡ธ Microsoft

Mar 2023 1.6B
Multimodal Proprietary
SR

Flan T5-XXL + BLIP-2

๐Ÿ‡บ๐Ÿ‡ธ Salesforce Research

Jan 2023 12.1B
Multimodal Open (Restricted)
DE

Adaptive Agent

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Jan 2023 533M
Multimodal Proprietary
DE

DreamerV3

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Jan 2023 200M
Multimodal Proprietary
DE

DeepNash

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Dec 2022
Multimodal Proprietary
MA

CICERO

๐Ÿ‡บ๐Ÿ‡ธ Meta AI

Nov 2022
Multimodal Open Weights
BA

AltCLIP_M9

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

Nov 2022
Multimodal Open (Restricted)
MA

Diplodocus

๐Ÿ‡บ๐Ÿ‡ธ Meta AI

Oct 2022
Multimodal Open Weights
Microsoft

BEIT-3

๐Ÿ‡บ๐Ÿ‡ธ Microsoft

Aug 2022 1.9B
Multimodal Open (Restricted)
AI

Unified-IO (XL)

๐Ÿ‡บ๐Ÿ‡ธ Allen Institute for AI

Jun 2022 2.9B
Multimodal Open (Restricted)
MR

MetaLM

๐Ÿ‡บ๐Ÿ‡ธ Microsoft Research

Jun 2022
Multimodal Proprietary
GO

LIMoE-H/14

๐Ÿ‡บ๐Ÿ‡ธ Google

Jun 2022 5.6B
Multimodal Proprietary
DE

Gato

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

May 2022 1.2B
Multimodal Proprietary
DE

Flamingo

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Apr 2022 80B
Multimodal Proprietary
LG

EXAONE 1.0

๐Ÿ‡ฐ๐Ÿ‡ท LG

Dec 2021 300B
Multimodal Proprietary
DE

Student of Games

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Dec 2021
Multimodal Proprietary
MR

NรœWA

๐Ÿ‡บ๐Ÿ‡ธ Microsoft Research

Nov 2021 870M
Multimodal Proprietary
TU

EfficientZero

๐Ÿ‡จ๐Ÿ‡ณ Tsinghua University

Oct 2021
Multimodal Proprietary
AL

M6-10T

๐Ÿ‡จ๐Ÿ‡ณ Alibaba

Oct 2021 10T
Multimodal Proprietary
CA

Zidong Taichu

๐Ÿ‡จ๐Ÿ‡ณ Chinese Academy of Sciences

Aug 2021 3.2B
Multimodal Open (Restricted)
DE

GOAT

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Jul 2021 3.5M
Multimodal Proprietary
GR

ALIGN

๐Ÿ‡บ๐Ÿ‡ธ Google Research

Jun 2021 820M
Multimodal Proprietary
BA

Wu Dao 2.0

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

May 2021 1.8T
Multimodal Proprietary
AL

M6-T

๐Ÿ‡จ๐Ÿ‡ณ Alibaba

Mar 2021 1T
Multimodal Proprietary
BA

Wu Dao - Wen Lan

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

Mar 2021 1B
Multimodal Proprietary
OpenAI

CLIP (ResNet-50)

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Jan 2021 88.6M
Multimodal Open (Restricted)
OpenAI

CLIP (ViT L/14@336px)

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Jan 2021 370M
Multimodal Open (Restricted)
UA

Go-explore

๐Ÿ‡บ๐Ÿ‡ธ Uber AI

Apr 2020
Multimodal Proprietary
DE

Agent57

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Mar 2020
Multimodal Proprietary
DE

Perceiver IO (optical flow)

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Feb 2020 27.9M
Multimodal Proprietary
OpenAI

OpenAI Five

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Dec 2019 159M
Multimodal Proprietary
OpenAI

OpenAI Five Rerun

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Dec 2019 159M
Multimodal Proprietary
DE

MuZero

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Nov 2019 36.9M
Multimodal Proprietary
DE

AlphaStar

๐Ÿ‡ฌ๐Ÿ‡ง DeepMind

Oct 2019 139M
Multimodal Proprietary
OpenAI

Hide and Seek

๐Ÿ‡บ๐Ÿ‡ธ OpenAI

Sep 2019 1.6M
Multimodal Proprietary
FA

Pluribus

๐Ÿ‡บ๐Ÿ‡ธ Facebook AI Research

Jul 2019
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.