Anthropic
OpenAI
Google
Meta
xAI
MistralWhich inputs and outputs matter?
Distinguish image understanding, image generation, OCR, visual question answering, and audio. The domain labels are not interchangeable.
Purpose guide · Multimodal
Which AI models belong on a multimodal shortlist?
Catalog current through July 21, 2026
192 models
Source-backed models whose Epoch records explicitly include a multimodal, vision, image, or audio domain. This is a capability map, not a quality ranking.
Selection rule
Included when the Epoch record explicitly lists multimodal, vision, image generation, or audio as a domain. Domain membership does not prove support for every media type or task.
Task-fit shortlist
Evidence first. Your evaluation next.
Ordered by release date for readability, newest first. The order is not a score or recommendation.
- 01
- Matching evidence
- Multimodal
- Domain
- Language · Multimodal
- Access
- API access; closed weights
- 02
- Matching evidence
- Multimodal
- Domain
- Language · Multimodal
- Access
- Unreleased; closed weights
- 03Microsoft
MAI-Image-2.5
Released June 2, 2026
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 04
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 05
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 06
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 07
- Matching evidence
- Audio
- Domain
- Audio
- Access
- API access; closed weights
- 08
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 09Alibaba
Qwen 3.6 Plus
Released April 1, 2026
- Matching evidence
- Multimodal
- Domain
- Language · Multimodal
- Access
- API access; closed weights
- 10Alibaba
Qwen3.5-Omni-Flash
Released March 29, 2026
- Matching evidence
- Multimodal · Vision · Audio
- Domain
- Language · Multimodal · Vision · Audio
- Access
- API access; closed weights
- 11Alibaba
Qwen3.5-Omni-Plus
Released March 29, 2026
- Matching evidence
- Multimodal · Vision · Audio
- Domain
- Language · Multimodal · Vision · Audio
- Access
- API access; closed weights
- 12
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 13
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 14
- Matching evidence
- Vision
- Domain
- Language · Vision
- Access
- API access; closed weights
- 15Alibaba
Qwen3.5 397B-A17B
Released February 13, 2026
- Matching evidence
- Vision
- Domain
- Language · Vision
- Access
- Open weights (unrestricted); open weights
- 16ByteDance
Seedance 2.0
Released February 12, 2026
- Matching evidence
- Image generation · Audio
- Domain
- Image generation · Video · Audio
- Access
- Hosted access (no API); closed weights
- 17
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 18
- Matching evidence
- Multimodal · Vision
- Domain
- Language · Multimodal · Vision
- Access
- API access; closed weights
- 19
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 20
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 21
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 22
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 23
- Matching evidence
- Vision
- Domain
- Video · Vision
- Access
- API access; closed weights
- 24
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 25
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Multimodal
- Access
- API access; closed weights
- 26
- Matching evidence
- Vision
- Domain
- Vision · Language · Speech
- Access
- API access; closed weights
- 27Alibaba
Qwen3-Omni-30B-A3B
Released September 22, 2025
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision · Speech · Video
- Access
- Open weights (unrestricted); open weights
- 28
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 29
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 30
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 31
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 32
- Matching evidence
- Multimodal · Vision
- Domain
- Language · Multimodal · Vision
- Access
- API access; closed weights
- 33Alibaba
Qwen Image
Released August 4, 2025
- Matching evidence
- Image generation · Vision
- Domain
- Image generation · Vision
- Access
- Open weights (unrestricted); open weights
- 34
- Matching evidence
- Multimodal · Vision · Audio
- Domain
- Language · Multimodal · Vision · Video · Audio · Mathematics
- Access
- Hosted access (no API); closed weights
- 35
- Matching evidence
- Multimodal · Vision
- Domain
- Language · Multimodal · Vision
- Access
- API access; closed weights
- 36ByteDance
Seed-1.6-Thinking
Released June 11, 2025
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 37
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Video · Multimodal · Speech
- Access
- API access; closed weights
- 38
- Matching evidence
- Multimodal · Vision
- Domain
- Language · Multimodal · Vision
- Access
- API access; closed weights
- 39
- Matching evidence
- Multimodal · Vision
- Domain
- Language · Multimodal · Vision
- Access
- API access; closed weights
- 40
- Matching evidence
- Vision
- Domain
- Video · Vision
- Access
- API access; closed weights
- 41
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 42ByteDance
Seed1.5-VL
Released May 11, 2025
- Matching evidence
- Vision · Multimodal
- Domain
- Vision · Language · Multimodal · Video
- Access
- API access; closed weights
- 43
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Video · Multimodal · Speech
- Access
- API access; closed weights
- 44
- Matching evidence
- Image generation · Vision
- Domain
- Image generation · Vision
- Access
- API access; closed weights
- 45
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Unreleased; closed weights
- 46
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Open weights (restricted use); open weights
- 47
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Open weights (restricted use); open weights
- 48
- Matching evidence
- Multimodal · Audio · Vision
- Domain
- Multimodal · Language · Audio · Speech · Vision
- Access
- API access; closed weights
- 49
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Video · Multimodal · Speech
- Access
- API access; closed weights
- 50
- Matching evidence
- Multimodal
- Domain
- Language · Mathematics · Multimodal
- Access
- API access; closed weights
- 51Baidu
ERNIE-4.5-VL-424B-A47B (文心大模型4.5)
Released March 16, 2025
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision · Video
- Access
- Open weights (unrestricted); open weights
- 52
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Vision · Language
- Access
- API access; closed weights
- 53
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Multimodal
- Access
- API access; closed weights
- 54
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Multimodal
- Access
- API access; closed weights
- 55
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Multimodal
- Access
- API access; closed weights
- 56
- Matching evidence
- Multimodal · Audio · Vision
- Domain
- Multimodal · Language · Audio · Speech · Vision
- Access
- API access; closed weights
- 57
- Matching evidence
- Vision · Multimodal
- Domain
- Vision · Language · Multimodal
- Access
- Hosted access (no API); closed weights
- 58Moonshot AI
Kimi k1.5
Released January 22, 2025
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 59
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Multimodal
- Access
- API access; closed weights
- 60
- Matching evidence
- Vision
- Domain
- Video · Vision
- Access
- API access; closed weights
- 61
- Matching evidence
- Multimodal · Vision · Audio
- Domain
- Language · Multimodal · Vision · Video · Audio
- Access
- Hosted access (no API); closed weights
- 62
- Matching evidence
- Vision
- Domain
- Video · Vision
- Access
- Hosted access (no API); closed weights
- 63ByteDance
Infinity
Released December 5, 2024
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Open weights (unrestricted); open weights
- 64
- Matching evidence
- Multimodal
- Domain
- Language · Mathematics · Multimodal
- Access
- API access; closed weights
- 65Amazon
Amazon Nova Pro
Released December 3, 2024
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Video · Vision
- Access
- API access; closed weights
- 66NVIDIA
Fugatto 1
Released November 25, 2024
- Matching evidence
- Multimodal · Audio
- Domain
- Multimodal · Language · Audio
- Access
- Unreleased; closed weights
- 67
- Matching evidence
- Multimodal · Audio · Vision
- Domain
- Multimodal · Language · Audio · Speech · Vision
- Access
- API access; closed weights
- 68
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Open weights (non-commercial); open weights
- 69ByteDance
SeedEdit
Released November 11, 2024
- Matching evidence
- Vision
- Domain
- Vision
- Access
- API access; closed weights
- 70NVIDIA
NVLM-D 72B
Released October 22, 2024
- Matching evidence
- Vision
- Domain
- Vision · Language
- Access
- Open weights (non-commercial); open weights
- 71NVIDIA
NVLM-H 72B
Released October 22, 2024
- Matching evidence
- Vision
- Domain
- Vision · Language
- Access
- Unreleased; closed weights
- 72NVIDIA
NVLM-X 72B
Released October 22, 2024
- Matching evidence
- Vision
- Domain
- Vision · Language
- Access
- Unreleased; closed weights
- 73
- Matching evidence
- Vision
- Domain
- Video · Vision
- Access
- Unreleased; closed weights
- 74
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Vision · Language
- Access
- Open weights (restricted use); open weights
- 75ByteDance
PixelDance
Released September 24, 2024
- Matching evidence
- Vision
- Domain
- Video · Vision
- Access
- Hosted access (no API); closed weights
- 76
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Multimodal
- Access
- API access; closed weights
- 77
- Matching evidence
- Multimodal · Audio · Vision
- Domain
- Multimodal · Language · Audio · Speech · Vision
- Access
- API access; closed weights
- 78
- Matching evidence
- Multimodal · Vision
- Domain
- Language · Multimodal · Vision
- Access
- API access; closed weights
- 79Baidu
Ernie 4.0 Turbo
Released June 28, 2024
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 80
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 81
- Matching evidence
- Multimodal · Audio · Vision
- Domain
- Multimodal · Language · Audio · Speech · Vision
- Access
- API access; closed weights
- 82
- Matching evidence
- Multimodal · Vision · Image generation
- Domain
- Multimodal · Vision · Language · Image generation
- Access
- API access; closed weights
- 83
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 84
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 85
- Matching evidence
- Multimodal
- Domain
- Language · Multimodal
- Access
- API access; closed weights
- 86
- Matching evidence
- Vision
- Domain
- Video · Vision
- Access
- Hosted access (no API); closed weights
- 87Alibaba
Qwen-VL-Max
Released January 25, 2024
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 88
- Matching evidence
- Multimodal · Vision · Audio
- Domain
- Multimodal · Language · Vision · Audio
- Access
- API access; closed weights
- 89
- Matching evidence
- Multimodal · Vision · Audio
- Domain
- Multimodal · Language · Vision · Audio
- Access
- Unreleased; closed weights
- 90
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 91
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 92Alibaba
Qwen-Audio-Chat
Released November 14, 2023
- Matching evidence
- Audio
- Domain
- Language · Speech · Audio
- Access
- Open weights (restricted use); open weights
- 93Alibaba
mPLUG-Owl2
Released November 7, 2023
- Matching evidence
- Vision · Multimodal
- Domain
- Vision · Language · Multimodal
- Access
- Open weights (unrestricted); open weights
- 94
- Matching evidence
- Multimodal · Vision · Image generation
- Domain
- Multimodal · Vision · Language · Image generation
- Access
- API access; closed weights
- 95Z.ai
ChatGLM3-6B
Released October 27, 2023
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Open weights (restricted use); open weights
- 96
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 97Baidu
ERNIE 4.0
Released October 17, 2023
- Matching evidence
- Multimodal · Image generation
- Domain
- Multimodal · Language · Video · Image generation
- Access
- API access; closed weights
- 98
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Unreleased; closed weights
- 99Amazon
Amazon Titan
Released September 28, 2023
- Matching evidence
- Image generation
- Domain
- Language · Image generation
- Access
- API access; closed weights
- 100
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Vision · Language
- Access
- API access; closed weights
- 101Alibaba
Qwen-VL
Released August 24, 2023
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Open weights (restricted use); open weights
- 102
- Matching evidence
- Audio
- Domain
- Audio
- Access
- Unreleased; closed weights
- 103
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 104
- Matching evidence
- Audio
- Domain
- Audio
- Access
- Open weights (non-commercial); open weights
- 105
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision · Video
- Access
- Unreleased; closed weights
- 106
- Matching evidence
- Multimodal · Vision · Audio · Image generation
- Domain
- Multimodal · Vision · Audio · Language · Image generation · Speech
- Access
- Open weights (non-commercial); open weights
- 107
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 108
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 109
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- API access; closed weights
- 110
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 111
- Matching evidence
- Audio
- Domain
- Audio
- Access
- Unreleased; closed weights
- 112Microsoft
VALL-E
Released January 5, 2023
- Matching evidence
- Audio
- Domain
- Audio · Speech
- Access
- Unreleased; closed weights
- 113NVIDIA
eDiff-I
Released November 2, 2022
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Unreleased; closed weights
- 114
- Matching evidence
- Audio
- Domain
- Audio
- Access
- Open weights (non-commercial); open weights
- 115
- Matching evidence
- Vision · Multimodal
- Domain
- Language · Vision · Multimodal
- Access
- Unreleased; closed weights
- 116Microsoft
BEIT-3
Released August 22, 2022
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Vision · Language
- Access
- Open weights (unrestricted); open weights
- 117
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Unreleased; closed weights
- 118
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 119Microsoft
MetaLM
Released June 13, 2022
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Unreleased; closed weights
- 120
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 121
- Matching evidence
- Multimodal
- Domain
- Multimodal · Robotics · Games · Language
- Access
- Unreleased; closed weights
- 122
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Vision · Language · Video
- Access
- Unreleased; closed weights
- 123
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Unreleased; closed weights
- 124
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 125
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Unreleased; closed weights
- 126
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Not documented in source; weight status unknown
- 127
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 128Baidu
ERNIE-ViLG
Released December 31, 2021
- Matching evidence
- Multimodal · Image generation · Vision
- Domain
- Multimodal · Image generation · Vision · Language
- Access
- Unreleased; closed weights
- 129Microsoft
Florence
Released November 22, 2021
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 130
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 131
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 132
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (non-commercial); open weights
- 133
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 134
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Vision · Language
- Access
- Unreleased; closed weights
- 135
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 136
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 137
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 138
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Open weights (non-commercial); open weights
- 139Alibaba
M6-T
Released March 5, 2021
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Unreleased; closed weights
- 140
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 141
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Vision · Language · Video
- Access
- Open weights (unrestricted); open weights
- 142
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Vision · Language · Video
- Access
- Open weights (unrestricted); open weights
- 143
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- API access; closed weights
- 144
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 145
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 146
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 147
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 148
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 149
- Matching evidence
- Multimodal · Vision
- Domain
- Multimodal · Language · Vision
- Access
- Unreleased; closed weights
- 150
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 151
- Matching evidence
- Vision · Image generation
- Domain
- Vision · Image generation
- Access
- Open weights (non-commercial); open weights
- 152
- Matching evidence
- Vision · Image generation
- Domain
- Vision · Image generation
- Access
- Open weights (unrestricted); open weights
- 153
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (non-commercial); open weights
- 154
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 155
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (non-commercial); open weights
- 156
- Matching evidence
- Audio
- Domain
- Audio
- Access
- Unreleased; closed weights
- 157NVIDIA
StyleGAN
Released December 12, 2018
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Open weights (non-commercial); open weights
- 158
- Matching evidence
- Image generation
- Domain
- Image generation
- Access
- Unreleased; closed weights
- 159IBM
Big-Little Net
Released July 10, 2018
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 160
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 161
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (non-commercial); open weights
- 162
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 163Baidu
DL scaling Image
Released December 1, 2017
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 164
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 165NVIDIA
ProgressiveGAN
Released October 27, 2017
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 166
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 167
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 168
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 169
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 170
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 171
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 172
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Unreleased; closed weights
- 173
- Matching evidence
- Vision
- Domain
- Vision · Games
- Access
- Not documented in source; weight status unknown
- 174
- Matching evidence
- Vision · Image generation
- Domain
- Vision · Image generation
- Access
- Unreleased; closed weights
- 175
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 176
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 177Microsoft
ResNet-1001
Released September 17, 2016
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 178
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 179
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 180
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 181Microsoft
ResNet-101 (ImageNet)
Released December 10, 2015
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 182Microsoft
ResNet-110 (CIFAR-10)
Released December 10, 2015
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 183Microsoft
ResNet-152 (ImageNet)
Released December 10, 2015
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 184
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 185Microsoft
Faster R-CNN
Released June 4, 2015
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Open weights (unrestricted); open weights
- 186Microsoft
Fast R-CNN
Released April 30, 2015
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 187Microsoft
MSRA (C, PReLU)
Released February 6, 2015
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 188Microsoft
Deeply-supervised nets
Released September 18, 2014
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 189
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 190
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 191
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
- 192IBM
Print Recognition Logic
Released January 1, 1963
- Matching evidence
- Vision
- Domain
- Vision
- Access
- Not documented in source; weight status unknown
Decision criteria
Questions the source can help you ask.
These checks narrow an evaluation plan. They do not replace hands-on testing.
Is the workflow mixed-media?
Test the full handoff between text, files, images, and tools. A multimodal label alone does not establish format support or reliable cross-modal reasoning.
What must you test yourself?
Evaluate media quality, instruction following, file limits, safety, latency, and cost on representative assets. Those measurements are not in this catalog.
Evidence boundary
What this guide does not know.
No task-level quality, price, context window, latency, or reliability measurement is present in the Epoch catalog. We therefore publish no ranking or “best” label.
Catalog source: Epoch AI Data on AI Models, retrieved 2026-07-21, CC-BY. Artifact sha256:80ad08198183693101ceb246a5ea272d64c06c147fc41f72cca8526b32483424.
Each listed model links to its model-level source above. See the publication source ledger and methodology for the wider evidence policy.