Skip to main content
Returns a list of all available model architectures based on config files. These are the built-in model configurations that can be used with create_model.

​
Signature

def list_models() -> list:
    ...

​
Parameters

This function takes no parameters.

​
Returns

​
models
list
List of available model architecture names (strings). Each name corresponds to a model configuration that can be loaded.

​
Example

import open_clip

# Get all available model architectures
models = open_clip.list_models()
print(f"Total models available: {len(models)}")
print(f"First 10 models: {models[:10]}")

# Check if a specific model is available
if 'ViT-B-32' in models:
    print("ViT-B-32 is available")

# Filter models by pattern
vit_models = [m for m in models if 'ViT' in m]
print(f"ViT models: {vit_models}")

resnet_models = [m for m in models if 'RN' in m]
print(f"ResNet models: {resnet_models}")

siglip_models = [m for m in models if 'SigLIP' in m]
print(f"SigLIP models: {siglip_models}")

​
Common Model Architectures

The function returns model names including but not limited to:
  • Vision Transformers: ViT-B-32, ViT-B-16, ViT-L-14, ViT-L-14-336, ViT-H-14, ViT-g-14, ViT-bigG-14
  • ResNet: RN50, RN101, RN50x4, RN50x16, RN50x64
  • ConvNeXt: convnext_base, convnext_base_w, convnext_large_d, convnext_xxlarge
  • SigLIP: ViT-B-16-SigLIP, ViT-L-16-SigLIP-256, ViT-SO400M-14-SigLIP-384
  • EVA: EVA01-g-14, EVA02-B-16, EVA02-L-14, EVA02-E-14
  • CoCa: coca_ViT-B-32, coca_ViT-L-14
  • Multilingual: roberta-ViT-B-32, xlm-roberta-base-ViT-B-32, xlm-roberta-large-ViT-H-14

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us