open_clip

src/open_clip/model_configs/ViT-B-16-SigLIP-384.json

+{
+    "embed_dim": 768,
+    "init_logit_bias": -10,
+    "custom_text": true,
+    "vision_cfg": {
+        "image_size": 384,
+        "timm_model_name": "vit_base_patch16_siglip_384",
+        "timm_model_pretrained": false,
+        "timm_pool": "map",
+        "timm_proj": "none"
+    },
+    "text_cfg": {
+        "context_length": 64,
+        "vocab_size": 32000,
+        "hf_tokenizer_name": "timm/ViT-B-16-SigLIP",
+        "tokenizer_kwargs": {
+            "clean": "canonicalize"
+        },
+        "width": 768,
+        "heads": 12,
+        "layers": 12,
+        "no_causal_mask": true,
+        "proj_bias": true,
+        "pool_type": "last",
+        "norm_kwargs":{
+            "eps": 1e-6
+        }
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-16-SigLIP-512.json

+{
+    "embed_dim": 768,
+    "init_logit_bias": -10,
+    "custom_text": true,
+    "vision_cfg": {
+        "image_size": 512,
+        "timm_model_name": "vit_base_patch16_siglip_512",
+        "timm_model_pretrained": false,
+        "timm_pool": "map",
+        "timm_proj": "none"
+    },
+    "text_cfg": {
+        "context_length": 64,
+        "vocab_size": 32000,
+        "hf_tokenizer_name": "timm/ViT-B-16-SigLIP",
+        "tokenizer_kwargs": {
+            "clean": "canonicalize"
+        },
+        "width": 768,
+        "heads": 12,
+        "layers": 12,
+        "no_causal_mask": true,
+        "proj_bias": true,
+        "pool_type": "last",
+        "norm_kwargs":{
+            "eps": 1e-6
+        }
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-16-SigLIP-i18n-256.json

+{
+    "embed_dim": 768,
+    "init_logit_bias": -10,
+    "custom_text": true,
+    "vision_cfg": {
+        "image_size": 256,
+        "timm_model_name": "vit_base_patch16_siglip_256",
+        "timm_model_pretrained": false,
+        "timm_pool": "map",
+        "timm_proj": "none"
+    },
+    "text_cfg": {
+        "context_length": 64,
+        "vocab_size": 250000,
+        "hf_tokenizer_name": "timm/ViT-B-16-SigLIP-i18n-256",
+        "tokenizer_kwargs": {
+            "clean": "canonicalize"
+        },
+        "width": 768,
+        "heads": 12,
+        "layers": 12,
+        "no_causal_mask": true,
+        "proj_bias": true,
+        "pool_type": "last",
+        "norm_kwargs":{
+            "eps": 1e-6
+        }
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-16-SigLIP.json

+{
+    "embed_dim": 768,
+    "init_logit_bias": -10,
+    "custom_text": true,
+    "vision_cfg": {
+        "image_size": 224,
+        "timm_model_name": "vit_base_patch16_siglip_224",
+        "timm_model_pretrained": false,
+        "timm_pool": "map",
+        "timm_proj": "none"
+    },
+    "text_cfg": {
+        "context_length": 64,
+        "vocab_size": 32000,
+        "hf_tokenizer_name": "timm/ViT-B-16-SigLIP",
+        "tokenizer_kwargs": {
+            "clean": "canonicalize"
+        },
+        "width": 768,
+        "heads": 12,
+        "layers": 12,
+        "no_causal_mask": true,
+        "proj_bias": true,
+        "pool_type": "last",
+        "norm_kwargs":{
+            "eps": 1e-6
+        }
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-16-plus-240.json

+{
+    "embed_dim": 640,
+    "vision_cfg": {
+        "image_size": 240,
+        "layers": 12,
+        "width": 896,
+        "patch_size": 16
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 640,
+        "heads": 10,
+        "layers": 12
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-16-plus.json

+{
+    "embed_dim": 640,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 12,
+        "width": 896,
+        "patch_size": 16
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 640,
+        "heads": 10,
+        "layers": 12
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-16-quickgelu.json

+{
+    "embed_dim": 512,
+    "quick_gelu": true,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 12,
+        "width": 768,
+        "patch_size": 16
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 512,
+        "heads": 8,
+        "layers": 12
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-16.json

+{
+    "embed_dim": 512,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 12,
+        "width": 768,
+        "patch_size": 16
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 512,
+        "heads": 8,
+        "layers": 12
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-32-256.json

+{
+    "embed_dim": 512,
+    "vision_cfg": {
+        "image_size": 256,
+        "layers": 12,
+        "width": 768,
+        "patch_size": 32
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 512,
+        "heads": 8,
+        "layers": 12
+    }
+}

src/open_clip/model_configs/ViT-B-32-plus-256.json

+{
+    "embed_dim": 640,
+    "vision_cfg": {
+        "image_size": 256,
+        "layers": 12,
+        "width": 896,
+        "patch_size": 32
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 640,
+        "heads": 10,
+        "layers": 12
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-32-quickgelu.json

+{
+    "embed_dim": 512,
+    "quick_gelu": true,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 12,
+        "width": 768,
+        "patch_size": 32
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 512,
+        "heads": 8,
+        "layers": 12
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-B-32.json

+{
+    "embed_dim": 512,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 12,
+        "width": 768,
+        "patch_size": 32
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 512,
+        "heads": 8,
+        "layers": 12
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-H-14-378-quickgelu.json

+{
+    "embed_dim": 1024,
+    "quick_gelu": true,
+    "vision_cfg": {
+        "image_size": 378,
+        "layers": 32,
+        "width": 1280,
+        "head_width": 80,
+        "patch_size": 14
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 1024,
+        "heads": 16,
+        "layers": 24
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-H-14-CLIPA-336.json

+{
+    "embed_dim": 1024,
+    "vision_cfg": {
+        "image_size": 336,
+        "layers": 32,
+        "width": 1280,
+        "head_width": 80,
+        "patch_size": 14,
+        "no_ln_pre": true,
+        "pool_type": "avg",
+        "final_ln_after_pool": true
+    },
+    "text_cfg": {
+        "context_length": 32,
+        "vocab_size": 32000,
+        "hf_tokenizer_name": "bert-base-uncased",
+        "tokenizer_kwargs": {
+            "strip_sep_token": true
+        },
+        "width": 1024,
+        "heads": 16,
+        "layers": 24,
+        "pool_type": "last",
+        "no_causal_mask": true
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-H-14-CLIPA.json

+{
+    "embed_dim": 1024,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 32,
+        "width": 1280,
+        "head_width": 80,
+        "patch_size": 14,
+        "no_ln_pre": true,
+        "pool_type": "avg",
+        "final_ln_after_pool": true
+    },
+    "text_cfg": {
+        "context_length": 32,
+        "vocab_size": 32000,
+        "hf_tokenizer_name": "bert-base-uncased",
+        "tokenizer_kwargs": {
+            "strip_sep_token": true
+        },
+        "width": 1024,
+        "heads": 16,
+        "layers": 24,
+        "pool_type": "last",
+        "no_causal_mask": true
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-H-14-quickgelu.json

+{
+    "embed_dim": 1024,
+    "quick_gelu": true,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 32,
+        "width": 1280,
+        "head_width": 80,
+        "patch_size": 14
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 1024,
+        "heads": 16,
+        "layers": 24
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-H-14.json

+{
+    "embed_dim": 1024,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 32,
+        "width": 1280,
+        "head_width": 80,
+        "patch_size": 14
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 1024,
+        "heads": 16,
+        "layers": 24
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-H-16.json

+{
+    "embed_dim": 1024,
+    "vision_cfg": {
+        "image_size": 224,
+        "layers": 32,
+        "width": 1280,
+        "head_width": 80,
+        "patch_size": 16
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 1024,
+        "heads": 16,
+        "layers": 24
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-L-14-280.json

+{
+    "embed_dim": 768,
+    "vision_cfg": {
+        "image_size": 280,
+        "layers": 24,
+        "width": 1024,
+        "patch_size": 14
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 768,
+        "heads": 12,
+        "layers": 12
+    }
+}
\ No newline at end of file

src/open_clip/model_configs/ViT-L-14-336.json

+{
+    "embed_dim": 768,
+    "vision_cfg": {
+        "image_size": 336,
+        "layers": 24,
+        "width": 1024,
+        "patch_size": 14
+    },
+    "text_cfg": {
+        "context_length": 77,
+        "vocab_size": 49408,
+        "width": 768,
+        "heads": 12,
+        "layers": 12
+    }
+}
\ No newline at end of file