Hugging Face's logo

tiiuae
/
Falcon-OCR

Image-to-Text
Transformers
Safetensors
falcon_ocr
text-generation
falcon
ocr
vision-language
document-understanding
custom_code
Eval Results
Model card Files
xet
 Community
14

Add updated images

#4
by wamreyaz - opened
base: refs/heads/main
←
from: refs/pr/4
Discussion Files changed
+23
-7
Files changed (1) hide show
  1. README.md +23 -7
README.md CHANGED
@@ -42,6 +42,9 @@ print(texts[0])
42
 
43
  > The first `generate()` call is slower due to `torch.compile` building optimized kernels. Subsequent calls are much faster.
44
 
 
 
 
45
  ## Categories
46
 
47
  By default, category is `"plain"` (general text extraction). You can specify a category to use a task-specific prompt:
@@ -97,21 +100,34 @@ for det in results[0]:
97
  ```
98
 
99
  ## Capabilities
100
- ### Handwriting
101
- <img src="https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/zX9GbTqZ2Jbazey4AOBGH.png" width="600" />
102
 
103
- ### Real-world images
104
- <img src="https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/xvyXcFXMfNkCM-3r0_-fu.png" width="600" />
 
 
 
 
105
 
106
  ### Tables
107
 
108
- <img src="https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/3VqNXV6awVZynM0X2QE9U.png" width="600" />
 
 
 
 
 
 
 
 
 
109
 
110
  ### Complex Layout
111
 
112
- <img src="https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/iGCYI2y1eaCj1wMHrmNds.png" width="600" />
113
 
114
- <!-- ![image](https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/iGCYI2y1eaCj1wMHrmNds.png) -->
 
 
 
115
 
116
  ## Citation
117
 
 
42
 
43
  > The first `generate()` call is slower due to `torch.compile` building optimized kernels. Subsequent calls are much faster.
44
 
45
+ By default, we use PagedInference which is quite fast for most interactive tasks, but for large-scale deployment, check the Deployment section
46
+ which provides a vLLM backend.
47
+
48
  ## Categories
49
 
50
  By default, category is `"plain"` (general text extraction). You can specify a category to use a task-specific prompt:
 
100
  ```
101
 
102
  ## Capabilities
 
 
103
 
104
+ ### Handwriting and Real World Images
105
+
106
+ <!-- ![realworld](https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/ySAuT6k3_0s5fJ29FaQYe.png) -->
107
+
108
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/ySAuT6k3_0s5fJ29FaQYe.png" width="600" />
109
+
110
 
111
  ### Tables
112
 
113
+
114
+ <!-- ![tables](https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/2yZjZJAEHVVpd_jfyyDcQ.png) -->
115
+
116
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/2yZjZJAEHVVpd_jfyyDcQ.png" width="600" />
117
+
118
+ ### Formulas
119
+
120
+ <!-- ![formulas](https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/__XMb0GyGO02IPKlQsPQx.png) -->
121
+
122
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/__XMb0GyGO02IPKlQsPQx.png" width="600" />
123
 
124
  ### Complex Layout
125
 
 
126
 
127
+ <!-- ![scientific](https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/kTR7nI7ogEqdI1SQtXpTu.png) -->
128
+
129
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/62fe441427c98b09b503a4e3/kTR7nI7ogEqdI1SQtXpTu.png" width="600" />
130
+
131
 
132
  ## Citation
133