Image-to-Text
Transformers
Safetensors
falcon_ocr
text-generation
falcon
ocr
vision-language
document-understanding
custom_code
Eval Results
Instructions to use tiiuae/Falcon-OCR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use tiiuae/Falcon-OCR with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("image-to-text", model="tiiuae/Falcon-OCR", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("tiiuae/Falcon-OCR", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Add OlmOCRBench evaluation results (#8)
Browse files- Add OlmOCRBench evaluation results (8013e23267e6daf8eb698f0de08280410ea16655)
- Add note (a51b32294cd1f45ced32798332c2ad5900fa56cf)
- Update .eval_results/olmocrbench.yaml (5f0dfb8ea4cbeeb4391e6a2d48e9b48d7ae55b2f)
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
.eval_results/olmocrbench.yaml
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
- dataset:
|
| 2 |
+
id: allenai/olmOCR-bench
|
| 3 |
+
task_id: overall
|
| 4 |
+
value: 80.3
|
| 5 |
+
source:
|
| 6 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 7 |
+
name: Falcon-OCR Model Card
|
| 8 |
+
user: nielsr
|
| 9 |
+
notes: English-subset only
|
| 10 |
+
|
| 11 |
+
- dataset:
|
| 12 |
+
id: allenai/olmOCR-bench
|
| 13 |
+
task_id: arxiv_math
|
| 14 |
+
value: 80.5
|
| 15 |
+
source:
|
| 16 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 17 |
+
name: Falcon-OCR Model Card
|
| 18 |
+
user: nielsr
|
| 19 |
+
notes: English-subset only
|
| 20 |
+
|
| 21 |
+
- dataset:
|
| 22 |
+
id: allenai/olmOCR-bench
|
| 23 |
+
task_id: old_scans_math
|
| 24 |
+
value: 69.2
|
| 25 |
+
source:
|
| 26 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 27 |
+
name: Falcon-OCR Model Card
|
| 28 |
+
user: nielsr
|
| 29 |
+
notes: English-subset only
|
| 30 |
+
|
| 31 |
+
- dataset:
|
| 32 |
+
id: allenai/olmOCR-bench
|
| 33 |
+
task_id: table_tests
|
| 34 |
+
value: 90.3
|
| 35 |
+
source:
|
| 36 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 37 |
+
name: Falcon-OCR Model Card
|
| 38 |
+
user: nielsr
|
| 39 |
+
notes: English-subset only
|
| 40 |
+
|
| 41 |
+
- dataset:
|
| 42 |
+
id: allenai/olmOCR-bench
|
| 43 |
+
task_id: old_scans
|
| 44 |
+
value: 43.5
|
| 45 |
+
source:
|
| 46 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 47 |
+
name: Falcon-OCR Model Card
|
| 48 |
+
user: nielsr
|
| 49 |
+
notes: English-subset only
|
| 50 |
+
|
| 51 |
+
- dataset:
|
| 52 |
+
id: allenai/olmOCR-bench
|
| 53 |
+
task_id: headers_footers
|
| 54 |
+
value: 94.0
|
| 55 |
+
source:
|
| 56 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 57 |
+
name: Falcon-OCR Model Card
|
| 58 |
+
user: nielsr
|
| 59 |
+
notes: English-subset only
|
| 60 |
+
|
| 61 |
+
- dataset:
|
| 62 |
+
id: allenai/olmOCR-bench
|
| 63 |
+
task_id: multi_column
|
| 64 |
+
value: 87.1
|
| 65 |
+
source:
|
| 66 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 67 |
+
name: Falcon-OCR Model Card
|
| 68 |
+
user: nielsr
|
| 69 |
+
notes: English-subset only
|
| 70 |
+
|
| 71 |
+
- dataset:
|
| 72 |
+
id: allenai/olmOCR-bench
|
| 73 |
+
task_id: long_tiny_text
|
| 74 |
+
value: 78.5
|
| 75 |
+
source:
|
| 76 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 77 |
+
name: Falcon-OCR Model Card
|
| 78 |
+
user: nielsr
|
| 79 |
+
notes: English-subset only
|
| 80 |
+
|
| 81 |
+
- dataset:
|
| 82 |
+
id: allenai/olmOCR-bench
|
| 83 |
+
task_id: baseline
|
| 84 |
+
value: 99.5
|
| 85 |
+
source:
|
| 86 |
+
url: https://huggingface.co/tiiuae/Falcon-OCR
|
| 87 |
+
name: Falcon-OCR Model Card
|
| 88 |
+
user: nielsr
|
| 89 |
+
notes: English-subset only
|