Image-to-Text
Transformers
Safetensors
mistral3
text-generation
ocr
document-understanding
vision-language
pdf
tables
forms
Eval Results
🇪🇺 Region: EU
Instructions to use lightonai/LightOnOCR-1B-1025 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use lightonai/LightOnOCR-1B-1025 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="lightonai/LightOnOCR-1B-1025")# Load model directly from transformers import AutoProcessor, AutoModelForSeq2SeqLM processor = AutoProcessor.from_pretrained("lightonai/LightOnOCR-1B-1025") model = AutoModelForSeq2SeqLM.from_pretrained("lightonai/LightOnOCR-1B-1025", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update .eval_results/olmocrbench.yaml
Browse files
.eval_results/olmocrbench.yaml
CHANGED
|
@@ -6,7 +6,7 @@
|
|
| 6 |
source:
|
| 7 |
url: https://huggingface.co/papers/2601.14251
|
| 8 |
name: LightOnOCR technical report
|
| 9 |
-
user:
|
| 10 |
- dataset:
|
| 11 |
id: allenai/olmOCR-bench
|
| 12 |
task_id: arxiv_math
|
|
@@ -14,7 +14,7 @@
|
|
| 14 |
source:
|
| 15 |
url: https://huggingface.co/papers/2601.14251
|
| 16 |
name: LightOnOCR technical report
|
| 17 |
-
user:
|
| 18 |
- dataset:
|
| 19 |
id: allenai/olmOCR-bench
|
| 20 |
task_id: old_scans_math
|
|
@@ -22,7 +22,7 @@
|
|
| 22 |
source:
|
| 23 |
url: https://huggingface.co/papers/2601.14251
|
| 24 |
name: LightOnOCR technical report
|
| 25 |
-
user:
|
| 26 |
- dataset:
|
| 27 |
id: allenai/olmOCR-bench
|
| 28 |
task_id: table_tests
|
|
@@ -30,7 +30,7 @@
|
|
| 30 |
source:
|
| 31 |
url: https://huggingface.co/papers/2601.14251
|
| 32 |
name: LightOnOCR technical report
|
| 33 |
-
user:
|
| 34 |
- dataset:
|
| 35 |
id: allenai/olmOCR-bench
|
| 36 |
task_id: old_scans
|
|
@@ -38,7 +38,7 @@
|
|
| 38 |
source:
|
| 39 |
url: https://huggingface.co/papers/2601.14251
|
| 40 |
name: LightOnOCR technical report
|
| 41 |
-
user:
|
| 42 |
- dataset:
|
| 43 |
id: allenai/olmOCR-bench
|
| 44 |
task_id: multi_column
|
|
@@ -46,7 +46,7 @@
|
|
| 46 |
source:
|
| 47 |
url: https://huggingface.co/papers/2601.14251
|
| 48 |
name: LightOnOCR technical report
|
| 49 |
-
user:
|
| 50 |
- dataset:
|
| 51 |
id: allenai/olmOCR-bench
|
| 52 |
task_id: long_tiny_text
|
|
@@ -54,7 +54,7 @@
|
|
| 54 |
source:
|
| 55 |
url: https://huggingface.co/papers/2601.14251
|
| 56 |
name: LightOnOCR technical report
|
| 57 |
-
user:
|
| 58 |
- dataset:
|
| 59 |
id: allenai/olmOCR-bench
|
| 60 |
task_id: headers_footers
|
|
@@ -63,7 +63,7 @@
|
|
| 63 |
source:
|
| 64 |
url: https://huggingface.co/papers/2601.14251
|
| 65 |
name: LightOnOCR technical report
|
| 66 |
-
user:
|
| 67 |
- dataset:
|
| 68 |
id: allenai/olmOCR-bench
|
| 69 |
task_id: baseline
|
|
@@ -71,4 +71,4 @@
|
|
| 71 |
source:
|
| 72 |
url: https://huggingface.co/papers/2601.14251
|
| 73 |
name: LightOnOCR technical report
|
| 74 |
-
user:
|
|
|
|
| 6 |
source:
|
| 7 |
url: https://huggingface.co/papers/2601.14251
|
| 8 |
name: LightOnOCR technical report
|
| 9 |
+
user: Bapt120
|
| 10 |
- dataset:
|
| 11 |
id: allenai/olmOCR-bench
|
| 12 |
task_id: arxiv_math
|
|
|
|
| 14 |
source:
|
| 15 |
url: https://huggingface.co/papers/2601.14251
|
| 16 |
name: LightOnOCR technical report
|
| 17 |
+
user: Bapt120
|
| 18 |
- dataset:
|
| 19 |
id: allenai/olmOCR-bench
|
| 20 |
task_id: old_scans_math
|
|
|
|
| 22 |
source:
|
| 23 |
url: https://huggingface.co/papers/2601.14251
|
| 24 |
name: LightOnOCR technical report
|
| 25 |
+
user: Bapt120
|
| 26 |
- dataset:
|
| 27 |
id: allenai/olmOCR-bench
|
| 28 |
task_id: table_tests
|
|
|
|
| 30 |
source:
|
| 31 |
url: https://huggingface.co/papers/2601.14251
|
| 32 |
name: LightOnOCR technical report
|
| 33 |
+
user: Bapt120
|
| 34 |
- dataset:
|
| 35 |
id: allenai/olmOCR-bench
|
| 36 |
task_id: old_scans
|
|
|
|
| 38 |
source:
|
| 39 |
url: https://huggingface.co/papers/2601.14251
|
| 40 |
name: LightOnOCR technical report
|
| 41 |
+
user: Bapt120
|
| 42 |
- dataset:
|
| 43 |
id: allenai/olmOCR-bench
|
| 44 |
task_id: multi_column
|
|
|
|
| 46 |
source:
|
| 47 |
url: https://huggingface.co/papers/2601.14251
|
| 48 |
name: LightOnOCR technical report
|
| 49 |
+
user: Bapt120
|
| 50 |
- dataset:
|
| 51 |
id: allenai/olmOCR-bench
|
| 52 |
task_id: long_tiny_text
|
|
|
|
| 54 |
source:
|
| 55 |
url: https://huggingface.co/papers/2601.14251
|
| 56 |
name: LightOnOCR technical report
|
| 57 |
+
user: Bapt120
|
| 58 |
- dataset:
|
| 59 |
id: allenai/olmOCR-bench
|
| 60 |
task_id: headers_footers
|
|
|
|
| 63 |
source:
|
| 64 |
url: https://huggingface.co/papers/2601.14251
|
| 65 |
name: LightOnOCR technical report
|
| 66 |
+
user: Bapt120
|
| 67 |
- dataset:
|
| 68 |
id: allenai/olmOCR-bench
|
| 69 |
task_id: baseline
|
|
|
|
| 71 |
source:
|
| 72 |
url: https://huggingface.co/papers/2601.14251
|
| 73 |
name: LightOnOCR technical report
|
| 74 |
+
user: Bapt120
|