Update README.md
Browse files
README.md
CHANGED
|
@@ -28,6 +28,12 @@ We provide several versions of different sizes, all based on pre-trained models
|
|
| 28 |
|
| 29 |
❗️Note: `NuExtract-2.0-2B` is based on Qwen2-VL rather than Qwen2.5-VL because the smallest Qwen2.5-VL model (3B) has a more restrictive, non-commercial license. We therefore include `NuExtract-2.0-2B` as a small model option that can be used commercially.
|
| 30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
## Overview
|
| 32 |
|
| 33 |
To use the model, provide an input text/image and a JSON template describing the information you need to extract. The template should be a JSON object, specifying field names and their expected type.
|
|
|
|
| 28 |
|
| 29 |
❗️Note: `NuExtract-2.0-2B` is based on Qwen2-VL rather than Qwen2.5-VL because the smallest Qwen2.5-VL model (3B) has a more restrictive, non-commercial license. We therefore include `NuExtract-2.0-2B` as a small model option that can be used commercially.
|
| 30 |
|
| 31 |
+
## Benchmark
|
| 32 |
+
Performance on collection of ~1,000 diverse extraction examples containing both text and image inputs.
|
| 33 |
+
<a href="https://nuextract.ai/">
|
| 34 |
+
<img src="nuextract2_bench.png" width="500"/>
|
| 35 |
+
</a>
|
| 36 |
+
|
| 37 |
## Overview
|
| 38 |
|
| 39 |
To use the model, provide an input text/image and a JSON template describing the information you need to extract. The template should be a JSON object, specifying field names and their expected type.
|