# Optical Character Recognition on AMEX (test)

Source: [Optical Character Recognition on AMEX (test)](<https://www.sota2.com/research/sota/optical-character-recognition-on-amex-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-09T09:00:34.124Z

Evaluation of Optical Character Recognition performance on the test split of the AMEX dataset.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Optical Character Recognition |
| Dataset (including any stated split) | AMEX (test) |
| Metric | OCR Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [OCR Accuracy](<https://www.sota2.com/research/sota/optical-character-recognition-on-amex-test.md?metric=OCR%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | OCR Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| CoME | 79.67 | Not recorded | 69a519842d603e6de3e3c9b6 |
| Qwen2VL | 78.87 | Not recorded | 69a519842d603e6de3e3c9b6 |
| Ultars | 78.87 | Not recorded | 69a519842d603e6de3e3c9b6 |
| MoE | 78.75 | Not recorded | 69a519842d603e6de3e3c9b6 |

### CoME

OCR Accuracy: 79.67. Result ID: 69ae8c1b6dbb7075b956dd27.

[Reporting paper](<https://www.sota2.com/research/paper/69a519842d603e6de3e3c9b6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a519842d603e6de3e3c9b6.md>)

Recorded evaluation configuration: {"Model Size":"5B"}. This record does not assert independent reproduction.

### Qwen2VL

OCR Accuracy: 78.87. Result ID: 69ae8c1b6dbb7075b956dd24.

[Reporting paper](<https://www.sota2.com/research/paper/69a519842d603e6de3e3c9b6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75151e209131cea65976.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-VL>)

Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction.

### Ultars

OCR Accuracy: 78.87. Result ID: 69ae8c1b6dbb7075b956dd25.

[Reporting paper](<https://www.sota2.com/research/paper/69a519842d603e6de3e3c9b6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a61e209131cea6e2ba.md>)

Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction.

### MoE

OCR Accuracy: 78.75. Result ID: 69ae8c1b6dbb7075b956dd26.

[Reporting paper](<https://www.sota2.com/research/paper/69a519842d603e6de3e3c9b6.md>)

Recorded evaluation configuration: {"Model Size":"A3B"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.