# PDF parsing with OCSR support on Tiny (test)

Source: [PDF parsing with OCSR support on Tiny (test)](<https://www.sota2.com/research/sota/pdf-parsing-with-ocsr-support-on-tiny-test>)

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

Data updated: 2026-02-26T06:01:15.391Z

Evaluation of PDF parsing capabilities, specifically focusing on performance when Optical Character Recognition (OCSR) is required, measured by recall, OCR success rate, ID matching accuracy, and processing time on the 'Tiny' test set.

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 | PDF parsing with OCSR support |
| Dataset (including any stated split) | Tiny (test) |
| Metric | Recall |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Recall](<https://www.sota2.com/research/sota/pdf-parsing-with-ocsr-support-on-tiny-test.md?metric=Recall>): Higher is better; unit: PERCENT.

- [OCSR Success Rate](<https://www.sota2.com/research/sota/pdf-parsing-with-ocsr-support-on-tiny-test.md?metric=OCSR%20Success%20Rate>): Higher is better; unit: PERCENT.

- [OCSR Accuracy](<https://www.sota2.com/research/sota/pdf-parsing-with-ocsr-support-on-tiny-test.md?metric=OCSR%20Accuracy>): Higher is better; unit: PERCENT.

- [ID Match Accuracy](<https://www.sota2.com/research/sota/pdf-parsing-with-ocsr-support-on-tiny-test.md?metric=ID%20Match%20Accuracy>): Higher is better; unit: PERCENT.

- [Latency (s)](<https://www.sota2.com/research/sota/pdf-parsing-with-ocsr-support-on-tiny-test.md?metric=Latency%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | Recall | Result date | Report paper |
| --- | --- | --- | --- |
| Uni-Parser | 100 | Not recorded | 69438a6d423ea42265729329 |
| MathPix | 100 | Not recorded | 69438a6d423ea42265729329 |
| MinerU.Chem | 66.7 | Not recorded | 69438a6d423ea42265729329 |

### Uni-Parser

Recall: 100. Result ID: 69983a125698665ec495f57c.

[Reporting paper](<https://www.sota2.com/research/paper/69438a6d423ea42265729329.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69438a6d423ea42265729329.md>)

Recorded evaluation configuration: {"Test Date":"2025-08-27"}. This record does not assert independent reproduction.

### MathPix

Recall: 100. Result ID: 69983a125698665ec495f57d.

[Reporting paper](<https://www.sota2.com/research/paper/69438a6d423ea42265729329.md>)

Recorded evaluation configuration: {"Test Date":"2025-08-27"}. This record does not assert independent reproduction.

### MinerU.Chem

Recall: 66.7. Result ID: 69983a125698665ec495f57e.

[Reporting paper](<https://www.sota2.com/research/paper/69438a6d423ea42265729329.md>)

Recorded evaluation configuration: {"Test Date":"2025-09-17"}. This record does not assert independent reproduction.

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