# Optical Character Recognition on Lens Source: [Optical Character Recognition on Lens]() Collection index: [llms.txt]() Data updated: 2026-05-14T08:25:59.123Z Evaluation of Optical Character Recognition performance on the Lens dataset, reporting accuracy. 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) | Lens | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Qwen2.5-VL | 68.51 | Not recorded | 68fa78b91e209131cea7a10f | | Qwen2.5-VL | 65.64 | Not recorded | 68fa78b91e209131cea7a10f | | Gemini2.5-Pro | 61.95 | Not recorded | 68fa78b91e209131cea7a10f | | Qwen2.5-VL | 61.65 | Not recorded | 68fa78b91e209131cea7a10f | | QVQ-Max | 58.67 | Not recorded | 68fa78b91e209131cea7a10f | | InternVL3 | 54.21 | Not recorded | 68fa78b91e209131cea7a10f | | Qwen2.5-VL | 52.43 | Not recorded | 68fa78b91e209131cea7a10f | | InternVL3 | 51.72 | Not recorded | 68fa78b91e209131cea7a10f | | InternVL3 | 48.79 | Not recorded | 68fa78b91e209131cea7a10f | | Deepseek-VL2 | 48.76 | Not recorded | 68fa78b91e209131cea7a10f | | InternVL3 | 45.51 | Not recorded | 68fa78b91e209131cea7a10f | | Deepseek-VL2-tiny | 44.01 | Not recorded | 68fa78b91e209131cea7a10f | | GPT-4o | 42.86 | Not recorded | 68fa78b91e209131cea7a10f | | Gemma3 | 33.98 | Not recorded | 68fa78b91e209131cea7a10f | | Kimi-VL-thinking | 30.21 | Not recorded | 68fa78b91e209131cea7a10f | | Gemma3 | 27.03 | Not recorded | 68fa78b91e209131cea7a10f | ### Qwen2.5-VL Accuracy: 68.51. Result ID: 6a05870447f34ce924ec7fd1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"72B"}. This record does not assert independent reproduction. ### Qwen2.5-VL Accuracy: 65.64. Result ID: 6a05870447f34ce924ec7fd0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"32B"}. This record does not assert independent reproduction. ### Gemini2.5-Pro Accuracy: 61.95. Result ID: 6a05870447f34ce924ec7fc5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"-"}. This record does not assert independent reproduction. ### Qwen2.5-VL Accuracy: 61.65. Result ID: 6a05870447f34ce924ec7fcf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"7B"}. This record does not assert independent reproduction. ### QVQ-Max Accuracy: 58.67. Result ID: 6a05870447f34ce924ec7fd2. [Reporting paper]() Recorded evaluation configuration: {"Model size":"72B"}. This record does not assert independent reproduction. ### InternVL3 Accuracy: 54.21. Result ID: 6a05870447f34ce924ec7fcd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"78B"}. This record does not assert independent reproduction. ### Qwen2.5-VL Accuracy: 52.43. Result ID: 6a05870447f34ce924ec7fce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"3B"}. This record does not assert independent reproduction. ### InternVL3 Accuracy: 51.72. Result ID: 6a05870447f34ce924ec7fcc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"38B"}. This record does not assert independent reproduction. ### InternVL3 Accuracy: 48.79. Result ID: 6a05870447f34ce924ec7fcb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"9B"}. This record does not assert independent reproduction. ### Deepseek-VL2 Accuracy: 48.76. Result ID: 6a05870447f34ce924ec7fc7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"MoE 4.5B/27B"}. This record does not assert independent reproduction. ### InternVL3 Accuracy: 45.51. Result ID: 6a05870447f34ce924ec7fca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"2B"}. This record does not assert independent reproduction. ### Deepseek-VL2-tiny Accuracy: 44.01. Result ID: 6a05870447f34ce924ec7fc6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"MoE 1B/3B"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 42.86. Result ID: 6a05870447f34ce924ec7fc4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model size":"-"}. This record does not assert independent reproduction. ### Gemma3 Accuracy: 33.98. Result ID: 6a05870447f34ce924ec7fc9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"12B"}. This record does not assert independent reproduction. ### Kimi-VL-thinking Accuracy: 30.21. Result ID: 6a05870447f34ce924ec7fd3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"MoE 2.8B/16B"}. This record does not assert independent reproduction. ### Gemma3 Accuracy: 27.03. Result ID: 6a05870447f34ce924ec7fc8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model size":"4B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.