# Real-world Multimodal Evaluation on MME-RW (test) Source: [Real-world Multimodal Evaluation on MME-RW (test)]() Collection index: [llms.txt]() Data updated: 2026-04-07T10:32:07.995Z Evaluation of multimodal reasoning capabilities using the test split of the MME-RW dataset, reporting the overall score. 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 | Real-world Multimodal Evaluation | | Dataset (including any stated split) | MME-RW (test) | | Metric | Overall Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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 - [Overall Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Overall Score | Result date | Report paper | | --- | --- | --- | --- | | Qwen2.5-VL-7B + Saliency-R1 | 62.9 | Not recorded | 69d48fb37d0172140ec744be | | Qwen2.5-VL-7B + SFT | 60.6 | Not recorded | 69d48fb37d0172140ec744be | | Qwen2.5-VL-7B | 58.7 | Not recorded | 69d48fb37d0172140ec744be | | InternVL2.5-8B | 58.1 | Not recorded | 69d48fb37d0172140ec744be | | Qwen2.5-VL-3B + Saliency-R1 | 57.7 | Not recorded | 69d48fb37d0172140ec744be | | LLaVA-OneVision-7B | 57.4 | Not recorded | 69d48fb37d0172140ec744be | | Qwen2.5-VL-3B + SFT | 56 | Not recorded | 69d48fb37d0172140ec744be | | InternVL2-8B | 53.5 | Not recorded | 69d48fb37d0172140ec744be | | Qwen2.5-VL-3B | 52 | Not recorded | 69d48fb37d0172140ec744be | | Claude-3.5 Sonnet | 51.6 | Not recorded | 69d48fb37d0172140ec744be | | IXC-2.5 | 50 | Not recorded | 69d48fb37d0172140ec744be | | MiniCPM-Llama-V-2.5-8B | 45.6 | Not recorded | 69d48fb37d0172140ec744be | | GPT-4o | 45.2 | Not recorded | 69d48fb37d0172140ec744be | | Cambrain-1-8B | 42.7 | Not recorded | 69d48fb37d0172140ec744be | | Gemini-1.5-Pro | 38.2 | Not recorded | 69d48fb37d0172140ec744be | ### Qwen2.5-VL-7B + Saliency-R1 Overall Score: 62.9. Result ID: 69d4dd2447f34ce924bd54c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Stage":"Saliency-R1"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B + SFT Overall Score: 60.6. Result ID: 69d4dd2447f34ce924bd54c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Stage":"SFT"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Overall Score: 58.7. Result ID: 69d4dd2447f34ce924bd54c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Stage":"Base"}. This record does not assert independent reproduction. ### InternVL2.5-8B Overall Score: 58.1. Result ID: 69d4dd2447f34ce924bd54b9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-VL-3B + Saliency-R1 Overall Score: 57.7. Result ID: 69d4dd2447f34ce924bd54bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Stage":"Saliency-R1"}. This record does not assert independent reproduction. ### LLaVA-OneVision-7B Overall Score: 57.4. Result ID: 69d4dd2447f34ce924bd54ba. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-VL-3B + SFT Overall Score: 56. Result ID: 69d4dd2447f34ce924bd54be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Stage":"SFT"}. This record does not assert independent reproduction. ### InternVL2-8B Overall Score: 53.5. Result ID: 69d4dd2447f34ce924bd54b8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-VL-3B Overall Score: 52. Result ID: 69d4dd2447f34ce924bd54bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Stage":"Base"}. This record does not assert independent reproduction. ### Claude-3.5 Sonnet Overall Score: 51.6. Result ID: 69d4dd2447f34ce924bd54b5. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### IXC-2.5 Overall Score: 50. Result ID: 69d4dd2447f34ce924bd54bc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MiniCPM-Llama-V-2.5-8B Overall Score: 45.6. Result ID: 69d4dd2447f34ce924bd54bb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-4o Overall Score: 45.2. Result ID: 69d4dd2447f34ce924bd54b4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Cambrain-1-8B Overall Score: 42.7. Result ID: 69d4dd2447f34ce924bd54b7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Gemini-1.5-Pro Overall Score: 38.2. Result ID: 69d4dd2447f34ce924bd54b6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.