# Fine-grained visual reasoning on MME Realworld Lite Source: [Fine-grained visual reasoning on MME Realworld Lite]() Collection index: [llms.txt]() Data updated: 2026-04-10T00:19:54.048Z Evaluation of model performance on the Fine-grained visual reasoning task using the MME Realworld Lite dataset, measured by Avg@1. 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 | Fine-grained visual reasoning | | Dataset (including any stated split) | MME Realworld Lite | | Metric | Avg@1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Avg@1](): Higher is better; unit: PERCENT. ## Recorded results | Method | Avg@1 | Result date | Report paper | | --- | --- | --- | --- | | MAPO | 55.8 | Not recorded | 69d732b77d0172140ec76cda | | Ovis2.5-9B + GRPO | 55.5 | Not recorded | 69d732b77d0172140ec76cda | | GPT-5 | 55.3 | Not recorded | 69d732b77d0172140ec76cda | | Thyme | 55.2 | Not recorded | 69d732b77d0172140ec76cda | | Ovis2.5-9B + GSPO | 54.9 | Not recorded | 69d732b77d0172140ec76cda | | Ovis2.5-9B + DAPO | 50.3 | Not recorded | 69d732b77d0172140ec76cda | | Mini-o3 | 49.4 | Not recorded | 69d732b77d0172140ec76cda | | Gemini 2.5 Pro | 49.2 | Not recorded | 69d732b77d0172140ec76cda | | DeepEyes | 48.4 | Not recorded | 69d732b77d0172140ec76cda | | Ovis2.5-9B + Coldstart SFT | 47.9 | Not recorded | 69d732b77d0172140ec76cda | | Ovis2.5-9B | 46.1 | Not recorded | 69d732b77d0172140ec76cda | | Ovis2.5-9B + PPO | 43 | Not recorded | 69d732b77d0172140ec76cda | ### MAPO Avg@1: 55.8. Result ID: 69d8421447f34ce924c04b70. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"backbone":"Ovis2.5-9B"}. This record does not assert independent reproduction. ### Ovis2.5-9B + GRPO Avg@1: 55.5. Result ID: 69d8421447f34ce924c04b6d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training":"GRPO"}. This record does not assert independent reproduction. ### GPT-5 Avg@1: 55.3. Result ID: 69d8421447f34ce924c04b65. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Thyme Avg@1: 55.2. Result ID: 69d8421447f34ce924c04b68. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"results\_source":"original paper"}. This record does not assert independent reproduction. ### Ovis2.5-9B + GSPO Avg@1: 54.9. Result ID: 69d8421447f34ce924c04b6f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"training":"GSPO"}. This record does not assert independent reproduction. ### Ovis2.5-9B + DAPO Avg@1: 50.3. Result ID: 69d8421447f34ce924c04b6e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"training":"DAPO"}. This record does not assert independent reproduction. ### Mini-o3 Avg@1: 49.4. Result ID: 69d8421447f34ce924c04b69. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"reproduced\_by\_authors":"true"}. This record does not assert independent reproduction. ### Gemini 2.5 Pro Avg@1: 49.2. Result ID: 69d8421447f34ce924c04b66. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DeepEyes Avg@1: 48.4. Result ID: 69d8421447f34ce924c04b67. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"reproduced\_by\_authors":"true"}. This record does not assert independent reproduction. ### Ovis2.5-9B + Coldstart SFT Avg@1: 47.9. Result ID: 69d8421447f34ce924c04b6b. [Reporting paper]() Recorded evaluation configuration: {"training":"Supervised Fine-Tuning"}. This record does not assert independent reproduction. ### Ovis2.5-9B Avg@1: 46.1. Result ID: 69d8421447f34ce924c04b6a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Ovis2.5-9B + PPO Avg@1: 43. Result ID: 69d8421447f34ce924c04b6c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"training":"PPO"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.