# Multimodal Search on MMSearch-Plus single-image Source: [Multimodal Search on MMSearch-Plus single-image]() Collection index: [llms.txt]() Data updated: 2026-04-15T10:22:56.773Z Evaluation of model performance on the single-image subset of the MMSearch-Plus dataset for Multimodal Search tasks, typically measured by Success Rate. 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 | Multimodal Search | | Dataset (including any stated split) | MMSearch-Plus single-image | | Metric | Success Rate | | 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 - [Success Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Success Rate | Result date | Report paper | | --- | --- | --- | --- | | Seed-1.8 | 46.7 | Not recorded | 69df1bc57d0172140ec7bb9b | | Kimi-K2.5 | 39.2 | Not recorded | 69df1bc57d0172140ec7bb9b | | GPT-5 | 34.8 | Not recorded | 69df1bc57d0172140ec7bb9b | | LMM-Searcher-30B | 34.8 | Not recorded | 69df1bc57d0172140ec7bb9b | | LMM-Searcher-30B | 32.9 | Not recorded | 69df1bc57d0172140ec7bb9b | | Vision-DeepResearch-30B | 28.5 | Not recorded | 69df1bc57d0172140ec7bb9b | | Gemini-2.5-Pro | 28.1 | Not recorded | 69df1bc57d0172140ec7bb9b | | REDSearcher-MM-30B | 26.6 | Not recorded | 69df1bc57d0172140ec7bb9b | | GPT-5 | 19.1 | Not recorded | 69df1bc57d0172140ec7bb9b | | Gemini-2.5-Pro | 14.5 | Not recorded | 69df1bc57d0172140ec7bb9b | | Qwen3-VL-30B-A3B-Thinking | 14.4 | Not recorded | 69df1bc57d0172140ec7bb9b | | Gemini-2.5-Flash | 14 | Not recorded | 69df1bc57d0172140ec7bb9b | | Kimi-K2.5 | 12.6 | Not recorded | 69df1bc57d0172140ec7bb9b | | Seed-1.8 | 8.6 | Not recorded | 69df1bc57d0172140ec7bb9b | | Gemini-2.5-Flash | 8.1 | Not recorded | 69df1bc57d0172140ec7bb9b | | Qwen3-VL-30B-A3B-Thinking | 2.7 | Not recorded | 69df1bc57d0172140ec7bb9b | ### Seed-1.8 Success Rate: 46.7. Result ID: 69df659e47f34ce924c8b7cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Agentic Search"}. This record does not assert independent reproduction. ### Kimi-K2.5 Success Rate: 39.2. Result ID: 69df659e47f34ce924c8b7ce. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Agentic Search"}. This record does not assert independent reproduction. ### GPT-5 Success Rate: 34.8. Result ID: 69df659e47f34ce924c8b7cc. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Agentic Search"}. This record does not assert independent reproduction. ### LMM-Searcher-30B Success Rate: 34.8. Result ID: 69df659e47f34ce924c8b7d5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Method Category":"Multimodal Search Agent","Evaluation Turns":"100","Context Management":"Enabled"}. This record does not assert independent reproduction. ### LMM-Searcher-30B Success Rate: 32.9. Result ID: 69df659e47f34ce924c8b7d4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Method Category":"Multimodal Search Agent"}. This record does not assert independent reproduction. ### Vision-DeepResearch-30B Success Rate: 28.5. Result ID: 69df659e47f34ce924c8b7d2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Method Category":"Multimodal Search Agent"}. This record does not assert independent reproduction. ### Gemini-2.5-Pro Success Rate: 28.1. Result ID: 69df659e47f34ce924c8b7cf. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Agentic Search"}. This record does not assert independent reproduction. ### REDSearcher-MM-30B Success Rate: 26.6. Result ID: 69df659e47f34ce924c8b7d3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Method Category":"Multimodal Search Agent"}. This record does not assert independent reproduction. ### GPT-5 Success Rate: 19.1. Result ID: 69df659e47f34ce924c8b7c6. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Direct Answer"}. This record does not assert independent reproduction. ### Gemini-2.5-Pro Success Rate: 14.5. Result ID: 69df659e47f34ce924c8b7c9. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Direct Answer"}. This record does not assert independent reproduction. ### Qwen3-VL-30B-A3B-Thinking Success Rate: 14.4. Result ID: 69df659e47f34ce924c8b7d1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Agentic Search"}. This record does not assert independent reproduction. ### Gemini-2.5-Flash Success Rate: 14. Result ID: 69df659e47f34ce924c8b7d0. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Agentic Search"}. This record does not assert independent reproduction. ### Kimi-K2.5 Success Rate: 12.6. Result ID: 69df659e47f34ce924c8b7c8. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Direct Answer"}. This record does not assert independent reproduction. ### Seed-1.8 Success Rate: 8.6. Result ID: 69df659e47f34ce924c8b7c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Direct Answer"}. This record does not assert independent reproduction. ### Gemini-2.5-Flash Success Rate: 8.1. Result ID: 69df659e47f34ce924c8b7ca. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Direct Answer"}. This record does not assert independent reproduction. ### Qwen3-VL-30B-A3B-Thinking Success Rate: 2.7. Result ID: 69df659e47f34ce924c8b7cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Direct Answer"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.