# Multimodal Reasoning on MMStar (Score %) Source: [Multimodal Reasoning on MMStar (Score %)]() Collection index: [llms.txt]() Data updated: 2026-05-19T07:20:52.616Z Evaluation of model performance on the MMStar dataset for Multimodal Reasoning, focusing specifically on the Score metric reported as a percentage. 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 Reasoning | | Dataset (including any stated split) | MMStar | | Metric | Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Score | Result date | Report paper | | --- | --- | --- | --- | | CAVE-7B | 65.2 | Not recorded | 6a0bee87dbc836b87d82c372 | | InternVL3.5-8B | 64.5 | Not recorded | 6a0bee87dbc836b87d82c372 | | Qwen3-VL-8B-Instruct | 64.3 | Not recorded | 6a0bee87dbc836b87d82c372 | | DeepEyes-7B | 63.6 | Not recorded | 6a0bee87dbc836b87d82c372 | | Pelican-Unified | 63.3 | Not recorded | 6a06a8cadbc836b87d82a374 | | Qwen3-VL-4B-Instruct | 62.9 | Not recorded | 6a06a8cadbc836b87d82a374 | | Qwen2.5-VL | 61.5 | Not recorded | 6a0bee87dbc836b87d82c372 | | IADA +LoRA | 51.2 | Not recorded | 69ca036f0636e5bf4afc929f | | AttnRes+LoRA | 48.4 | Not recorded | 69ca036f0636e5bf4afc929f | | Pre-trained | 45.4 | Not recorded | 69ca036f0636e5bf4afc929f | | LoRA-only | 44.8 | Not recorded | 69ca036f0636e5bf4afc929f | | Gemma3-4B-IT | 37.1 | Not recorded | 6a06a8cadbc836b87d82a374 | | LLaVA-1.5-7B | 33.3 | Not recorded | 69d48f997d0172140ec7434b | | π0.5 | 21.7 | Not recorded | 6a06a8cadbc836b87d82a374 | | MolmoAct | 1.2 | Not recorded | 6a06a8cadbc836b87d82a374 | | LSG | 0.9 | Not recorded | 69d48f997d0172140ec7434b | | LSG | 0.8 | Not recorded | 69d48f997d0172140ec7434b | | LSG | 0.7 | Not recorded | 69d48f997d0172140ec7434b | ### CAVE-7B Score: 65.2. Result ID: 6a0c0f37fbc373592c01ad5c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### InternVL3.5-8B Score: 64.5. Result ID: 6a0c0f37fbc373592c01ad5a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"8B"}. This record does not assert independent reproduction. ### Qwen3-VL-8B-Instruct Score: 64.3. Result ID: 6a0c0f37fbc373592c01ad59. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"8B","Instruct":"true"}. This record does not assert independent reproduction. ### DeepEyes-7B Score: 63.6. Result ID: 6a0c0f37fbc373592c01ad5b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### Pelican-Unified Score: 63.3. Result ID: 6a06eecd47f34ce924ef2008. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Our Unified Model"}. This record does not assert independent reproduction. ### Qwen3-VL-4B-Instruct Score: 62.9. Result ID: 6a06eecd47f34ce924ef2007. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Vision-Language Models"}. This record does not assert independent reproduction. ### Qwen2.5-VL Score: 61.5. Result ID: 6a0c0f37fbc373592c01ad58. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"7B"}. This record does not assert independent reproduction. ### IADA +LoRA Score: 51.2. Result ID: 69ca27e547f34ce924b278cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Trainable Params":"7.7M"}. This record does not assert independent reproduction. ### AttnRes+LoRA Score: 48.4. Result ID: 69ca27e547f34ce924b278cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Trainable Params":"7.7M"}. This record does not assert independent reproduction. ### Pre-trained Score: 45.4. Result ID: 69ca27e547f34ce924b278ca. [Reporting paper]() Recorded evaluation configuration: {"Trainable Params":"0"}. This record does not assert independent reproduction. ### LoRA-only Score: 44.8. Result ID: 69ca27e547f34ce924b278cb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Trainable Params":"7.6M"}. This record does not assert independent reproduction. ### Gemma3-4B-IT Score: 37.1. Result ID: 6a06eecd47f34ce924ef2006. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Vision-Language Models"}. This record does not assert independent reproduction. ### LLaVA-1.5-7B Score: 33.3. Result ID: 69d4b5a847f34ce924bc3a8a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Gating layer configuration":"None (Baseline)"}. This record does not assert independent reproduction. ### π0.5 Score: 21.7. Result ID: 6a06eecd47f34ce924ef2005. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Vision-Language-Action Models"}. This record does not assert independent reproduction. ### MolmoAct Score: 1.2. Result ID: 6a06eecd47f34ce924ef2004. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Vision-Language-Action Models"}. This record does not assert independent reproduction. ### LSG Score: 0.9. Result ID: 69d4b5a847f34ce924bc3a8b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Gating layer configuration":"L2","Base model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### LSG Score: 0.8. Result ID: 69d4b5a847f34ce924bc3a8d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Gating layer configuration":"L19","Base model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### LSG Score: 0.7. Result ID: 69d4b5a847f34ce924bc3a8c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Gating layer configuration":"L10","Base model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.