# Spatial Reasoning on MINDCUBE 1.0 (test) Source: [Spatial Reasoning on MINDCUBE 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-04-01T06:41:54.370Z Evaluation of model performance on the test split of the MINDCUBE 1.0 dataset, focusing on Spatial Reasoning capabilities. 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 | Spatial Reasoning | | Dataset (including any stated split) | MINDCUBE 1.0 (test) | | Metric | Overall Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 19 of 19 | | 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. - [Rotation Accuracy](): Higher is better; unit: PERCENT. - [Among Accuracy](): Higher is better; unit: PERCENT. - [Around Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Overall Score | Result date | Report paper | | --- | --- | --- | --- | | DeepSeek-VL2-Small | 47.62 | Not recorded | 68fa79a41e209131cea7ea3f | | GPT-4o-2024-08-06 | 47.59 | Not recorded | 68fa79a41e209131cea7ea3f | | LLaVA-Onevision-7B | 47.43 | Not recorded | 68fa79a41e209131cea7ea3f | | Gemini-1.5-pro-002 | 47.05 | Not recorded | 68fa79a41e209131cea7ea3f | | Gemma-2-9B-it | 46.67 | Not recorded | 68fa79a41e209131cea7ea3f | | mPLUG-Owl3-7B-241101 | 44.85 | Not recorded | 68fa79a41e209131cea7ea3f | | Claude-3.5-Sonnet-20240620 | 44.75 | Not recorded | 68fa79a41e209131cea7ea3f | | LLaVA-Video-Qwen-7B | 41.96 | Not recorded | 68fa79a41e209131cea7ea3f | | Mantis-8B (SigLip) | 41.05 | Not recorded | 68fa79a41e209131cea7ea3f | | InternVL2-8B | 37.5 | Not recorded | 68fa79a41e209131cea7ea3f | | RoboBrain | 37.38 | Not recorded | 68fa79a41e209131cea7ea3f | | Space-Qwen | 33.28 | Not recorded | 68fa79a41e209131cea7ea3f | | Qwen2-VL-2B-Instruct | 33.21 | Not recorded | 68fa79a41e209131cea7ea3f | | Random (frequency) | 33.02 | Not recorded | 68fa79a41e209131cea7ea3f | | Random (chance) | 32.35 | Not recorded | 68fa79a41e209131cea7ea3f | | Spatial-MLLM | 32.06 | Not recorded | 68fa79a41e209131cea7ea3f | | LongVA-7B | 29.46 | Not recorded | 68fa79a41e209131cea7ea3f | | Qwen2-VL-7B-Instruct | 29.26 | Not recorded | 68fa79a41e209131cea7ea3f | | SpaceMantis | 22.81 | Not recorded | 68fa79a41e209131cea7ea3f | ### DeepSeek-VL2-Small Overall Score: 47.62. Result ID: 69ccbc6f47f34ce924b5c7cf. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### GPT-4o-2024-08-06 Overall Score: 47.59. Result ID: 69ccbc6f47f34ce924b5c7d2. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Proprietary Model"}. This record does not assert independent reproduction. ### LLaVA-Onevision-7B Overall Score: 47.43. Result ID: 69ccbc6f47f34ce924b5c7c8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### Gemini-1.5-pro-002 Overall Score: 47.05. Result ID: 69ccbc6f47f34ce924b5c7d3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Proprietary Model"}. This record does not assert independent reproduction. ### Gemma-2-9B-it Overall Score: 46.67. Result ID: 69ccbc6f47f34ce924b5c7d0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### mPLUG-Owl3-7B-241101 Overall Score: 44.85. Result ID: 69ccbc6f47f34ce924b5c7cb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### Claude-3.5-Sonnet-20240620 Overall Score: 44.75. Result ID: 69ccbc6f47f34ce924b5c7d4. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Proprietary Model"}. This record does not assert independent reproduction. ### LLaVA-Video-Qwen-7B Overall Score: 41.96. Result ID: 69ccbc6f47f34ce924b5c7c9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### Mantis-8B (SigLip) Overall Score: 41.05. Result ID: 69ccbc6f47f34ce924b5c7d1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### InternVL2-8B Overall Score: 37.5. Result ID: 69ccbc6f47f34ce924b5c7cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### RoboBrain Overall Score: 37.38. Result ID: 69ccbc6f47f34ce924b5c7d5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Spatial Model"}. This record does not assert independent reproduction. ### Space-Qwen Overall Score: 33.28. Result ID: 69ccbc6f47f34ce924b5c7d8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Spatial Model"}. This record does not assert independent reproduction. ### Qwen2-VL-2B-Instruct Overall Score: 33.21. Result ID: 69ccbc6f47f34ce924b5c7ce. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### Random (frequency) Overall Score: 33.02. Result ID: 69ccbc6f47f34ce924b5c7c7. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Baseline"}. This record does not assert independent reproduction. ### Random (chance) Overall Score: 32.35. Result ID: 69ccbc6f47f34ce924b5c7c6. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Baseline"}. This record does not assert independent reproduction. ### Spatial-MLLM Overall Score: 32.06. Result ID: 69ccbc6f47f34ce924b5c7d7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Spatial Model"}. This record does not assert independent reproduction. ### LongVA-7B Overall Score: 29.46. Result ID: 69ccbc6f47f34ce924b5c7ca. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### Qwen2-VL-7B-Instruct Overall Score: 29.26. Result ID: 69ccbc6f47f34ce924b5c7cd. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Model"}. This record does not assert independent reproduction. ### SpaceMantis Overall Score: 22.81. Result ID: 69ccbc6f47f34ce924b5c7d6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Spatial Model"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.