# Spatial Reasoning on MindCube (Detailed Accuracy Metrics) Source: [Spatial Reasoning on MindCube (Detailed Accuracy Metrics)]() Collection index: [llms.txt]() Data updated: 2026-06-04T09:40:50.441Z Evaluation of model performance on the MindCube dataset for Spatial Reasoning, reporting Overall Accuracy, Rotation Accuracy, Among Accuracy, and Around Accuracy. 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 | | Metric | Overall Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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 Accuracy](): 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 Accuracy | Result date | Report paper | | --- | --- | --- | --- | | DCRL | 43.52 | Not recorded | 6a2106a5c10484bac23f02fc | | Qwen3-VL-8B | 40.01 | Not recorded | 6a2106a5c10484bac23f02fc | | GPT-4o | 38.81 | Not recorded | 6a2106a5c10484bac23f02fc | | RoboBrain | 37.38 | Not recorded | 6a2106a5c10484bac23f02fc | | Idefics3-8B-Llama3 | 35.86 | Not recorded | 6a2106a5c10484bac23f02fc | | LongVA-7B | 29.46 | Not recorded | 6a2106a5c10484bac23f02fc | | Qwen2.5-VL-7B | 29.26 | Not recorded | 6a2106a5c10484bac23f02fc | | SpatialVLM | 22.81 | Not recorded | 6a2106a5c10484bac23f02fc | | InternVL2.5-8B | 18.68 | Not recorded | 6a2106a5c10484bac23f02fc | ### DCRL Overall Accuracy: 43.52. Result ID: 6a214814fbc373592c1d47c4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Our Models"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Overall Accuracy: 40.01. Result ID: 6a214814fbc373592c1d47c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Our Models"}. This record does not assert independent reproduction. ### GPT-4o Overall Accuracy: 38.81. Result ID: 6a214814fbc373592c1d47c0. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Proprietary Model"}. This record does not assert independent reproduction. ### RoboBrain Overall Accuracy: 37.38. Result ID: 6a214814fbc373592c1d47c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Spatial Models"}. This record does not assert independent reproduction. ### Idefics3-8B-Llama3 Overall Accuracy: 35.86. Result ID: 6a214814fbc373592c1d47bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Models"}. This record does not assert independent reproduction. ### LongVA-7B Overall Accuracy: 29.46. Result ID: 6a214814fbc373592c1d47bc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Models"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Overall Accuracy: 29.26. Result ID: 6a214814fbc373592c1d47be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Models"}. This record does not assert independent reproduction. ### SpatialVLM Overall Accuracy: 22.81. Result ID: 6a214814fbc373592c1d47c2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Spatial Models"}. This record does not assert independent reproduction. ### InternVL2.5-8B Overall Accuracy: 18.68. Result ID: 6a214814fbc373592c1d47bd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Open-Weight Multi Image Models"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.