EmbSpatialBench
Qwen3.5-Omni-Plus 85.4% leads since Mar 2026 Fading Models answer multiple-choice questions on six spatial relations between objects in egocentric embodied scenes. Scored by accuracy.
In the authors’ words
The recent rapid development of Large Vision-Language Models (LVLMs) has indicated their potential for embodied tasks.However, the critical skill of spatial understanding in embodied environments has not been thoroughly evaluated, leaving the gap between current LVLMs and qualified embodied intelligence unknown. Therefore, we construct EmbSpatial-Bench, a benchmark for evaluating embodied spatial understanding of LVLMs.The benchmark is automatically derived from embodied scenes and covers 6 spatial relationships from an egocentric perspective.Experiments expose the insufficient capacity of current LVLMs (even GPT-4V). We further present EmbSpatial-SFT, an instruction-tuning dataset designed to improve LVLMs' embodied spatial understanding.
- Level
- Everyday Most adults, without training
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Multiple choice
Qwen3.5-Omni-Plus leads with 85.4% among models released by Jun 2026
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in Jun 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 16 more
- 1 Qwen3.5-Omni-Plus 85.4%
- 2 Qwen3.6 27B 84.6%
- 3 Qwen3.5 397B-A17B 84.5%
- 4 Qwen3-VL-235B-A22B 84.3%
- 5 Qwen 3.6 35B-A3B 84.3%
- 6 Qwen3.5-122B-A10B 83.9%
- 7
Seed2.1 Pro 83.4% - 8
Keye-VL-2.0-30B-A3B 83.2% - 9
SenseNova-SI-1.3-InternVL3-8B 81% - 10
gemini-robotics-er-1.5-preview 78.4%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Qwen3.5-Omni-Plus | 85.4% | Mar 2026 | Model card | |
| 2 | Qwen3.6 27B | 84.6% | Apr 2026 | Model card | |
| 3 | Qwen3.5 397B-A17B | 84.5% | Feb 2026 | Model card | |
| 4 | Qwen3-VL-235B-A22B | 84.3% | Sep 2025 | Model card | |
| 5 | Qwen 3.6 35B-A3B | 84.3% | Apr 2026 | Model card | |
| 6 | Qwen3.5-122B-A10B | 83.9% | Feb 2026 | Model card | |
| 7 | Seed2.1 Pro | 83.4% | Jun 2026 | Model card | |
| 8 | Keye-VL-2.0-30B-A3B | 83.2% | May 2026 | Model card | |
| 9 | SenseNova-SI-1.3-InternVL3-8B | 81% | Jan 2026 | Model card | |
| 10 | gemini-robotics-er-1.5-preview | 78.4% | Sep 2025 | Model card |
3 models have held number one since its release in Jun 2024
Lab claims count from the day the card came out, independent runs from the model’s release.
The best score passed 85% in Mar 2026
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Labs headline it less often now, last in May 2026
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 25 May 2026 | Keye-VL-2.0-30B-A3B | 83.2 | EmbSpatialBench |
| 22 Apr 2026 | Qwen3.6 27B | 84.6 | EmbSpatialBench |
| 15 Apr 2026 | Qwen 3.6 35B-A3B | 84.3 | EmbSpatialBench |
| 30 Mar 2026 | Qwen3.5-Omni-Plus | 85.4 | EmbSpatialBench |
| 24 Feb 2026 | Qwen3.5-122B-A10B | 83.9 | EmbSpatialBench |
| 16 Feb 2026 | Qwen3.5 397B-A17B | 84.5 | EmbSpatialBench |
| 8 Jan 2026 | SenseNova-SI-1.3-InternVL3-8B | 81 | EmbSpatial-Bench |
| 23 Sep 2025 | Qwen3-VL-235B-A22B | 84.3 | EmbSpatialBench |
5 system cards report it, from 4 labs
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 23 Jun 2026 | Seed2.1 Pro model card | 83.4 | EmbSpatialBench (Table 9) | Capabilities |
| 9 Jun 2026 | Keye-VL-2.0-30B-A3B technical report | 83.2 | EmbSpatialBench (Figure 5, printed value) | Capabilities |
| 17 Apr 2026 | Qwen3.5-Omni-Plus technical report | 85.4 | EmbSpatialBench (Table 6) | Capabilities |
| 26 Nov 2025 | Qwen3-VL-235B-A22B technical report | 84.3 | EmbSpatialBench (Table 2) | Capabilities |
| 25 Sep 2025 | gemini-robotics-er-1.5-preview technical report | 78.4 | EmbSpatial (Table 19, thinking on). Thinking off 73.4 | Capabilities |
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. GPT-5.1 85.4%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. Kimi K3.1 Pro 84.1%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
- ChartQA In use Models answer human-written and generated questions about chart images that need visual and arithmetic reasoning. Scored by relaxed accuracy with a 5 percent numeric tolerance.
Command A Vision 90.9% - DocVQA In use Models answer 50,000 questions about 12,000 scanned document images by reading their text and layout. Scoring is ANLS, a string similarity to reference answers. Seed1.5-VL 96.9%