# Multimodal Spatial Reasoning on Our-Bench SpatialTree-Bench 1.0 (test)

Source: [Multimodal Spatial Reasoning on Our-Bench SpatialTree-Bench 1.0 (test)](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:35:51.970Z

Evaluation of model performance on the test split of Our-Bench SpatialTree-Bench 1.0, focusing on Multimodal Spatial Reasoning capabilities across detailed sub-metrics like Geometry, Motion, and Causal Reasoning.

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 Spatial Reasoning |
| Dataset (including any stated split) | Our-Bench SpatialTree-Bench 1.0 (test) |
| Metric | Average Score |
| 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

- [Average Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

- [Geometry Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Geometry%20Score>): Higher is better; unit: PERCENT.

- [Motion Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Motion%20Score>): Higher is better; unit: PERCENT.

- [Relational Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Relational%20Score>): Higher is better; unit: PERCENT.

- [Local Context Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Local%20Context%20Score>): Higher is better; unit: PERCENT.

- [Orientation Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Orientation%20Score>): Higher is better; unit: PERCENT.

- [Understanding Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Understanding%20Score>): Higher is better; unit: PERCENT.

- [Memory Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Memory%20Score>): Higher is better; unit: PERCENT.

- [Causal Reasoning Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Causal%20Reasoning%20Score>): Higher is better; unit: PERCENT.

- [Sequence Planning Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Sequence%20Planning%20Score>): Higher is better; unit: PERCENT.

- [Goal Execution Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Goal%20Execution%20Score>): Higher is better; unit: PERCENT.

- [Open Explanation Score](<https://www.sota2.com/research/sota/multimodal-spatial-reasoning-on-our-bench-spatialtree-bench-1-0-test.md?metric=Open%20Explanation%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Score | Result date | Report paper |
| --- | --- | --- | --- |
| Gemini3-Flash | 57.8 | Not recorded | 694b739d53fdd5c361739bfd |
| Gemini3-Pro | 56.5 | Not recorded | 694b739d53fdd5c361739bfd |
| Seed1.8 | 50.3 | Not recorded | 694b739d53fdd5c361739bfd |
| Gemini2.5-Pro | 50.1 | Not recorded | 694b739d53fdd5c361739bfd |
| Gemini2.5 Pro | 41.4 | Not recorded | 694b739d53fdd5c361739bfd |
| Seed1.5vl | 41.3 | Not recorded | 694b739d53fdd5c361739bfd |
| Gemini2.5-Flash | 41.1 | Not recorded | 694b739d53fdd5c361739bfd |
| Qwen3VL-235B | 40 | Not recorded | 694b739d53fdd5c361739bfd |
| GLM4.5V | 36 | Not recorded | 694b739d53fdd5c361739bfd |
| Gemini2.5 Flash | 35.8 | Not recorded | 694b739d53fdd5c361739bfd |
| Qwen3VL-30B | 35.3 | Not recorded | 694b739d53fdd5c361739bfd |
| Qwen2.5VL-72B | 33 | Not recorded | 694b739d53fdd5c361739bfd |
| GPT-4o | 31.9 | Not recorded | 694b739d53fdd5c361739bfd |
| Qwen2.5VL-32B | 27.9 | Not recorded | 694b739d53fdd5c361739bfd |
| Qwen2.5VL-7B | 27.5 | Not recorded | 694b739d53fdd5c361739bfd |
| Kimi-VL-A3B | 24.4 | Not recorded | 694b739d53fdd5c361739bfd |

### Gemini3-Flash

Average Score: 57.8. Result ID: 6998141b5698665ec46de4a7.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

Recorded evaluation configuration: {"Thinking Category":"Thinking Models","Model Category":"Proprietary","Rank":1}. This record does not assert independent reproduction.

### Gemini3-Pro

Average Score: 56.5. Result ID: 6998141b5698665ec46de4a6.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

Recorded evaluation configuration: {"Thinking Category":"Thinking Models","Model Category":"Proprietary","Rank":2}. This record does not assert independent reproduction.

### Seed1.8

Average Score: 50.3. Result ID: 6998141b5698665ec46de4a2.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

Recorded evaluation configuration: {"Thinking Category":"Thinking Models","Model Category":"Proprietary","Rank":3}. This record does not assert independent reproduction.

### Gemini2.5-Pro

Average Score: 50.1. Result ID: 6998141b5698665ec46de4a4.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c311e209131cea8cef5.md>)

Recorded evaluation configuration: {"Thinking Category":"Thinking Models","Thinking Mode":"Thinking","Model Category":"Proprietary","Rank":4}. This record does not assert independent reproduction.

### Gemini2.5 Pro

Average Score: 41.4. Result ID: 6998141b5698665ec46de4a0.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c311e209131cea8cef5.md>)

Recorded evaluation configuration: {"Thinking Category":"Non-Thinking","Thinking Mode":"Non-Thinking (NT)","Model Category":"Proprietary","Rank":5}. This record does not assert independent reproduction.

### Seed1.5vl

Average Score: 41.3. Result ID: 6998141b5698665ec46de4a1.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78791e209131cea788a1.md>)

Recorded evaluation configuration: {"Thinking Category":"Thinking Models","Model Category":"Proprietary","Rank":6}. This record does not assert independent reproduction.

### Gemini2.5-Flash

Average Score: 41.1. Result ID: 6998141b5698665ec46de4a5.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c311e209131cea8cef5.md>)

Recorded evaluation configuration: {"Thinking Category":"Thinking Models","Thinking Mode":"Thinking","Model Category":"Proprietary","Rank":7}. This record does not assert independent reproduction.

### Qwen3VL-235B

Average Score: 40. Result ID: 6998141b5698665ec46de4ac.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Thinking Category":"Open-source Models","Rank":8}. This record does not assert independent reproduction.

### GLM4.5V

Average Score: 36. Result ID: 6998141b5698665ec46de4a3.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

Recorded evaluation configuration: {"Thinking Category":"Thinking Models","Model Category":"Proprietary","Rank":9}. This record does not assert independent reproduction.

### Gemini2.5 Flash

Average Score: 35.8. Result ID: 6998141b5698665ec46de49f.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c311e209131cea8cef5.md>)

Recorded evaluation configuration: {"Thinking Category":"Non-Thinking","Thinking Mode":"Non-Thinking (NT)","Model Category":"Proprietary","Rank":10}. This record does not assert independent reproduction.

### Qwen3VL-30B

Average Score: 35.3. Result ID: 6998141b5698665ec46de4ab.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Thinking Category":"Open-source Models","Rank":11}. This record does not assert independent reproduction.

### Qwen2.5VL-72B

Average Score: 33. Result ID: 6998141b5698665ec46de4aa.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Thinking Category":"Open-source Models","Rank":12}. This record does not assert independent reproduction.

### GPT-4o

Average Score: 31.9. Result ID: 6998141b5698665ec46de49e.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75681e209131cea67a98.md>)

Recorded evaluation configuration: {"Thinking Category":"Non-Thinking","Model Category":"Proprietary","Rank":13}. This record does not assert independent reproduction.

### Qwen2.5VL-32B

Average Score: 27.9. Result ID: 6998141b5698665ec46de4a9.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Thinking Category":"Open-source Models","Rank":14}. This record does not assert independent reproduction.

### Qwen2.5VL-7B

Average Score: 27.5. Result ID: 6998141b5698665ec46de4a8.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Thinking Category":"Open-source Models","Rank":15}. This record does not assert independent reproduction.

### Kimi-VL-A3B

Average Score: 24.4. Result ID: 6998141b5698665ec46de4ad.

[Reporting paper](<https://www.sota2.com/research/paper/694b739d53fdd5c361739bfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79a81e209131cea7eb7f.md>)

Recorded evaluation configuration: {"Thinking Category":"Open-source Models","Rank":16}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
