# Spatial Reasoning on 3DSRBench (LLM-Judge)

Source: [Spatial Reasoning on 3DSRBench (LLM-Judge)](<https://www.sota2.com/research/sota/spatial-reasoning-on-3dsrbench-llm-judge>)

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

Data updated: 2026-04-21T10:19:04.562Z

Evaluation of model performance on the 3DSRBench dataset for Spatial Reasoning tasks, measured specifically using LLM-Judge 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) | 3DSRBench |
| Metric | LLM-Judge Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [LLM-Judge Accuracy](<https://www.sota2.com/research/sota/spatial-reasoning-on-3dsrbench-llm-judge.md?metric=LLM-Judge%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | LLM-Judge Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SpatialImaginer | 54.3 | Not recorded | 69e704a07d0172140ec7ed9d |
| VLM-3R-7B | 51.5 | Not recorded | 69e704a07d0172140ec7ed9d |
| ViLaSR-7B | 46.6 | Not recorded | 69e704a07d0172140ec7ed9d |
| Cambrian-S-7B | 45 | Not recorded | 69e704a07d0172140ec7ed9d |
| SpatialLadder-3B | 42.8 | Not recorded | 69e704a07d0172140ec7ed9d |
| SpaceR-7B | 40.5 | Not recorded | 69e704a07d0172140ec7ed9d |
| Intern8B | 37.2 | Not recorded | 69c21a640636e5bf4afc463a |
| Intern8B | 36.58 | Not recorded | 69c21a640636e5bf4afc463a |
| Spatial-MLLM-4B | 36.2 | Not recorded | 69e704a07d0172140ec7ed9d |
| Qwen7B | 34.95 | Not recorded | 69c21a640636e5bf4afc463a |
| Qwen7B | 34.16 | Not recorded | 69c21a640636e5bf4afc463a |
| Llama11B | 29.4 | Not recorded | 69c21a640636e5bf4afc463a |
| Llama11B | 28.87 | Not recorded | 69c21a640636e5bf4afc463a |
| MindCube-3B | 2.8 | Not recorded | 69e704a07d0172140ec7ed9d |

### SpatialImaginer

LLM-Judge Accuracy: 54.3. Result ID: 69e74f0847f34ce924ce2e08.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### VLM-3R-7B

LLM-Judge Accuracy: 51.5. Result ID: 69e74f0847f34ce924ce2e01.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa790d1e209131cea7be1f.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ViLaSR-7B

LLM-Judge Accuracy: 46.6. Result ID: 69e74f0847f34ce924ce2e06.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79831e209131cea7defc.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Cambrian-S-7B

LLM-Judge Accuracy: 45. Result ID: 69e74f0847f34ce924ce2e07.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/690d7cfb2828b7cc6937cd50.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SpatialLadder-3B

LLM-Judge Accuracy: 42.8. Result ID: 69e74f0847f34ce924ce2e03.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bfb1e209131cea8b978.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SpaceR-7B

LLM-Judge Accuracy: 40.5. Result ID: 69e74f0847f34ce924ce2e05.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a41e209131cea798a1.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Intern8B

LLM-Judge Accuracy: 37.2. Result ID: 69c2683447f34ce924aa5c1a.

[Reporting paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

Recorded evaluation configuration: {"Inter Head":"after"}. This record does not assert independent reproduction.

### Intern8B

LLM-Judge Accuracy: 36.58. Result ID: 69c2683447f34ce924aa5c19.

[Reporting paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

Recorded evaluation configuration: {"Inter Head":"before"}. This record does not assert independent reproduction.

### Spatial-MLLM-4B

LLM-Judge Accuracy: 36.2. Result ID: 69e74f0847f34ce924ce2e04.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78f11e209131cea7b58a.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen7B

LLM-Judge Accuracy: 34.95. Result ID: 69c2683447f34ce924aa5c16.

[Reporting paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

Recorded evaluation configuration: {"Inter Head":"after"}. This record does not assert independent reproduction.

### Qwen7B

LLM-Judge Accuracy: 34.16. Result ID: 69c2683447f34ce924aa5c15.

[Reporting paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

Recorded evaluation configuration: {"Inter Head":"before"}. This record does not assert independent reproduction.

### Llama11B

LLM-Judge Accuracy: 29.4. Result ID: 69c2683447f34ce924aa5c18.

[Reporting paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

Recorded evaluation configuration: {"Inter Head":"after"}. This record does not assert independent reproduction.

### Llama11B

LLM-Judge Accuracy: 28.87. Result ID: 69c2683447f34ce924aa5c17.

[Reporting paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c21a640636e5bf4afc463a.md>)

Recorded evaluation configuration: {"Inter Head":"before"}. This record does not assert independent reproduction.

### MindCube-3B

LLM-Judge Accuracy: 2.8. Result ID: 69e74f0847f34ce924ce2e02.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a07d0172140ec7ed9d.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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