spatial reasoning benchmark
RefSpatialBench Leaderboard
RefSpatialBench evaluates spatial reference understanding and grounding.
Updated Sep 5, 2026
Models6
Model coverage6
MetricScore
EvidenceB
RefSpatialBench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score0.70 points | Percentile100.00% | Participants6 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score0.70 points | Percentile80.00% | Participants6 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score0.69 points | Percentile60.00% | Participants6 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score0.68 points | Percentile40.00% | Participants6 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score0.64 points | Percentile20.00% | Participants6 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score0.64 points | Percentile0.00% | Participants6 | EvidenceC | Evaluated |
RefSpatialBench Highlights
The leading models and scores on this benchmark.
RefSpatialBench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for RefSpatialBench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is RefSpatialBench?
What RefSpatialBench measures and how its scores work.
RefSpatialBench is a benchmark for evaluating spatial reference understanding and grounding.
It measures spatial reference understanding and grounding using Score in points.
Scores are shown in points. This benchmark is not independently verified and has an evidence level of B.
- Family
- RefSpatialBench
- Modality
- image
- Primary category
- spatial reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about RefSpatialBench.
Which model scores highest on RefSpatialBench?
Qwen3.6-27B is currently ranked first with 0.70 points.
What are the top three models on RefSpatialBench?
The current leaders are Qwen3.6-27B (0.70 points), Qwen3 VL 235B A22B Thinking (0.70 points), and Qwen3.5-122B-A10B (0.69 points).
Which RefSpatialBench model has the lowest official input price?
Qwen3.6-35B-A3B has the lowest matched official input price at $0.25 input / $1.5 output per 1M tokens.
Which models are fastest among RefSpatialBench results?
No matched runtime record is currently available.
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.