spatial reasoning benchmark
RefCOCO-avg
RefCOCO-avg measures object grounding accuracy averaged across RefCOCO, RefCOCO+, and RefCOCOg benchmarks.
Updated Aug 7, 2026
Published models7
Registry coverage7
MetricScore
EvidenceB
RefCOCO-avg leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 0.935 points | 100.0% | 7 | C | |
| 2 | AC | 0.925 points | 83.3% | 7 | C | |
| 3 | AC | 0.924 points | 66.7% | 7 | C | |
| 4 | AC | 0.92 points | 50.0% | 7 | C | |
| 5 | AC | 0.913 points | 33.3% | 7 | C | |
| 6 | AC | 0.909 points | 16.7% | 7 | C | |
| 7 | AC | 0.892 points | 0.0% | 7 | C |
Score distribution
Top published rows on the benchmark's original scale.
RefCOCO-avg highlights
The top published results on this benchmark's own scale.
What is RefCOCO-avg?
Definition and scoring fields from the benchmark registry.
RefCOCO-avg measures object grounding accuracy averaged across RefCOCO, RefCOCO+, and RefCOCOg benchmarks.
Scores are shown in points. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- RefCOCO-avg
- Modality
- image
- Primary category
- spatial reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- refcoco-avg|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about RefCOCO-avg.
Which model scores highest on RefCOCO-avg?
Qwen3.6 Plus is currently ranked first with 0.935 points.
What does RefCOCO-avg measure?
RefCOCO-avg measures object grounding accuracy averaged across RefCOCO, RefCOCO+, and RefCOCOg benchmarks.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
7 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.