reasoning benchmark
CountBench Leaderboard
CountBench evaluates object counting capabilities in visual understanding.
Updated Oct 23, 2026
Models9
Model coverage9
MetricScore
EvidenceB
CountBench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelBY | Score96.30 points | Percentile100.00% | Participants9 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score0.98 points | Percentile87.50% | Participants9 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score0.98 points | Percentile75.00% | Participants9 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score0.98 points | Percentile62.50% | Participants9 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score0.98 points | Percentile50.00% | Participants9 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score0.97 points | Percentile37.50% | Participants9 | EvidenceC | Evaluated |
| Rank07 | ModelBY | Score0.96 points | Percentile25.00% | Participants9 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score0.94 points | Percentile12.50% | Participants9 | EvidenceC | Evaluated |
| Rank09 | ModelCO | Score0.73 points | Percentile0.00% | Participants9 | EvidenceC | Evaluated |
CountBench Highlights
The leading models and scores on this benchmark.
CountBench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for CountBench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is CountBench?
What CountBench measures and how its scores work.
CountBench is a benchmark for evaluating object counting capabilities in visual understanding.
It measures object counting capabilities in visual understanding using Score, reported in points.
Scores are shown in points. This benchmark is not independently verified and has an evidence level of B.
- Family
- CountBench
- Modality
- image
- Primary category
- 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 CountBench.
Which model scores highest on CountBench?
Seed 1.8 is currently ranked first with 96.30 points.
What are the top three models on CountBench?
The current leaders are Seed 1.8 (96.30 points), Qwen3.5-27B (0.98 points), and Qwen3.5-35B-A3B (0.98 points).
Which CountBench model has the lowest official input price?
Qwen3.5-35B-A3B has the lowest matched official input price at $0.25 input / $2.0 output per 1M tokens.
Which models are fastest among CountBench 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.