# Object Counting on Counting-Tricks

Source: [Object Counting on Counting-Tricks](<https://www.sota2.com/research/sota/object-counting-on-counting-tricks>)

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

Data updated: 2026-04-14T07:51:31.043Z

Evaluation of model performance on the Counting-Tricks dataset for the Object Counting task, measured using Average 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 | Object Counting |
| Dataset (including any stated split) | Counting-Tricks |
| Metric | Average Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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 Accuracy](<https://www.sota2.com/research/sota/object-counting-on-counting-tricks.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen2.5-VL-7B | 50.52 | Not recorded | 69ddc9ee7d0172140ec79868 |
| Ovis-8B | 47.9 | Not recorded | 69ddc9ee7d0172140ec79868 |
| LLaVA-OneVision-Qwen2-7B | 44.44 | Not recorded | 69ddc9ee7d0172140ec79868 |
| Phi-4-Multimodal | 39.83 | Not recorded | 69ddc9ee7d0172140ec79868 |
| Qwen2.5-VL-3B | 36.01 | Not recorded | 69ddc9ee7d0172140ec79868 |
| Gemma-3-4B | 27.37 | Not recorded | 69ddc9ee7d0172140ec79868 |
| Llama-3.2-11B | 24 | Not recorded | 69ddc9ee7d0172140ec79868 |
| InternVL3-8B | 21.72 | Not recorded | 69ddc9ee7d0172140ec79868 |
| LLaVA-1.6-Vicuna-7B | 14.2 | Not recorded | 69ddc9ee7d0172140ec79868 |
| LLaVA-1.5-7B | 11.82 | Not recorded | 69ddc9ee7d0172140ec79868 |

### Qwen2.5-VL-7B

Average Accuracy: 50.52. Result ID: 69ddf1e747f34ce924c52409.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

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

Recorded evaluation configuration: {"Number of Parameters":"7B"}. This record does not assert independent reproduction.

### Ovis-8B

Average Accuracy: 47.9. Result ID: 69ddf1e747f34ce924c52408.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

Recorded evaluation configuration: {"Number of Parameters":"8B"}. This record does not assert independent reproduction.

### LLaVA-OneVision-Qwen2-7B

Average Accuracy: 44.44. Result ID: 69ddf1e747f34ce924c52407.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

Recorded evaluation configuration: {"Number of Parameters":"7B"}. This record does not assert independent reproduction.

### Phi-4-Multimodal

Average Accuracy: 39.83. Result ID: 69ddf1e747f34ce924c52406.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

Recorded evaluation configuration: {"Config":"Multimodal"}. This record does not assert independent reproduction.

### Qwen2.5-VL-3B

Average Accuracy: 36.01. Result ID: 69ddf1e747f34ce924c52405.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

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

Recorded evaluation configuration: {"Number of Parameters":"3B"}. This record does not assert independent reproduction.

### Gemma-3-4B

Average Accuracy: 27.37. Result ID: 69ddf1e747f34ce924c52404.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

Recorded evaluation configuration: {"Number of Parameters":"4B"}. This record does not assert independent reproduction.

### Llama-3.2-11B

Average Accuracy: 24. Result ID: 69ddf1e747f34ce924c52403.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

Recorded evaluation configuration: {"Number of Parameters":"11B"}. This record does not assert independent reproduction.

### InternVL3-8B

Average Accuracy: 21.72. Result ID: 69ddf1e747f34ce924c52402.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

Recorded evaluation configuration: {"Number of Parameters":"8B"}. This record does not assert independent reproduction.

### LLaVA-1.6-Vicuna-7B

Average Accuracy: 14.2. Result ID: 69ddf1e747f34ce924c52401.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

Recorded evaluation configuration: {"Number of Parameters":"7B"}. This record does not assert independent reproduction.

### LLaVA-1.5-7B

Average Accuracy: 11.82. Result ID: 69ddf1e747f34ce924c52400.

[Reporting paper](<https://www.sota2.com/research/paper/69ddc9ee7d0172140ec79868.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea51970.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Number of Parameters":"7B"}. This record does not assert independent reproduction.

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