# Multi-modal Understanding on Cambrian

Source: [Multi-modal Understanding on Cambrian](<https://www.sota2.com/research/sota/multi-modal-understanding-on-cambrian>)

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

Data updated: 2026-05-19T07:41:21.557Z

Evaluation of model performance on the Cambrian dataset focusing on Multi-modal Understanding capabilities, measured by 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 | Multi-modal Understanding |
| Dataset (including any stated split) | Cambrian |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/multi-modal-understanding-on-cambrian.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| InternVL3.5-8B | 76.22 | Not recorded | 6a0bee84dbc836b87d82c339 |
| CDPruner | 75.74 | Not recorded | 6a0bee84dbc836b87d82c339 |
| F3A | 75.61 | Not recorded | 6a0bee84dbc836b87d82c339 |
| VisionZip | 74.85 | Not recorded | 6a0bee84dbc836b87d82c339 |
| DivPrune | 74.75 | Not recorded | 6a0bee84dbc836b87d82c339 |
| F3A | 74.7 | Not recorded | 6a0bee84dbc836b87d82c339 |
| CDPruner | 74.28 | Not recorded | 6a0bee84dbc836b87d82c339 |
| VisionZip | 74.15 | Not recorded | 6a0bee84dbc836b87d82c339 |
| FastV | 74.14 | Not recorded | 6a0bee84dbc836b87d82c339 |
| DivPrune | 74.08 | Not recorded | 6a0bee84dbc836b87d82c339 |
| FastV | 73.53 | Not recorded | 6a0bee84dbc836b87d82c339 |
| F3A | 73.27 | Not recorded | 6a0bee84dbc836b87d82c339 |
| FastV | 72.96 | Not recorded | 6a0bee84dbc836b87d82c339 |
| CDPruner | 72.77 | Not recorded | 6a0bee84dbc836b87d82c339 |
| VisionZip | 72.03 | Not recorded | 6a0bee84dbc836b87d82c339 |
| DivPrune | 71.49 | Not recorded | 6a0bee84dbc836b87d82c339 |

### InternVL3.5-8B

Accuracy: 76.22. Result ID: 6a0c13f3fbc373592c01d53d.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71de1e209131cea5370d.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Ratio":"100%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### CDPruner

Accuracy: 75.74. Result ID: 6a0c13f3fbc373592c01d53e.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6cb.md>)

Recorded evaluation configuration: {"Ratio":"60%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### F3A

Accuracy: 75.61. Result ID: 6a0c13f3fbc373592c01d542.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

Recorded evaluation configuration: {"Ratio":"60%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### VisionZip

Accuracy: 74.85. Result ID: 6a0c13f3fbc373592c01d541.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa760f1e209131cea6b33b.md>)

[Implementation](<https://github.com/dvlab-research/VisionZip>)

Recorded evaluation configuration: {"Ratio":"60%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### DivPrune

Accuracy: 74.75. Result ID: 6a0c13f3fbc373592c01d540.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77c41e209131cea7532a.md>)

Recorded evaluation configuration: {"Ratio":"60%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### F3A

Accuracy: 74.7. Result ID: 6a0c13f3fbc373592c01d547.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

Recorded evaluation configuration: {"Ratio":"40%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### CDPruner

Accuracy: 74.28. Result ID: 6a0c13f3fbc373592c01d543.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6cb.md>)

Recorded evaluation configuration: {"Ratio":"40%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### VisionZip

Accuracy: 74.15. Result ID: 6a0c13f3fbc373592c01d546.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa760f1e209131cea6b33b.md>)

[Implementation](<https://github.com/dvlab-research/VisionZip>)

Recorded evaluation configuration: {"Ratio":"40%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### FastV

Accuracy: 74.14. Result ID: 6a0c13f3fbc373592c01d53f.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632b7.md>)

Recorded evaluation configuration: {"Ratio":"60%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### DivPrune

Accuracy: 74.08. Result ID: 6a0c13f3fbc373592c01d545.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77c41e209131cea7532a.md>)

Recorded evaluation configuration: {"Ratio":"40%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### FastV

Accuracy: 73.53. Result ID: 6a0c13f3fbc373592c01d544.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632b7.md>)

Recorded evaluation configuration: {"Ratio":"40%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### F3A

Accuracy: 73.27. Result ID: 6a0c13f3fbc373592c01d54c.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

Recorded evaluation configuration: {"Ratio":"20%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### FastV

Accuracy: 72.96. Result ID: 6a0c13f3fbc373592c01d549.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632b7.md>)

Recorded evaluation configuration: {"Ratio":"20%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### CDPruner

Accuracy: 72.77. Result ID: 6a0c13f3fbc373592c01d548.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6cb.md>)

Recorded evaluation configuration: {"Ratio":"20%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### VisionZip

Accuracy: 72.03. Result ID: 6a0c13f3fbc373592c01d54b.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa760f1e209131cea6b33b.md>)

[Implementation](<https://github.com/dvlab-research/VisionZip>)

Recorded evaluation configuration: {"Ratio":"20%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

### DivPrune

Accuracy: 71.49. Result ID: 6a0c13f3fbc373592c01d54a.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee84dbc836b87d82c339.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77c41e209131cea7532a.md>)

Recorded evaluation configuration: {"Ratio":"20%","Backbone":"InternVL3.5-8B"}. This record does not assert independent reproduction.

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