# Navigation on EmbodiedBench Vanilla

Source: [Navigation on EmbodiedBench Vanilla](<https://www.sota2.com/research/sota/navigation-on-embodiedbench-vanilla>)

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

Data updated: 2026-06-04T10:52:01.967Z

Evaluation of agent performance on the EmbodiedBench Vanilla environment specifically for navigation tasks, measured across base, common, complex, and visual sub-metrics.

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 | Navigation |
| Dataset (including any stated split) | EmbodiedBench Vanilla |
| Metric | Base Success Rate |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Base Success Rate](<https://www.sota2.com/research/sota/navigation-on-embodiedbench-vanilla.md?metric=Base%20Success%20Rate>): Higher is better; unit: PERCENT.

- [Common Success Rate](<https://www.sota2.com/research/sota/navigation-on-embodiedbench-vanilla.md?metric=Common%20Success%20Rate>): Higher is better; unit: PERCENT.

- [Complex Success Rate](<https://www.sota2.com/research/sota/navigation-on-embodiedbench-vanilla.md?metric=Complex%20Success%20Rate>): Higher is better; unit: PERCENT.

- [Visual Success Rate](<https://www.sota2.com/research/sota/navigation-on-embodiedbench-vanilla.md?metric=Visual%20Success%20Rate>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Base Success Rate | Result date | Report paper |
| --- | --- | --- | --- |
| SceneDiver (gemini-2.5-flash) | 71.7 | Not recorded | 6a2106ddc10484bac23f04f3 |
| gemini-2.5-flash (base) | 68.3 | Not recorded | 6a2106ddc10484bac23f04f3 |
| SceneDiver (Qwen2.5-VL-7B-Instruct-AWQ) | 40.7 | Not recorded | 6a2106ddc10484bac23f04f3 |
| Qwen2.5-VL-7B-Instruct-AWQ (base) | 36.3 | Not recorded | 6a2106ddc10484bac23f04f3 |

### SceneDiver (gemini-2.5-flash)

Base Success Rate: 71.7. Result ID: 6a2158c2fbc373592c1ddc38.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106ddc10484bac23f04f3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2106ddc10484bac23f04f3.md>)

Recorded evaluation configuration: {"Model":"gemini-2.5-flash","Configuration":"Ours"}. This record does not assert independent reproduction.

### gemini-2.5-flash (base)

Base Success Rate: 68.3. Result ID: 6a2158c2fbc373592c1ddc37.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106ddc10484bac23f04f3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c311e209131cea8cef5.md>)

Recorded evaluation configuration: {"Model":"gemini-2.5-flash","Configuration":"Base"}. This record does not assert independent reproduction.

### SceneDiver (Qwen2.5-VL-7B-Instruct-AWQ)

Base Success Rate: 40.7. Result ID: 6a2158c2fbc373592c1ddc36.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106ddc10484bac23f04f3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2106ddc10484bac23f04f3.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B-Instruct-AWQ","Configuration":"Ours"}. This record does not assert independent reproduction.

### Qwen2.5-VL-7B-Instruct-AWQ (base)

Base Success Rate: 36.3. Result ID: 6a2158c2fbc373592c1ddc35.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106ddc10484bac23f04f3.md>)

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

Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B-Instruct-AWQ","Configuration":"Base"}. This record does not assert independent reproduction.

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