# Embodied Task Success on EmbodiedBench EB-Habitat 1.0 (test)

Source: [Embodied Task Success on EmbodiedBench EB-Habitat 1.0 (test)](<https://www.sota2.com/research/sota/embodied-task-success-on-embodiedbench-eb-habitat-1-0-test>)

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

Data updated: 2026-05-12T15:32:52.530Z

Evaluation of embodied agent performance on task success within the EmbodiedBench EB-Habitat 1.0 environment (test split), reporting scores across Base, Common, Complex, Long, Spatial, and Visual sub-tasks, summarized by the Average score.

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 | Embodied Task Success |
| Dataset (including any stated split) | EmbodiedBench EB-Habitat 1.0 (test) |
| Metric | Average Task Success |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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 Task Success](<https://www.sota2.com/research/sota/embodied-task-success-on-embodiedbench-eb-habitat-1-0-test.md?metric=Average%20Task%20Success>): Higher is better; unit: PERCENT.

- [Base Task Success](<https://www.sota2.com/research/sota/embodied-task-success-on-embodiedbench-eb-habitat-1-0-test.md?metric=Base%20Task%20Success>): Higher is better; unit: PERCENT.

- [Common Task Success](<https://www.sota2.com/research/sota/embodied-task-success-on-embodiedbench-eb-habitat-1-0-test.md?metric=Common%20Task%20Success>): Higher is better; unit: PERCENT.

- [Complex Task Success](<https://www.sota2.com/research/sota/embodied-task-success-on-embodiedbench-eb-habitat-1-0-test.md?metric=Complex%20Task%20Success>): Higher is better; unit: PERCENT.

- [Visual Task Success](<https://www.sota2.com/research/sota/embodied-task-success-on-embodiedbench-eb-habitat-1-0-test.md?metric=Visual%20Task%20Success>): Higher is better; unit: PERCENT.

- [Spatial Task Success](<https://www.sota2.com/research/sota/embodied-task-success-on-embodiedbench-eb-habitat-1-0-test.md?metric=Spatial%20Task%20Success>): Higher is better; unit: PERCENT.

- [Long Task Success](<https://www.sota2.com/research/sota/embodied-task-success-on-embodiedbench-eb-habitat-1-0-test.md?metric=Long%20Task%20Success>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Task Success | Result date | Report paper |
| --- | --- | --- | --- |
| EmbodiSkill | 52.33 | Not recorded | 6a02b477dbc836b87d824bc0 |
| EmbodiSkill | 50.33 | Not recorded | 6a02b477dbc836b87d824bc0 |
| Gemini-3-flash | 46 | Not recorded | 6a02b477dbc836b87d824bc0 |
| EmbodiSkill | 45.33 | Not recorded | 6a02b477dbc836b87d824bc0 |
| G-Memory | 45 | Not recorded | 6a02b477dbc836b87d824bc0 |
| EmbodiSkill | 41 | Not recorded | 6a02b477dbc836b87d824bc0 |
| GPT-5.2 | 40 | Not recorded | 6a02b477dbc836b87d824bc0 |
| Mem0 | 38.33 | Not recorded | 6a02b477dbc836b87d824bc0 |
| LangMem | 38.33 | Not recorded | 6a02b477dbc836b87d824bc0 |
| No memory | 34.67 | Not recorded | 6a02b477dbc836b87d824bc0 |
| G-Memory | 25.66 | Not recorded | 6a02b477dbc836b87d824bc0 |
| No memory | 24 | Not recorded | 6a02b477dbc836b87d824bc0 |
| Mem0 | 20.33 | Not recorded | 6a02b477dbc836b87d824bc0 |
| LangMem | 19.67 | Not recorded | 6a02b477dbc836b87d824bc0 |

### EmbodiSkill

Average Task Success: 52.33. Result ID: 6a03469f47f34ce924e86a3d.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-32B-Instruct","Skill Evolution Model":"Gemini-3-flash"}. This record does not assert independent reproduction.

### EmbodiSkill

Average Task Success: 50.33. Result ID: 6a03469f47f34ce924e86a3c.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-32B-Instruct","Skill Evolution Model":"GPT-5.2"}. This record does not assert independent reproduction.

### Gemini-3-flash

Average Task Success: 46. Result ID: 6a03469f47f34ce924e86a31.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Direct (Closed-source)"}. This record does not assert independent reproduction.

### EmbodiSkill

Average Task Success: 45.33. Result ID: 6a03469f47f34ce924e86a36.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-8B-Instruct","Skill Evolution Model":"GPT-5.2"}. This record does not assert independent reproduction.

### G-Memory

Average Task Success: 45. Result ID: 6a03469f47f34ce924e86a3a.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79731e209131cea7db10.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-32B-Instruct","Memory Strategy":"G-Memory"}. This record does not assert independent reproduction.

### EmbodiSkill

Average Task Success: 41. Result ID: 6a03469f47f34ce924e86a37.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-8B-Instruct","Skill Evolution Model":"Gemini-3-flash"}. This record does not assert independent reproduction.

### GPT-5.2

Average Task Success: 40. Result ID: 6a03469f47f34ce924e86a30.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Direct (Closed-source)"}. This record does not assert independent reproduction.

### Mem0

Average Task Success: 38.33. Result ID: 6a03469f47f34ce924e86a39.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa784b1e209131cea77ad6.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-32B-Instruct","Memory Strategy":"Mem0"}. This record does not assert independent reproduction.

### LangMem

Average Task Success: 38.33. Result ID: 6a03469f47f34ce924e86a3b.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-32B-Instruct","Memory Strategy":"LangMem"}. This record does not assert independent reproduction.

### No memory

Average Task Success: 34.67. Result ID: 6a03469f47f34ce924e86a38.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-32B-Instruct","Memory Strategy":"None"}. This record does not assert independent reproduction.

### G-Memory

Average Task Success: 25.66. Result ID: 6a03469f47f34ce924e86a34.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79731e209131cea7db10.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-8B-Instruct","Memory Strategy":"G-Memory"}. This record does not assert independent reproduction.

### No memory

Average Task Success: 24. Result ID: 6a03469f47f34ce924e86a32.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-8B-Instruct","Memory Strategy":"None"}. This record does not assert independent reproduction.

### Mem0

Average Task Success: 20.33. Result ID: 6a03469f47f34ce924e86a33.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa784b1e209131cea77ad6.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-8B-Instruct","Memory Strategy":"Mem0"}. This record does not assert independent reproduction.

### LangMem

Average Task Success: 19.67. Result ID: 6a03469f47f34ce924e86a35.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b477dbc836b87d824bc0.md>)

Recorded evaluation configuration: {"Executor":"Qwen3-VL-8B-Instruct","Memory Strategy":"LangMem"}. This record does not assert independent reproduction.

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