# Driving with language reasoning on DriveLM (leaderboard)

Source: [Driving with language reasoning on DriveLM (leaderboard)](<https://www.sota2.com/research/sota/driving-with-language-reasoning-on-drivelm-leaderboard>)

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

Data updated: 2026-05-01T09:10:33.260Z

Evaluation of model performance on the DriveLM dataset for tasks requiring language-based reasoning within driving scenarios, specifically reflecting leaderboard results.

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 | Driving with language reasoning |
| Dataset (including any stated split) | DriveLM (leaderboard) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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/driving-with-language-reasoning-on-drivelm-leaderboard.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [GPT Score](<https://www.sota2.com/research/sota/driving-with-language-reasoning-on-drivelm-leaderboard.md?metric=GPT%20Score>): Higher is better; unit: SCALAR.

- [Match Rate](<https://www.sota2.com/research/sota/driving-with-language-reasoning-on-drivelm-leaderboard.md?metric=Match%20Rate>): Higher is better; unit: PERCENT.

- [Final Score](<https://www.sota2.com/research/sota/driving-with-language-reasoning-on-drivelm-leaderboard.md?metric=Final%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| HERMES++ | 83 | Not recorded | 69f433d155a3c32af89fd422 |
| HERMES++ | 83 | Not recorded | 69f433d155a3c32af89fd422 |
| Team NVIDIA | 78 | Not recorded | 69f433d155a3c32af89fd422 |
| Omni-Q | 78 | Not recorded | 69f433d155a3c32af89fd422 |
| Team NVIDIA | 78 | Not recorded | 69f433d155a3c32af89fd422 |
| Omni-Q | 78 | Not recorded | 69f433d155a3c32af89fd422 |
| FSDrive | 72 | Not recorded | 69f433d155a3c32af89fd422 |
| FSDrive | 72 | Not recorded | 69f433d155a3c32af89fd422 |
| MMFM AD | 67 | Not recorded | 69f433d155a3c32af89fd422 |
| MMFM AD | 67 | Not recorded | 69f433d155a3c32af89fd422 |
| DriveLM | 65 | Not recorded | 69f433d155a3c32af89fd422 |
| DriveLM | 65 | Not recorded | 69f433d155a3c32af89fd422 |

### HERMES++

Accuracy: 83. Result ID: 69f46df647f34ce924da9f08.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Implementation](<https://github.com/H-EmbodVis/HERMESV2>)

Recorded evaluation configuration: {"Reference":"-"}. This record does not assert independent reproduction.

### HERMES++

Accuracy: 83. Result ID: 69f46df647f34ce924da9f20.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Implementation](<https://github.com/H-EmbodVis/HERMESV2>)

Recorded evaluation configuration: {"Reference":"-"}. This record does not assert independent reproduction.

### Team NVIDIA

Accuracy: 78. Result ID: 69f46df647f34ce924da9f04.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

Recorded evaluation configuration: {"Reference":"CVPRW 24 \[92\]"}. This record does not assert independent reproduction.

### Omni-Q

Accuracy: 78. Result ID: 69f46df647f34ce924da9f06.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78031e209131cea76545.md>)

Recorded evaluation configuration: {"Reference":"CVPR 25 \[15\]"}. This record does not assert independent reproduction.

### Team NVIDIA

Accuracy: 78. Result ID: 69f46df647f34ce924da9f1c.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

Recorded evaluation configuration: {"Reference":"CVPRW 24"}. This record does not assert independent reproduction.

### Omni-Q

Accuracy: 78. Result ID: 69f46df647f34ce924da9f1e.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78031e209131cea76545.md>)

Recorded evaluation configuration: {"Reference":"CVPR 25"}. This record does not assert independent reproduction.

### FSDrive

Accuracy: 72. Result ID: 69f46df647f34ce924da9f07.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78b71e209131cea79ff0.md>)

Recorded evaluation configuration: {"Reference":"NeurIPS 25 \[54\]"}. This record does not assert independent reproduction.

### FSDrive

Accuracy: 72. Result ID: 69f46df647f34ce924da9f1f.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78b71e209131cea79ff0.md>)

Recorded evaluation configuration: {"Reference":"NeurIPS 25"}. This record does not assert independent reproduction.

### MMFM AD

Accuracy: 67. Result ID: 69f46df647f34ce924da9f05.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

Recorded evaluation configuration: {"Reference":"CVPRW 24 \[92\]"}. This record does not assert independent reproduction.

### MMFM AD

Accuracy: 67. Result ID: 69f46df647f34ce924da9f1d.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

Recorded evaluation configuration: {"Reference":"CVPRW 24"}. This record does not assert independent reproduction.

### DriveLM

Accuracy: 65. Result ID: 69f46df647f34ce924da9f03.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa769b1e209131cea6e195.md>)

Recorded evaluation configuration: {"Reference":"ECCV 24 \[16\]"}. This record does not assert independent reproduction.

### DriveLM

Accuracy: 65. Result ID: 69f46df647f34ce924da9f1b.

[Reporting paper](<https://www.sota2.com/research/paper/69f433d155a3c32af89fd422.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa769b1e209131cea6e195.md>)

Recorded evaluation configuration: {"Reference":"ECCV 24"}. This record does not assert independent reproduction.

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