# Web Navigation Question Answering on WebWalker QA (avg@3)

Source: [Web Navigation Question Answering on WebWalker QA (avg@3)](<https://www.sota2.com/research/sota/web-navigation-question-answering-on-webwalker-qa-avg-at-3>)

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

Data updated: 2026-04-23T08:33:20.298Z

Evaluation of model performance on the WebWalker QA dataset for Web Navigation Question Answering, specifically reporting the avg@3 metric.

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 | Web Navigation Question Answering |
| Dataset (including any stated split) | WebWalker QA |
| Metric | Avg@3 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Avg@3](<https://www.sota2.com/research/sota/web-navigation-question-answering-on-webwalker-qa-avg-at-3.md?metric=Avg%403>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Avg@3 | Result date | Report paper |
| --- | --- | --- | --- |
| Tongyi-DeepResearch-30B | 72.2 | Not recorded | 691bfdce6750be83fe5cea20 |
| OpenAI-o3 | 71.7 | Not recorded | 691bfdce6750be83fe5cea20 |
| AFM-32B-RL | 63 | Not recorded | 691bfdce6750be83fe5cea20 |
| WebExplorer-8B-RL | 62.7 | Not recorded | 691bfdce6750be83fe5cea20 |
| MiroThinker-v1.0-72B | 62.1 | Not recorded | 691bfdce6750be83fe5cea20 |
| Claude-4-Sonnet | 61.7 | Not recorded | 691bfdce6750be83fe5cea20 |
| DeepSeek-V3.1 | 61.2 | Not recorded | 691bfdce6750be83fe5cea20 |
| MiroThinker-v1.0-30B | 61 | Not recorded | 691bfdce6750be83fe5cea20 |
| MiroThinker-v1.0-8B | 60.6 | Not recorded | 691bfdce6750be83fe5cea20 |

### Tongyi-DeepResearch-30B

Avg@3: 72.2. Result ID: 69e9d94747f34ce924d10081.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b7f.md>)

Recorded evaluation configuration: {"Type":"Research Agents"}. This record does not assert independent reproduction.

### OpenAI-o3

Avg@3: 71.7. Result ID: 69e9d94747f34ce924d1007e.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

Recorded evaluation configuration: {"Type":"Foundation Models with Tools"}. This record does not assert independent reproduction.

### AFM-32B-RL

Avg@3: 63. Result ID: 69e9d94747f34ce924d10080.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7abc1e209131cea84e86.md>)

Recorded evaluation configuration: {"Type":"Research Agents"}. This record does not assert independent reproduction.

### WebExplorer-8B-RL

Avg@3: 62.7. Result ID: 69e9d94747f34ce924d1007f.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89abb.md>)

Recorded evaluation configuration: {"Type":"Research Agents"}. This record does not assert independent reproduction.

### MiroThinker-v1.0-72B

Avg@3: 62.1. Result ID: 69e9d94747f34ce924d10084.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Implementation](<https://github.com/MiroMindAI/MiroThinker>)

Recorded evaluation configuration: {"Parameters":"72B"}. This record does not assert independent reproduction.

### Claude-4-Sonnet

Avg@3: 61.7. Result ID: 69e9d94747f34ce924d1007d.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

Recorded evaluation configuration: {"Type":"Foundation Models with Tools"}. This record does not assert independent reproduction.

### DeepSeek-V3.1

Avg@3: 61.2. Result ID: 69e9d94747f34ce924d1007c.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77001e209131cea7066a.md>)

Recorded evaluation configuration: {"Type":"Foundation Models with Tools"}. This record does not assert independent reproduction.

### MiroThinker-v1.0-30B

Avg@3: 61. Result ID: 69e9d94747f34ce924d10083.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Implementation](<https://github.com/MiroMindAI/MiroThinker>)

Recorded evaluation configuration: {"Parameters":"30B"}. This record does not assert independent reproduction.

### MiroThinker-v1.0-8B

Avg@3: 60.6. Result ID: 69e9d94747f34ce924d10082.

[Reporting paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691bfdce6750be83fe5cea20.md>)

[Implementation](<https://github.com/MiroMindAI/MiroThinker>)

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

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