# Deep Research on GAIA (Pass@1, Pass@3)

Source: [Deep Research on GAIA (Pass@1, Pass@3)](<https://www.sota2.com/research/sota/deep-research-on-gaia-pass-at-1-pass-at-3>)

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

Data updated: 2026-02-26T05:57:31.220Z

Evaluation of Deep Research capabilities on the GAIA dataset, measured using Pass@1 and Pass@3.

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 | Deep Research |
| Dataset (including any stated split) | GAIA |
| Metric | Pass@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Pass@1](<https://www.sota2.com/research/sota/deep-research-on-gaia-pass-at-1-pass-at-3.md?metric=Pass%401>): Higher is better; unit: PERCENT.

- [Pass@3](<https://www.sota2.com/research/sota/deep-research-on-gaia-pass-at-1-pass-at-3.md?metric=Pass%403>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Pass@1 | Result date | Report paper |
| --- | --- | --- | --- |
| OpenAI-o3 | 70.5 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| Claude-4-Sonnet | 68.3 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| DeepSeek-V3.1 | 63.1 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| Kimi-K2 | 57.7 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| WebSailor | 53.2 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| WebDancer-QwQ | 51.5 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| WebThinker-RL | 48.5 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| Web-30B-E-GRPO | 48.5 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| Web-30B-GRPO | 47.6 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| Web-30B-SFT | 45 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| WebSailor | 37.9 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| Web-7B-E-GRPO | 36.9 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| Web-7B-GRPO | 33 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| Web-7B-SFT | 31.7 | Not recorded | 69019f853ef7ccd74e5e3b78 |
| R1-Searcher | 20.4 | Not recorded | 69019f853ef7ccd74e5e3b78 |

### OpenAI-o3

Pass@1: 70.5. Result ID: 699ea3615698665ec46e40d6.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c7b5.md>)

Recorded evaluation configuration: {"Model Category":"Advanced Models"}. This record does not assert independent reproduction.

### Claude-4-Sonnet

Pass@1: 68.3. Result ID: 699ea3615698665ec46e40d7.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c7b5.md>)

Recorded evaluation configuration: {"Model Category":"Advanced Models"}. This record does not assert independent reproduction.

### DeepSeek-V3.1

Pass@1: 63.1. Result ID: 699ea3615698665ec46e40d9.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c7b5.md>)

Recorded evaluation configuration: {"Model Category":"Advanced Models"}. This record does not assert independent reproduction.

### Kimi-K2

Pass@1: 57.7. Result ID: 699ea3615698665ec46e40d8.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c1e1e209131cea8c7b5.md>)

Recorded evaluation configuration: {"Model Category":"Advanced Models"}. This record does not assert independent reproduction.

### WebSailor

Pass@1: 53.2. Result ID: 699ea3615698665ec46e40de.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79c91e209131cea7f5e0.md>)

Recorded evaluation configuration: {"Model Size":"32B"}. This record does not assert independent reproduction.

### WebDancer-QwQ

Pass@1: 51.5. Result ID: 699ea3615698665ec46e40dc.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a8d1e209131cea83dce.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### WebThinker-RL

Pass@1: 48.5. Result ID: 699ea3615698665ec46e40db.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c191e209131cea8c4d7.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Web-30B-E-GRPO

Pass@1: 48.5. Result ID: 699ea3615698665ec46e40e4.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

Recorded evaluation configuration: {"Model Size":"30B","Training Strategy":"E-GRPO"}. This record does not assert independent reproduction.

### Web-30B-GRPO

Pass@1: 47.6. Result ID: 699ea3615698665ec46e40e3.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

Recorded evaluation configuration: {"Model Size":"30B","Training Strategy":"GRPO"}. This record does not assert independent reproduction.

### Web-30B-SFT

Pass@1: 45. Result ID: 699ea3615698665ec46e40e2.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

Recorded evaluation configuration: {"Model Size":"30B","Training Strategy":"SFT"}. This record does not assert independent reproduction.

### WebSailor

Pass@1: 37.9. Result ID: 699ea3615698665ec46e40dd.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79c91e209131cea7f5e0.md>)

Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction.

### Web-7B-E-GRPO

Pass@1: 36.9. Result ID: 699ea3615698665ec46e40e1.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

Recorded evaluation configuration: {"Model Size":"7B","Training Strategy":"E-GRPO"}. This record does not assert independent reproduction.

### Web-7B-GRPO

Pass@1: 33. Result ID: 699ea3615698665ec46e40e0.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

Recorded evaluation configuration: {"Model Size":"7B","Training Strategy":"GRPO"}. This record does not assert independent reproduction.

### Web-7B-SFT

Pass@1: 31.7. Result ID: 699ea3615698665ec46e40df.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

Recorded evaluation configuration: {"Model Size":"7B","Training Strategy":"SFT"}. This record does not assert independent reproduction.

### R1-Searcher

Pass@1: 20.4. Result ID: 699ea3615698665ec46e40da.

[Reporting paper](<https://www.sota2.com/research/paper/69019f853ef7ccd74e5e3b78.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77771e209131cea7356d.md>)

Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction.

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