# Long-Horizon Search Intelligence on GAIA (Pass@1, Pass@3)

Source: [Long-Horizon Search Intelligence on GAIA (Pass@1, Pass@3)](<https://www.sota2.com/research/sota/long-horizon-search-intelligence-on-gaia-pass-at-1-pass-at-3>)

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

Data updated: 2026-03-27T13:58:48.517Z

Evaluation of model performance on Long-Horizon Search Intelligence tasks using the GAIA dataset, measured by 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 | Long-Horizon Search Intelligence |
| Dataset (including any stated split) | GAIA |
| Metric | Pass@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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/long-horizon-search-intelligence-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/long-horizon-search-intelligence-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 |
| --- | --- | --- | --- |
| MiroThinker-32Bv0.1 | 57.3 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| ASearcher-QWQ-32B | 52.8 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebExplorer-8B | 50 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-30B | 48.5 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-30B | 48.5 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-30B | 48.2 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| MiroThinker-8Bv0.1 | 46.6 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-30B | 45 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| Qwen3-ARPO-14B | 43.7 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-7B | 42.4 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-3B | 38.5 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-3B | 37.9 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-7B | 37.2 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-30B | 35.7 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-7B | 34 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-7B | 31.7 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-3B | 28.5 | Not recorded | 68fa7c1e1e209131cea8c7b5 |
| WebSailor-3B | 25.6 | Not recorded | 68fa7c1e1e209131cea8c7b5 |

### MiroThinker-32Bv0.1

Pass@1: 57.3. Result ID: 69c68cf347f34ce924b0c482.

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

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

Recorded evaluation configuration: {"RL":"DPO","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### ASearcher-QWQ-32B

Pass@1: 52.8. Result ID: 69c68cf347f34ce924b0c483.

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

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

Recorded evaluation configuration: {"RL":"GRPO","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### WebExplorer-8B

Pass@1: 50. Result ID: 69c68cf347f34ce924b0c484.

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

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

Recorded evaluation configuration: {"RL":"GRPO","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### WebSailor-30B

Pass@1: 48.5. Result ID: 69c68cf347f34ce924b0c48f.

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

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

Recorded evaluation configuration: {"RL":"GRPO","Paradigm":"ReSum"}. This record does not assert independent reproduction.

### WebSailor-30B

Pass@1: 48.5. Result ID: 69c68cf347f34ce924b0c491.

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

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

Recorded evaluation configuration: {"RL":"ReSum-GRPO","Paradigm":"ReSum"}. This record does not assert independent reproduction.

### WebSailor-30B

Pass@1: 48.2. Result ID: 69c68cf347f34ce924b0c48e.

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

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

Recorded evaluation configuration: {"RL":"GRPO","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### MiroThinker-8Bv0.1

Pass@1: 46.6. Result ID: 69c68cf347f34ce924b0c481.

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

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

Recorded evaluation configuration: {"RL":"DPO","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### WebSailor-30B

Pass@1: 45. Result ID: 69c68cf347f34ce924b0c48d.

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

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

Recorded evaluation configuration: {"RL":"-","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### Qwen3-ARPO-14B

Pass@1: 43.7. Result ID: 69c68cf347f34ce924b0c480.

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

Recorded evaluation configuration: {"RL":"ARPO","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### WebSailor-7B

Pass@1: 42.4. Result ID: 69c68cf347f34ce924b0c48c.

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

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

Recorded evaluation configuration: {"RL":"ReSum-GRPO","Paradigm":"ReSum"}. This record does not assert independent reproduction.

### WebSailor-3B

Pass@1: 38.5. Result ID: 69c68cf347f34ce924b0c487.

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

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

Recorded evaluation configuration: {"RL":"GRPO","Paradigm":"ReSum"}. This record does not assert independent reproduction.

### WebSailor-3B

Pass@1: 37.9. Result ID: 69c68cf347f34ce924b0c488.

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

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

Recorded evaluation configuration: {"RL":"ReSum-GRPO","Paradigm":"ReSum"}. This record does not assert independent reproduction.

### WebSailor-7B

Pass@1: 37.2. Result ID: 69c68cf347f34ce924b0c48b.

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

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

Recorded evaluation configuration: {"RL":"GRPO","Paradigm":"ReSum"}. This record does not assert independent reproduction.

### WebSailor-30B

Pass@1: 35.7. Result ID: 69c68cf347f34ce924b0c490.

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

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

Recorded evaluation configuration: {"RL":"MEM1-GRPO","Paradigm":"MEM1"}. This record does not assert independent reproduction.

### WebSailor-7B

Pass@1: 34. Result ID: 69c68cf347f34ce924b0c48a.

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

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

Recorded evaluation configuration: {"RL":"GRPO","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### WebSailor-7B

Pass@1: 31.7. Result ID: 69c68cf347f34ce924b0c489.

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

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

Recorded evaluation configuration: {"RL":"-","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### WebSailor-3B

Pass@1: 28.5. Result ID: 69c68cf347f34ce924b0c486.

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

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

Recorded evaluation configuration: {"RL":"GRPO","Paradigm":"ReAct"}. This record does not assert independent reproduction.

### WebSailor-3B

Pass@1: 25.6. Result ID: 69c68cf347f34ce924b0c485.

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

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

Recorded evaluation configuration: {"RL":"-","Paradigm":"ReAct"}. This record does not assert independent reproduction.

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