# Multi-step reasoning and tool use on GAIA (val)

Source: [Multi-step reasoning and tool use on GAIA (val)](<https://www.sota2.com/research/sota/multi-step-reasoning-and-tool-use-on-gaia-val>)

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

Data updated: 2026-02-26T05:57:22.956Z

Evaluation of model performance on multi-step reasoning and tool use capabilities using the validation split of the GAIA dataset, measured by Pass@1, Pass@3, and Average Pass@1 scores.

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 | Multi-step reasoning and tool use |
| Dataset (including any stated split) | GAIA (val) |
| Metric | Pass@1 |
| 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

- [Pass@1](<https://www.sota2.com/research/sota/multi-step-reasoning-and-tool-use-on-gaia-val.md?metric=Pass%401>): Higher is better; unit: PERCENT.

- [Pass@3](<https://www.sota2.com/research/sota/multi-step-reasoning-and-tool-use-on-gaia-val.md?metric=Pass%403>): Higher is better; unit: PERCENT.

- [Average Pass@1](<https://www.sota2.com/research/sota/multi-step-reasoning-and-tool-use-on-gaia-val.md?metric=Average%20Pass%401>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Pass@1 | Result date | Report paper |
| --- | --- | --- | --- |
| AORCHESTRA | 80 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| AORCHESTRA | 67.87 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| OpenHands | 66.06 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| OpenHands | 63.64 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| AORCHESTRA | 60.61 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| Mini-SWE | 58.18 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| OpenHands | 54.55 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| Mini-SWE | 50.3 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| ReAct | 49.09 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| ReAct | 47.88 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| ReAct | 46.7 | Not recorded | 6982d2fba8b01e76cc8e9a12 |
| Mini-SWE | 40.61 | Not recorded | 6982d2fba8b01e76cc8e9a12 |

### AORCHESTRA

Pass@1: 80. Result ID: 699605ce5698665ec43cf104.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

Recorded evaluation configuration: {"Model Setup":"Gemini-3-Flash"}. This record does not assert independent reproduction.

### AORCHESTRA

Pass@1: 67.87. Result ID: 699605ce5698665ec43cf105.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

Recorded evaluation configuration: {"Model Setup":"DeepSeek-V3.2"}. This record does not assert independent reproduction.

### OpenHands

Pass@1: 66.06. Result ID: 699605ce5698665ec43cf0fe.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78221e209131cea76ed6.md>)

Recorded evaluation configuration: {"Model Setup":"Gemini-3-Flash"}. This record does not assert independent reproduction.

### OpenHands

Pass@1: 63.64. Result ID: 699605ce5698665ec43cf0ff.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78221e209131cea76ed6.md>)

Recorded evaluation configuration: {"Model Setup":"DeepSeek-V3.2"}. This record does not assert independent reproduction.

### AORCHESTRA

Pass@1: 60.61. Result ID: 699605ce5698665ec43cf106.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

Recorded evaluation configuration: {"Model Setup":"Claude-4.5-haiku"}. This record does not assert independent reproduction.

### Mini-SWE

Pass@1: 58.18. Result ID: 699605ce5698665ec43cf101.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b61e209131cea697de.md>)

Recorded evaluation configuration: {"Model Setup":"Gemini-3-Flash"}. This record does not assert independent reproduction.

### OpenHands

Pass@1: 54.55. Result ID: 699605ce5698665ec43cf100.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78221e209131cea76ed6.md>)

Recorded evaluation configuration: {"Model Setup":"Claude-4.5-haiku"}. This record does not assert independent reproduction.

### Mini-SWE

Pass@1: 50.3. Result ID: 699605ce5698665ec43cf102.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b61e209131cea697de.md>)

Recorded evaluation configuration: {"Model Setup":"DeepSeek-V3.2"}. This record does not assert independent reproduction.

### ReAct

Pass@1: 49.09. Result ID: 699605ce5698665ec43cf0fb.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Model Setup":"Gemini-3-Flash"}. This record does not assert independent reproduction.

### ReAct

Pass@1: 47.88. Result ID: 699605ce5698665ec43cf0fd.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Model Setup":"Claude-4.5-haiku"}. This record does not assert independent reproduction.

### ReAct

Pass@1: 46.7. Result ID: 699605ce5698665ec43cf0fc.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Model Setup":"DeepSeek-V3.2"}. This record does not assert independent reproduction.

### Mini-SWE

Pass@1: 40.61. Result ID: 699605ce5698665ec43cf103.

[Reporting paper](<https://www.sota2.com/research/paper/6982d2fba8b01e76cc8e9a12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75b61e209131cea697de.md>)

Recorded evaluation configuration: {"Model Setup":"Claude-4.5-haiku"}. This record does not assert independent reproduction.

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