# Multi-agent Game Playing on Avalon Homogeneous Teams (test)

Source: [Multi-agent Game Playing on Avalon Homogeneous Teams (test)](<https://www.sota2.com/research/sota/multi-agent-game-playing-on-avalon-homogeneous-teams-test>)

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

Data updated: 2026-04-13T09:40:14.537Z

Evaluation of multi-agent performance in the Avalon Homogeneous Teams environment, measured by win rates against specific baseline agents (Rand, ReCon, DS-R1, o4-mini) and the average win rate.

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-agent Game Playing |
| Dataset (including any stated split) | Avalon Homogeneous Teams (test) |
| Metric | Win Rate (vs. Rand) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Win Rate (vs. Rand)](<https://www.sota2.com/research/sota/multi-agent-game-playing-on-avalon-homogeneous-teams-test.md?metric=Win%20Rate%20(vs.%20Rand)>): Higher is better; unit: PERCENT.

- [Win Rate (vs. ReCon)](<https://www.sota2.com/research/sota/multi-agent-game-playing-on-avalon-homogeneous-teams-test.md?metric=Win%20Rate%20(vs.%20ReCon)>): Higher is better; unit: PERCENT.

- [Win Rate (vs. DS-R1)](<https://www.sota2.com/research/sota/multi-agent-game-playing-on-avalon-homogeneous-teams-test.md?metric=Win%20Rate%20(vs.%20DS-R1)>): Higher is better; unit: PERCENT.

- [Win Rate (vs. o4-mini)](<https://www.sota2.com/research/sota/multi-agent-game-playing-on-avalon-homogeneous-teams-test.md?metric=Win%20Rate%20(vs.%20o4-mini)>): Higher is better; unit: PERCENT.

- [Average Win Rate](<https://www.sota2.com/research/sota/multi-agent-game-playing-on-avalon-homogeneous-teams-test.md?metric=Average%20Win%20Rate>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Win Rate (vs. Rand) | Result date | Report paper |
| --- | --- | --- | --- |
| GRAIL | 95 | Not recorded | 68fa798e1e209131cea7e2de |
| DeepSeek-R1 | 90 | Not recorded | 68fa798e1e209131cea7e2de |
| ReCon | 80 | Not recorded | 68fa798e1e209131cea7e2de |
| GPT-o4-mini | 70 | Not recorded | 68fa798e1e209131cea7e2de |
| Rand | 0 | Not recorded | 68fa798e1e209131cea7e2de |

### GRAIL

Win Rate (vs. Rand): 95. Result ID: 69dcb95b47f34ce924c404be.

[Reporting paper](<https://www.sota2.com/research/paper/68fa798e1e209131cea7e2de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa798e1e209131cea7e2de.md>)

Recorded evaluation configuration: {"Team Type":"Homogeneous","Underlying LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### DeepSeek-R1

Win Rate (vs. Rand): 90. Result ID: 69dcb95b47f34ce924c404bb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa798e1e209131cea7e2de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

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

### ReCon

Win Rate (vs. Rand): 80. Result ID: 69dcb95b47f34ce924c404bd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa798e1e209131cea7e2de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f01e209131cea4e267.md>)

Recorded evaluation configuration: {"Team Type":"Homogeneous","Underlying LLM":"GPT-4.1"}. This record does not assert independent reproduction.

### GPT-o4-mini

Win Rate (vs. Rand): 70. Result ID: 69dcb95b47f34ce924c404bc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa798e1e209131cea7e2de.md>)

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

### Rand

Win Rate (vs. Rand): 0. Result ID: 69dcb95b47f34ce924c404ba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa798e1e209131cea7e2de.md>)

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

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