# Multi-Agent Reinforcement Learning on SMAC 3s5z v1 (test)

Source: [Multi-Agent Reinforcement Learning on SMAC 3s5z v1 (test)](<https://www.sota2.com/research/sota/multi-agent-reinforcement-learning-on-smac-3s5z-v1-test>)

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

Data updated: 2026-05-19T10:09:18.659Z

Evaluation of Multi-Agent Reinforcement Learning performance on the test split of the SMAC 3s5z v1 environment, reporting AUC and Win metrics.

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 Reinforcement Learning |
| Dataset (including any stated split) | SMAC 3s5z v1 (test) |
| Metric | Win Rate |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 13 of 13 |
| 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](<https://www.sota2.com/research/sota/multi-agent-reinforcement-learning-on-smac-3s5z-v1-test.md?metric=Win%20Rate>): Higher is better; unit: PERCENT.

- [AUC](<https://www.sota2.com/research/sota/multi-agent-reinforcement-learning-on-smac-3s5z-v1-test.md?metric=AUC>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Win Rate | Result date | Report paper |
| --- | --- | --- | --- |
| MAGI | 92.7 | Not recorded | 6a0bee9ddbc836b87d82c492 |
| AIB-only | 92.6 | Not recorded | 6a0bee9ddbc836b87d82c492 |
| HIBCG | 92.6 | Not recorded | 6a0bee9ddbc836b87d82c492 |
| BVME | 91.56 | Not recorded | 693f96196cfd488761a81c4b |
| CommFormer | 91.2 | Not recorded | 6a0bee9ddbc836b87d82c492 |
| HIB-flat | 90.8 | Not recorded | 6a0bee9ddbc836b87d82c492 |
| QMIX | 89.7 | Not recorded | 6a0bee9ddbc836b87d82c492 |
| GACG | 89.7 | Not recorded | 6a0bee9ddbc836b87d82c492 |
| BVME | 89.6 | Not recorded | 6a0bee9ddbc836b87d82c492 |
| BVME | 87.29 | Not recorded | 693f96196cfd488761a81c4b |
| GACG | 86.77 | Not recorded | 693f96196cfd488761a81c4b |
| BVME | 86.77 | Not recorded | 693f96196cfd488761a81c4b |
| ExpoComm | 85.5 | Not recorded | 6a0bee9ddbc836b87d82c492 |

### MAGI

Win Rate: 92.7. Result ID: 6a0c36bafbc373592c02db41.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

### AIB-only

Win Rate: 92.6. Result ID: 6a0c36bafbc373592c02db46.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

### HIBCG

Win Rate: 92.6. Result ID: 6a0c36bafbc373592c02db48.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

### BVME

Win Rate: 91.56. Result ID: 69996c425698665ec4d9f1b0.

[Reporting paper](<https://www.sota2.com/research/paper/693f96196cfd488761a81c4b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/693f96196cfd488761a81c4b.md>)

Recorded evaluation configuration: {"Bandwidth Ratio (r)":0.05,"Message Reduction vs 0.30":"83.3%"}. This record does not assert independent reproduction.

### CommFormer

Win Rate: 91.2. Result ID: 6a0c36bafbc373592c02db42.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73351e209131cea5b287.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

### HIB-flat

Win Rate: 90.8. Result ID: 6a0c36bafbc373592c02db47.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

### QMIX

Win Rate: 89.7. Result ID: 6a0c36bafbc373592c02db40.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64671e209131cea046e9.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

### GACG

Win Rate: 89.7. Result ID: 6a0c36bafbc373592c02db44.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73271e209131cea5ab8a.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

### BVME

Win Rate: 89.6. Result ID: 6a0c36bafbc373592c02db45.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

### BVME

Win Rate: 87.29. Result ID: 69996c425698665ec4d9f1b1.

[Reporting paper](<https://www.sota2.com/research/paper/693f96196cfd488761a81c4b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/693f96196cfd488761a81c4b.md>)

Recorded evaluation configuration: {"Bandwidth Ratio (r)":0.075,"Message Reduction vs 0.30":"75.0%"}. This record does not assert independent reproduction.

### GACG

Win Rate: 86.77. Result ID: 69996c425698665ec4d9f1af.

[Reporting paper](<https://www.sota2.com/research/paper/693f96196cfd488761a81c4b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73271e209131cea5ab8a.md>)

Recorded evaluation configuration: {"Bandwidth Ratio (r)":0.3,"Message Reduction vs 0.30":null}. This record does not assert independent reproduction.

### BVME

Win Rate: 86.77. Result ID: 69996c425698665ec4d9f1b2.

[Reporting paper](<https://www.sota2.com/research/paper/693f96196cfd488761a81c4b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/693f96196cfd488761a81c4b.md>)

Recorded evaluation configuration: {"Bandwidth Ratio (r)":0.1,"Message Reduction vs 0.30":"66.7%"}. This record does not assert independent reproduction.

### ExpoComm

Win Rate: 85.5. Result ID: 6a0c36bafbc373592c02db43.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee9ddbc836b87d82c492.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa772e1e209131cea719f9.md>)

Recorded evaluation configuration: {"Unit Composition":"2-type","Training Budget":"1.2M"}. This record does not assert independent reproduction.

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