# Model Evaluation on Aggregate All Tasks (AVG)

Source: [Model Evaluation on Aggregate All Tasks (AVG)](<https://www.sota2.com/research/sota/model-evaluation-on-aggregate-all-tasks-avg>)

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

Data updated: 2026-02-26T05:26:28.449Z

Overall evaluation score, reported as an average (AVG), across all tasks aggregated within the dataset.

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 | Model Evaluation |
| Dataset (including any stated split) | Aggregate All Tasks |
| Metric | AVG |
| 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

- [AVG](<https://www.sota2.com/research/sota/model-evaluation-on-aggregate-all-tasks-avg.md?metric=AVG>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AVG | Result date | Report paper |
| --- | --- | --- | --- |
| Aligned | 76 | Not recorded | 68fa7ba31e209131cea8984e |
| PRINMIX | 74 | Not recorded | 68fa7ba31e209131cea8984e |
| Delta-CoMe | 71.9 | Not recorded | 68fa7ba31e209131cea8984e |
| Low-Rank | 48.2 | Not recorded | 68fa7ba31e209131cea8984e |
| BitDelta | 40.5 | Not recorded | 68fa7ba31e209131cea8984e |

### Aligned

AVG: 76. Result ID: 699a67f45698665ec4f04621.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba31e209131cea8984e.md>)

Recorded evaluation configuration: {"α":"1"}. This record does not assert independent reproduction.

### PRINMIX

AVG: 74. Result ID: 699a67f45698665ec4f04625.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba31e209131cea8984e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba31e209131cea8984e.md>)

Recorded evaluation configuration: {"α":"1/16"}. This record does not assert independent reproduction.

### Delta-CoMe

AVG: 71.9. Result ID: 699a67f45698665ec4f04624.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba31e209131cea8984e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75e71e209131cea6a8ac.md>)

Recorded evaluation configuration: {"α":"1/16"}. This record does not assert independent reproduction.

### Low-Rank

AVG: 48.2. Result ID: 699a67f45698665ec4f04622.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba31e209131cea8984e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f931e209131cea4662a.md>)

Recorded evaluation configuration: {"α":"1/16"}. This record does not assert independent reproduction.

### BitDelta

AVG: 40.5. Result ID: 699a67f45698665ec4f04623.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba31e209131cea8984e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75371e209131cea66803.md>)

[Implementation](<https://github.com/FasterDecoding/BitDelta>)

Recorded evaluation configuration: {"α":"1/16"}. This record does not assert independent reproduction.

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