# Natural Language Understanding on EuroEval German

Source: [Natural Language Understanding on EuroEval German](<https://www.sota2.com/research/sota/natural-language-understanding-on-euroeval-german>)

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

Data updated: 2026-05-27T08:00:27.970Z

Evaluation of model performance on the EuroEval benchmark suite specifically for German language tasks, focusing on Natural Language Understanding capabilities.

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 | Natural Language Understanding |
| Dataset (including any stated split) | EuroEval German |
| Metric | EuroEval Score |
| Unit | SCALAR |
| Direction | Lower 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

- [EuroEval Score](<https://www.sota2.com/research/sota/natural-language-understanding-on-euroeval-german.md?metric=EuroEval%20Score>): Lower is better; unit: SCALAR.

## Recorded results

| Method | EuroEval Score | Result date | Report paper |
| --- | --- | --- | --- |
| aya-3B (Monolingual Paired DPO) | 0.1 | Not recorded | 6a167aa3dbd7150506f80031 |
| aya-3B (Multilingual Paired DPO) | 0.1 | Not recorded | 6a167aa3dbd7150506f80031 |
| EuroLLM-9B (Monolingual In-lang SFT) | 0.4 | Not recorded | 6a167aa3dbd7150506f80031 |
| MAPO (13B) | 0.7 | Not recorded | 6a167aa3dbd7150506f80031 |
| EuroLLM-9B (Multilingual Max-R SFT) | 0.9 | Not recorded | 6a167aa3dbd7150506f80031 |
| EuroLLM-9B (Monolingual Paired DPO) | 1 | Not recorded | 6a167aa3dbd7150506f80031 |
| EuroLLM-9B (Multilingual Paired DPO) | 1.2 | Not recorded | 6a167aa3dbd7150506f80031 |
| ICR (8B) | 1.7 | Not recorded | 6a167aa3dbd7150506f80031 |
| EuroLLM-9B (Monolingual Max-R SFT) | 2 | Not recorded | 6a167aa3dbd7150506f80031 |
| EuroLLM-9B (Multilingual All Lang SFT) | 3.8 | Not recorded | 6a167aa3dbd7150506f80031 |
| aya-3B (Monolingual In-lang SFT) | 4.2 | Not recorded | 6a167aa3dbd7150506f80031 |
| aya-3B (Monolingual Max-R SFT) | 5.7 | Not recorded | 6a167aa3dbd7150506f80031 |
| MAPO (13B) | 5.7 | Not recorded | 6a167aa3dbd7150506f80031 |
| ICR (8B) | 6.7 | Not recorded | 6a167aa3dbd7150506f80031 |
| aya-3B (Multilingual All Lang SFT) | 6.9 | Not recorded | 6a167aa3dbd7150506f80031 |
| aya-3B (Multilingual Max-R SFT) | 7.6 | Not recorded | 6a167aa3dbd7150506f80031 |
| aya-3B (Base) | 42.4 | Not recorded | 6a167aa3dbd7150506f80031 |
| EuroLLM-9B (Base) | 47.4 | Not recorded | 6a167aa3dbd7150506f80031 |

### aya-3B (Monolingual Paired DPO)

EuroEval Score: 0.1. Result ID: 6a16a41ffbc373592c0f9863.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Implementation](<https://github.com/jjzha/CroCo>)

Recorded evaluation configuration: {"Language model":"aya-3B","Regime":"Monolingual","Post-training":"DPO","Data Selection":"Paired"}. This record does not assert independent reproduction.

### aya-3B (Multilingual Paired DPO)

EuroEval Score: 0.1. Result ID: 6a16a41ffbc373592c0f9866.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Implementation](<https://github.com/jjzha/CroCo>)

Recorded evaluation configuration: {"Language model":"aya-3B","Regime":"Multilingual","Post-training":"DPO","Data Selection":"Paired"}. This record does not assert independent reproduction.

### EuroLLM-9B (Monolingual In-lang SFT)

EuroEval Score: 0.4. Result ID: 6a16a41ffbc373592c0f986a.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

Recorded evaluation configuration: {"Language model":"EuroLLM-9B","Regime":"Monolingual","Post-training":"SFT","Data Selection":"In-lang"}. This record does not assert independent reproduction.

### MAPO (13B)

EuroEval Score: 0.7. Result ID: 6a16a41ffbc373592c0f9860.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d91e209131cea58e60.md>)

Recorded evaluation configuration: {"Baseline Model":"MAPO","Parameters":"13B","Comparison":"aya-3B"}. This record does not assert independent reproduction.

### EuroLLM-9B (Multilingual Max-R SFT)

EuroEval Score: 0.9. Result ID: 6a16a41ffbc373592c0f986e.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Implementation](<https://github.com/jjzha/CroCo>)

Recorded evaluation configuration: {"Language model":"EuroLLM-9B","Regime":"Multilingual","Post-training":"SFT","Data Selection":"Max-R"}. This record does not assert independent reproduction.

### EuroLLM-9B (Monolingual Paired DPO)

EuroEval Score: 1. Result ID: 6a16a41ffbc373592c0f986c.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Implementation](<https://github.com/jjzha/CroCo>)

Recorded evaluation configuration: {"Language model":"EuroLLM-9B","Regime":"Monolingual","Post-training":"DPO","Data Selection":"Paired"}. This record does not assert independent reproduction.

### EuroLLM-9B (Multilingual Paired DPO)

EuroEval Score: 1.2. Result ID: 6a16a41ffbc373592c0f986f.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Implementation](<https://github.com/jjzha/CroCo>)

Recorded evaluation configuration: {"Language model":"EuroLLM-9B","Regime":"Multilingual","Post-training":"DPO","Data Selection":"Paired"}. This record does not assert independent reproduction.

### ICR (8B)

EuroEval Score: 1.7. Result ID: 6a16a41ffbc373592c0f985f.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79321e209131cea7c8f2.md>)

Recorded evaluation configuration: {"Baseline Model":"ICR","Parameters":"8B","Comparison":"aya-3B"}. This record does not assert independent reproduction.

### EuroLLM-9B (Monolingual Max-R SFT)

EuroEval Score: 2. Result ID: 6a16a41ffbc373592c0f986b.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Implementation](<https://github.com/jjzha/CroCo>)

Recorded evaluation configuration: {"Language model":"EuroLLM-9B","Regime":"Monolingual","Post-training":"SFT","Data Selection":"Max-R"}. This record does not assert independent reproduction.

### EuroLLM-9B (Multilingual All Lang SFT)

EuroEval Score: 3.8. Result ID: 6a16a41ffbc373592c0f986d.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

Recorded evaluation configuration: {"Language model":"EuroLLM-9B","Regime":"Multilingual","Post-training":"SFT","Data Selection":"All Lang"}. This record does not assert independent reproduction.

### aya-3B (Monolingual In-lang SFT)

EuroEval Score: 4.2. Result ID: 6a16a41ffbc373592c0f9861.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

Recorded evaluation configuration: {"Language model":"aya-3B","Regime":"Monolingual","Post-training":"SFT","Data Selection":"In-lang"}. This record does not assert independent reproduction.

### aya-3B (Monolingual Max-R SFT)

EuroEval Score: 5.7. Result ID: 6a16a41ffbc373592c0f9862.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Implementation](<https://github.com/jjzha/CroCo>)

Recorded evaluation configuration: {"Language model":"aya-3B","Regime":"Monolingual","Post-training":"SFT","Data Selection":"Max-R"}. This record does not assert independent reproduction.

### MAPO (13B)

EuroEval Score: 5.7. Result ID: 6a16a41ffbc373592c0f9869.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d91e209131cea58e60.md>)

Recorded evaluation configuration: {"Baseline Model":"MAPO","Parameters":"13B","Comparison":"EuroLLM-9B"}. This record does not assert independent reproduction.

### ICR (8B)

EuroEval Score: 6.7. Result ID: 6a16a41ffbc373592c0f9868.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79321e209131cea7c8f2.md>)

Recorded evaluation configuration: {"Baseline Model":"ICR","Parameters":"8B","Comparison":"EuroLLM-9B"}. This record does not assert independent reproduction.

### aya-3B (Multilingual All Lang SFT)

EuroEval Score: 6.9. Result ID: 6a16a41ffbc373592c0f9864.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

Recorded evaluation configuration: {"Language model":"aya-3B","Regime":"Multilingual","Post-training":"SFT","Data Selection":"All Lang"}. This record does not assert independent reproduction.

### aya-3B (Multilingual Max-R SFT)

EuroEval Score: 7.6. Result ID: 6a16a41ffbc373592c0f9865.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Implementation](<https://github.com/jjzha/CroCo>)

Recorded evaluation configuration: {"Language model":"aya-3B","Regime":"Multilingual","Post-training":"SFT","Data Selection":"Max-R"}. This record does not assert independent reproduction.

### aya-3B (Base)

EuroEval Score: 42.4. Result ID: 6a16a41ffbc373592c0f985e.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b399ef0636e5bf4afbc93b.md>)

Recorded evaluation configuration: {"Language model":"aya-3B","Regime":"Baseline"}. This record does not assert independent reproduction.

### EuroLLM-9B (Base)

EuroEval Score: 47.4. Result ID: 6a16a41ffbc373592c0f9867.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aa3dbd7150506f80031.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698575d8c97a0b3e3b5d5e8c.md>)

Recorded evaluation configuration: {"Language model":"EuroLLM-9B","Regime":"Baseline"}. This record does not assert independent reproduction.

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