# Multilingual Reward Modeling on m-RewardBench Avg. 23 langs

Source: [Multilingual Reward Modeling on m-RewardBench Avg. 23 langs](<https://www.sota2.com/research/sota/multilingual-reward-modeling-on-m-rewardbench-avg-23-langs>)

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

Data updated: 2026-03-12T07:53:43.890Z

Evaluation of model performance on the multilingual average subset (23 languages) of the m-RewardBench dataset for the Multilingual Reward Modeling task, measured using Accuracy.

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 | Multilingual Reward Modeling |
| Dataset (including any stated split) | m-RewardBench Avg. 23 langs |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/multilingual-reward-modeling-on-m-rewardbench-avg-23-langs.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DIBJudge-Qwen3-8B | 91.37 | Not recorded | 69b24878d030084c22e2f85c |
| DIBJudge-Qwen3-4B | 89.84 | Not recorded | 69b24878d030084c22e2f85c |
| Nemotron-Multi-49B | 88.83 | Not recorded | 69b24878d030084c22e2f85c |
| mR3-Qwen3-8B | 88.58 | Not recorded | 69b24878d030084c22e2f85c |
| Gemini-2.5-Flash | 88.06 | Not recorded | 69b24878d030084c22e2f85c |
| mR3-Qwen3-4B | 87.21 | Not recorded | 69b24878d030084c22e2f85c |
| Qwen3-8B | 86.12 | Not recorded | 69b24878d030084c22e2f85c |
| GPT-4o | 85.75 | Not recorded | 69b24878d030084c22e2f85c |
| Qwen3-4B | 85.06 | Not recorded | 69b24878d030084c22e2f85c |
| Think-as-Locals 7B | 84.51 | Not recorded | 69b24878d030084c22e2f85c |
| M-PROMETHEUS 7B | 78.03 | Not recorded | 69b24878d030084c22e2f85c |
| Qwen2.5-7B-Instruct | 77.89 | Not recorded | 69b24878d030084c22e2f85c |
| M-PROMETHEUS 3B | 68.45 | Not recorded | 69b24878d030084c22e2f85c |
| Qwen2.5-3B-Instruct | 66.97 | Not recorded | 69b24878d030084c22e2f85c |

### DIBJudge-Qwen3-8B

Accuracy: 91.37. Result ID: 69b270606dbb7075b95cca96.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

### DIBJudge-Qwen3-4B

Accuracy: 89.84. Result ID: 69b270606dbb7075b95cca95.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction.

### Nemotron-Multi-49B

Accuracy: 88.83. Result ID: 69b270606dbb7075b95cca8f.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa706f1e209131cea4af71.md>)

[Implementation](<https://github.com/i-Eval/FairEval>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models","Parameters":"49B"}. This record does not assert independent reproduction.

### mR3-Qwen3-8B

Accuracy: 88.58. Result ID: 69b270606dbb7075b95cca93.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc51e209131cea8a58d.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models","Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

### Gemini-2.5-Flash

Accuracy: 88.06. Result ID: 69b270606dbb7075b95cca8a.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c311e209131cea8cef5.md>)

Recorded evaluation configuration: {"Model Category":"Proprietary Models"}. This record does not assert independent reproduction.

### mR3-Qwen3-4B

Accuracy: 87.21. Result ID: 69b270606dbb7075b95cca92.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc51e209131cea8a58d.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models","Backbone":"Qwen3-4B"}. This record does not assert independent reproduction.

### Qwen3-8B

Accuracy: 86.12. Result ID: 69b270606dbb7075b95cca8e.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Model Category":"General Open Models","Parameters":"8B"}. This record does not assert independent reproduction.

### GPT-4o

Accuracy: 85.75. Result ID: 69b270606dbb7075b95cca89.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75681e209131cea67a98.md>)

Recorded evaluation configuration: {"Model Category":"Proprietary Models"}. This record does not assert independent reproduction.

### Qwen3-4B

Accuracy: 85.06. Result ID: 69b270606dbb7075b95cca8d.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Model Category":"General Open Models","Parameters":"4B"}. This record does not assert independent reproduction.

### Think-as-Locals 7B

Accuracy: 84.51. Result ID: 69b270606dbb7075b95cca94.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba81e209131cea89b47.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models","Parameters":"7B"}. This record does not assert independent reproduction.

### M-PROMETHEUS 7B

Accuracy: 78.03. Result ID: 69b270606dbb7075b95cca91.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77e71e209131cea75e93.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models","Parameters":"7B"}. This record does not assert independent reproduction.

### Qwen2.5-7B-Instruct

Accuracy: 77.89. Result ID: 69b270606dbb7075b95cca8c.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Model Category":"General Open Models","Parameters":"7B"}. This record does not assert independent reproduction.

### M-PROMETHEUS 3B

Accuracy: 68.45. Result ID: 69b270606dbb7075b95cca90.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77e71e209131cea75e93.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models","Parameters":"3B"}. This record does not assert independent reproduction.

### Qwen2.5-3B-Instruct

Accuracy: 66.97. Result ID: 69b270606dbb7075b95cca8b.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Model Category":"General Open Models","Parameters":"3B"}. This record does not assert independent reproduction.

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