Category ranking · 8 models · checked 2026-10-13
Multilingual and translation
Ranked by LMArena text, non-English category (Elo-style). Source: lmarena-ai/leaderboard-dataset, text/latest, published 2026-10-08 (CC-BY-4.0).
LMArena text, non-English category (Elo-style)
3 with a published score, 5 not yet measured.
# | Model | Parameters | Score | Source | Licence |
|---|---|---|---|---|---|
| 1 | Gemma 3 4B Google | 4.3 B | 1,273.2Sourced | Source gemma-3-4b-it, non_english, 95% CI 1259.2–1287.1, 1,599 votes. | Gemma Terms of Use |
| 2 | Aya Expanse 32B Cohere Labs | 32.3 B | 1,213.4Sourced | Source c4ai-aya-expanse-32b, non_english, 95% CI 1207.1–1219.7, 12,766 votes. | CC-BY-NC-4.0 Check terms before commercial use |
| 3 | Aya Expanse 8B Cohere Labs | 8.0 B | 1,180Sourced | Source c4ai-aya-expanse-8b, non_english, 95% CI 1170.9–1189.1, 4,321 votes. | CC-BY-NC-4.0 Check terms before commercial use |
| — | EuroLLM 9B UTTER project | 9.2 B | Not yet measured | — | Apache-2.0 |
| — | Hunyuan-MT 7B Tencent | 8.0 B | Not yet measured | — | Not declared in HF metadata (see repo) Check terms before commercial use |
| — | NLLB-200 600M Meta | — | Not yet measured | — | CC-BY-NC-4.0 Check terms before commercial use |
| — | Seed-X PPO 7B ByteDance Seed | 7.5 B | Not yet measured | — | OpenMDW |
| — | Tower+ 9B Unbabel | 9.2 B | Not yet measured | — | CC-BY-NC-SA-4.0 Check terms before commercial use |
How to read this: one public benchmark per category, so you can compare like with like. We copy the published number and link it; we do not re-run it, average different benchmarks or fill gaps. Vendor-reported numbers use each vendor's own settings, so treat small gaps as ties. “Not yet measured” means we found no comparable public score, not that the model is weak.
Every model was checked against the Hugging Face API on 2026-10-13: it exists, is public, and has the parameter count and licence shown. Other categories: Math and reasoning · Coding · Vision-language · OCR and documents · Speech to text · Text to speech · Image generation · Embeddings and rerankers · Small fast 'System 1' · Biomedical · Tool use and agents · Time series · Robotics (VLA).
Machine-readable: /api/v1/categories/multilingual.json