Benchmark
WMT24++ Benchmark
What is WMT24++?
WMT24++ is a comprehensive multilingual machine translation benchmark that expands the WMT24 dataset to cover 55 languages and dialects. It includes…
- Modality
- text
- Evaluates
- language
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 23
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/wmt24%2B%2B?top_n=500
WMT24++ results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Command A+ | Cohere | 0.81 | 2026-05-20 | evidence |
| Gemma 3 12B | 0.516 | 2025-03-12 | evidence | |
| Gemma 3 1B | 0.359 | 2025-03-12 | evidence | |
| Gemma 3 27B | 0.534 | 2025-03-12 | evidence | |
| Gemma 3 4B | 0.468 | 2025-03-12 | evidence | |
| Gemma 3n E2B Instructed | 0.427 | 2025-06-26 | evidence | |
| Gemma 3n E2B Instructed LiteRT (Preview) | 0.427 | 2025-05-20 | evidence | |
| Gemma 3n E4B Instructed | 0.501 | 2025-06-26 | evidence | |
| Gemma 3n E4B Instructed LiteRT Preview | 0.501 | 2025-05-20 | evidence | |
| Nemotron 3 Nano (30B A3B) | NVIDIA | 0.862 | 2025-12-15 | evidence |
| Nemotron 3 Super (120B A12B) | NVIDIA | 0.8667 | 2026-03-11 | evidence |
| Nemotron 3 Ultra (550B A55B) | NVIDIA | 0.837 | 2026-06-04 | evidence |
| Qwen3.5-0.8B | Qwen | 0.272 | 2026-03-02 | evidence |
| Qwen3.5-122B-A10B | Qwen | 0.783 | 2026-02-24 | evidence |
| Qwen3.5-27B | Qwen | 0.776 | 2026-02-24 | evidence |
| Qwen3.5-2B | Qwen | 0.458 | 2026-03-02 | evidence |
| Qwen3.5-35B-A3B | Qwen | 0.763 | 2026-02-24 | evidence |
| Qwen3.5-397B-A17B | Qwen | 0.789 | 2026-02-16 | evidence |
| Qwen3.5-4B | Qwen | 0.666 | 2026-03-02 | evidence |
| Qwen3.5-9B | Qwen | 0.726 | 2026-03-02 | evidence |
| Qwen3.6 Plus | Qwen | 0.843 | 2026-04-02 | evidence |
| Qwen3.7 Max | Qwen | 0.858 | 2026-05-19 | evidence |
| Qwen3.7-Plus | Qwen | 0.846 | 2026-05-31 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.