language benchmark
WMT24++ Leaderboard
WMT24++ is a comprehensive multilingual machine translation benchmark that expands the WMT24 dataset to cover 55 languages and dialects. It includes human-written references and post-edits across four domains (literary, news, social, and speech) to evaluate machine translation systems and large language models across diverse linguistic contexts.
Updated Aug 17, 2026
Models23
Model coverage23
MetricScore
EvidenceB
WMT24++ Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelNV | Score86.7% | Percentile100.0% | Participants23 | EvidenceC | Evaluated |
| Rank02 | ModelNV | Score86.2% | Percentile95.5% | Participants23 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score85.8% | Percentile90.9% | Participants23 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score84.6% | Percentile86.4% | Participants23 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score84.3% | Percentile81.8% | Participants23 | EvidenceC | Evaluated |
| Rank06 | ModelNV | Score83.7% | Percentile77.3% | Participants23 | EvidenceC | Evaluated |
| Rank07 | ModelCO | Score81.0% | Percentile72.7% | Participants23 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score78.9% | Percentile68.2% | Participants23 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score78.3% | Percentile63.6% | Participants23 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score77.6% | Percentile59.1% | Participants23 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score76.3% | Percentile54.5% | Participants23 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score72.6% | Percentile50.0% | Participants23 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score66.6% | Percentile45.5% | Participants23 | EvidenceC | Evaluated |
| Rank14 | ModelGO | Score53.4% | Percentile40.9% | Participants23 | EvidenceC | Evaluated |
| Rank15 | ModelGO | Score51.6% | Percentile36.4% | Participants23 | EvidenceC | Evaluated |
| Rank16 | ModelGO | Score50.1% | Percentile31.8% | Participants23 | EvidenceC | Evaluated |
| Rank17 | ModelGO | Score50.1% | Percentile27.3% | Participants23 | EvidenceC | Evaluated |
| Rank18 | ModelGO | Score46.8% | Percentile22.7% | Participants23 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score45.8% | Percentile18.2% | Participants23 | EvidenceC | Evaluated |
| Rank20 | ModelGO | Score42.7% | Percentile13.6% | Participants23 | EvidenceC | Evaluated |
| Rank21 | ModelGO | Score42.7% | Percentile9.1% | Participants23 | EvidenceC | Evaluated |
| Rank22 | ModelGO | Score35.9% | Percentile4.5% | Participants23 | EvidenceC | Evaluated |
| Rank23 | ModelAC | Score27.2% | Percentile0.0% | Participants23 | EvidenceC | Evaluated |
WMT24++ Highlights
The leading models and scores on this benchmark.
WMT24++ Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for WMT24++
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis wmt24++ AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01NV
- Score
- 86.7%
- Price
- $0.20 input / $0.80 output per 1M tokens
Strengths
- Ranks #1 of 23 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures WMT24++, not total model capability
- 02NV
- Score
- 86.2%
Strengths
- Ranks #2 of 23 compared models
- 95th percentile on this benchmark
- C evidence result
Considerations
- This result measures WMT24++, not total model capability
- 03ACQwen3.7 MaxAlibaba Cloud / Qwen Team
- Score
- 85.8%
- Price
- $2.5 input / $7.5 output per 1M tokens
- Speed
- Up to 5.8 tok/s via Together
Strengths
- Ranks #3 of 23 compared models
- 91th percentile on this benchmark
- C evidence result
Considerations
- This result measures WMT24++, not total model capability
- 04ACQwen3.7-PlusAlibaba Cloud / Qwen Team
- Score
- 84.6%
- Price
- $0.50 input / $3.0 output per 1M tokens
Strengths
- Ranks #4 of 23 compared models
- 86th percentile on this benchmark
- C evidence result
Considerations
- This result measures WMT24++, not total model capability
- 05ACQwen3.6 PlusAlibaba Cloud / Qwen Team
- Score
- 84.3%
- Price
- $0.50 input / $3.0 output per 1M tokens
- Speed
- Up to 16 tok/s via Together
Strengths
- Ranks #5 of 23 compared models
- 82th percentile on this benchmark
- C evidence result
Considerations
- This result measures WMT24++, not total model capability
Selection summary
Best AI Models for WMT24++
Nemotron 3 Super (120B A12B) currently leads WMT24++ with 86.7%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is WMT24++?
What WMT24++ measures and how its scores work.
WMT24++ is a comprehensive multilingual machine translation benchmark that expands the WMT24 dataset to cover 55 languages and dialects. It includes human-written references and post-edits across four domains (literary, news, social, and speech) to evaluate machine translation systems and large language models across diverse linguistic contexts.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- WMT24++
- Modality
- text
- Primary category
- language
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- wmt24++|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about WMT24++.
Which model scores highest on WMT24++?
Nemotron 3 Super (120B A12B) is currently ranked first with 86.7%.
What does WMT24++ measure?
WMT24++ is a comprehensive multilingual machine translation benchmark that expands the WMT24 dataset to cover 55 languages and dialects. It includes human-written references and post-edits across four domains (literary, news, social, and speech) to evaluate machine translation systems and large language models across diverse linguistic contexts.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
23 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.