general benchmark
NOVA-63
NOVA-63 is a multilingual evaluation benchmark covering 63 languages, designed to assess LLM performance across diverse linguistic contexts and tasks.
Updated Aug 7, 2026
Published models11
Registry coverage11
MetricScore
EvidenceB
NOVA-63 leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 59.1% | 100.0% | 11 | C | |
| 2 | AC | 59.0% | 90.0% | 11 | C | |
| 3 | AC | 58.8% | 80.0% | 11 | C | |
| 4 | AC | 58.6% | 70.0% | 11 | C | |
| 5 | AC | 58.1% | 60.0% | 11 | C | |
| 6 | AC | 57.9% | 50.0% | 11 | C | |
| 7 | AC | 57.1% | 40.0% | 11 | C | |
| 8 | AC | 55.9% | 30.0% | 11 | C | |
| 9 | AC | 54.3% | 20.0% | 11 | C | |
| 10 | AC | 46.4% | 10.0% | 11 | C | |
| 11 | AC | 42.4% | 0.0% | 11 | C |
Score distribution
Top published rows on the benchmark's original scale.
NOVA-63 highlights
The top published results on this benchmark's own scale.
What is NOVA-63?
Definition and scoring fields from the benchmark registry.
NOVA-63 is a multilingual evaluation benchmark covering 63 languages, designed to assess LLM performance across diverse linguistic contexts and tasks.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- NOVA-63
- Modality
- text
- Primary category
- general
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- nova-63|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about NOVA-63.
Which model scores highest on NOVA-63?
Qwen3.5-397B-A17B is currently ranked first with 59.1%.
What does NOVA-63 measure?
NOVA-63 is a multilingual evaluation benchmark covering 63 languages, designed to assess LLM performance across diverse linguistic contexts and tasks.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
11 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is marked as eligible for the current LLMBoard capability methodology.