reasoning benchmark
ProfBench
ProfBench evaluates models on professional-domain reasoning and knowledge-work tasks, including search-augmented question answering across expert fields.
Updated Aug 7, 2026
Published models1
Registry coverage1
MetricScore
EvidenceB
ProfBench leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | NV | 56.0% | 100.0% | 1 | C |
ProfBench highlights
The top published results on this benchmark's own scale.
What is ProfBench?
Definition and scoring fields from the benchmark registry.
ProfBench evaluates models on professional-domain reasoning and knowledge-work tasks, including search-augmented question answering across expert fields.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- ProfBench
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- profbench|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about ProfBench.
Which model scores highest on ProfBench?
Nemotron 3 Ultra (550B A55B) is currently ranked first with 56.0%.
What does ProfBench measure?
ProfBench evaluates models on professional-domain reasoning and knowledge-work tasks, including search-augmented question answering across expert fields.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.