reasoning benchmark
VITA-Bench
VITA-Bench evaluates AI agents on real-world virtual task automation, measuring their ability to complete complex multi-step tasks in simulated environments.
Updated Aug 7, 2026
Published models10
Registry coverage10
MetricScore
EvidenceB
VITA-Bench leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 49.7% | 100.0% | 10 | C | |
| 2 | AC | 47.9% | 88.9% | 10 | C | |
| 3 | AC | 45.6% | 77.8% | 10 | C | |
| 4 | AC | 44.3% | 66.7% | 10 | C | |
| 5 | AC | 41.9% | 55.6% | 10 | C | |
| 6 | AC | 35.6% | 44.4% | 10 | C | |
| 7 | AC | 33.6% | 33.3% | 10 | C | |
| 8 | AC | 31.9% | 22.2% | 10 | C | |
| 9 | AC | 29.8% | 11.1% | 10 | C | |
| 10 | AC | 22.0% | 0.0% | 10 | C |
Score distribution
Top published rows on the benchmark's original scale.
VITA-Bench highlights
The top published results on this benchmark's own scale.
What is VITA-Bench?
Definition and scoring fields from the benchmark registry.
VITA-Bench evaluates AI agents on real-world virtual task automation, measuring their ability to complete complex multi-step tasks in simulated environments.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- VITA-Bench
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- vita-bench|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about VITA-Bench.
Which model scores highest on VITA-Bench?
Qwen3.5-397B-A17B is currently ranked first with 49.7%.
What does VITA-Bench measure?
VITA-Bench evaluates AI agents on real-world virtual task automation, measuring their ability to complete complex multi-step tasks in simulated environments.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
10 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.