An interactive real-world agent benchmark grounded in practical consumer-service tasks such as delivery, in-store consumption, and online travel workflows.
As of March 2026, Qwen3.6 Plus leads the VITA-Bench leaderboard with 44.3% , followed by Qwen3.5 397B (43.7%) and GLM-5 (37.0%).
Qwen3.6 Plus
Alibaba
Qwen3.5 397B
Alibaba
GLM-5
Zhipu AI
According to BenchLM.ai, Qwen3.6 Plus leads the VITA-Bench benchmark with a score of 44.3%, followed by Qwen3.5 397B (43.7%) and GLM-5 (37.0%). There is significant spread across the leaderboard, making this benchmark effective at differentiating model capabilities.
8 models have been evaluated on VITA-Bench. The benchmark falls in the Agentic category. This category carries a 22% weight in BenchLM.ai's overall scoring system. VITA-Bench is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
Year
2025
Tasks
Interactive consumer-service agent tasks
Format
End-to-end interactive agent evaluation
Difficulty
Long-horizon real-world workflows
VITA-Bench is built to test realistic interactive agent behavior rather than toy tool calls. It stresses long-horizon coordination, tool selection, changing user intent, and domain switching across daily-life applications.
VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world ApplicationsVersion
VITA-Bench 2025
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
An interactive real-world agent benchmark grounded in practical consumer-service tasks such as delivery, in-store consumption, and online travel workflows.
Qwen3.6 Plus by Alibaba currently leads with a score of 44.3% on VITA-Bench.
8 AI models have been evaluated on VITA-Bench on BenchLM.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.