Benchmark
VITA-Bench
What is 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.
- Modality
- text
- Evaluates
- reasoning, agents
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 10
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/vita-bench?top_n=500
VITA-Bench results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Qwen3.5-122B-A10B | Qwen | 0.336 | 2026-02-24 | evidence |
| Qwen3.5-27B | Qwen | 0.419 | 2026-02-24 | evidence |
| Qwen3.5-35B-A3B | Qwen | 0.319 | 2026-02-24 | evidence |
| Qwen3.5-397B-A17B | Qwen | 0.497 | 2026-02-16 | evidence |
| Qwen3.5-4B | Qwen | 0.22 | 2026-03-02 | evidence |
| Qwen3.5-9B | Qwen | 0.298 | 2026-03-02 | evidence |
| Qwen3.6 Plus | Qwen | 0.443 | 2026-04-02 | evidence |
| Qwen3.6-35B-A3B | Qwen | 0.356 | 2026-04-16 | evidence |
| Qwen3.7 Max | Qwen | 0.479 | 2026-05-19 | evidence |
| Qwen3.7-Plus | Qwen | 0.456 | 2026-05-31 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.