Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Agentic benchmark
Every model the catalog carries a published VITA-Bench value for, ranked by that value.
An interactive real-world agent benchmark grounded in practical consumer-service tasks such as delivery, in-store consumption, and online travel workflows. 13 models in the ModelChoice catalog have a published value.
Each point is a model with a published VITA-Bench value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
13 models with a published VITA-Bench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | End-to-end interactive agent evaluation |
|---|---|---|---|
| 1 | Alibaba | 47.9 | |
| 2 | Alibaba | 45.6 | |
| 3 | InternScience | 45.4 |
What the benchmark measures and how its value is published, from the source’s own record.
VITA-Bench is an agentic benchmark published by VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applications. An interactive real-world agent benchmark grounded in practical consumer-service tasks such as delivery, in-store consumption, and online travel workflows. Its evaluation set is described by the source as interactive consumer-service agent tasks. The source rates its difficulty as long-horizon real-world workflows.
Each model's result is published by VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applications as end-to-end interactive agent evaluation. ModelChoice prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 13 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
| 4 | Alibaba | 44.3 |
| 5 | Alibaba | 43.7 |
| 6 | InternScience | 40.3 |
| 7 | InternScience | 38.8 |
| 8 | Alibaba | 35.6 |
| 9 | Anthropic | 23.3 |
| 10 | Meituan | 21.7 |
| 11 | DeepSeek | 18.5 |
| 12 | Anthropic | 17.0 |
| 13 | Z.AI | 15.5 |
Evidence key: Observed
Rows are ordered by the value VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applications published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.