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.
Multimodal & Grounded benchmark
Every model the catalog carries a published OfficeQA Pro value for, ranked by that value.
A benchmark for grounded reasoning over office-style documents, spreadsheets, charts, and business artifacts. 17 models in the ModelScale catalog have a published value.
Each point is a model with a published OfficeQA Pro value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
17 models with a published OfficeQA Pro 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 | Grounded QA over office artifacts |
|---|---|---|---|
| 1 | Anthropic | 72.9 | |
| 2 | Anthropic | 69 | |
| 3 | Anthropic | 67.7 |
What the benchmark measures and how its value is published, from the source’s own record.
OfficeQA Pro is a multimodal & grounded benchmark published by OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning. A benchmark for grounded reasoning over office-style documents, spreadsheets, charts, and business artifacts. Its evaluation set is described by the source as document and spreadsheet tasks. The source rates its difficulty as enterprise grounded reasoning.
Each model's result is published by OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning as grounded QA over office artifacts. ModelScale prints that value exactly as published, and never converts or rescales it. This page ranks the 17 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.
All leaderboards · Multimodal & Grounded capability leaderboard
| 4 | Moonshot AI | 67.4 |
| 5 | Anthropic | 66.9 |
| 6 | Anthropic | 66.2 |
| 6 | Tencent | 66.2 |
| 8 | Anthropic | 65.6 |
| 9 | Moonshot AI | 63.3 |
| 10 | Z.AI | 62.4 |
| 11 | StepFun | 60.3 |
| 12 | Anthropic | 57.9 |
| 13 | OpenAI | 54.1 |
| 14 | OpenAI | 53.2 |
| 15 | MiniMax | 45.1 |
| 16 | Anthropic | 43.6 |
| 16 | Anthropic | 43.6 |
Evidence key: ObservedLast good
Rows are ordered by the value OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning 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.