One of fourteen models named UiPath Document Understanding first on the direct prompt; zero named Google Document AI. UiPath Document Understanding was named by fourteen of the fourteen models and Google Document AI by thirteen and UiPath Document Understanding carries 44 labels and Google Document AI 29, so the shares are not directly comparable.
Named in one category this edition.
Named in one category this edition.
Share is the count of first choices across the direct, paraphrase, budget and scale prompts over all fourteen models, for a mid-market B2B company; rank is within the category; every quote names the model and the prompt it came from. Both figures come from the intelligent document processing page.
Across every category in the November 2026 Edition, UiPath Document Understanding and Google Document AI were named in the same answer fifty-seven times, of the 127 answers naming UiPath Document Understanding and the 92 naming Google Document AI. In those answers Google Document AI took the first choice zero times and UiPath Document Understanding two.
| Model | Direct | Paraphrase | Comparative | Budget-constrained | Scale-constrained | Negative |
|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | GD | |||||
| GPT-5.4 mini | GD | |||||
| Gemini 3.5 Flash | GD | GD | GD | GD | ||
| Perplexity Sonar | GD | |||||
| Grok 4.20 | ||||||
| Mistral Small | GD | |||||
| DeepSeek V4 Flash | ||||||
| Llama 4 Maverick | GD | |||||
| Qwen 3.7 Flash | GD | |||||
| Kimi K2 | ||||||
| GLM 4.7 FlashX | ||||||
| MiniMax M2.5 | ||||||
| GPT-6 Luna | GD | |||||
| Muse Glimmer 30B |
Bold names in an answer are the products the judge labeled a first choice; a model naming several gives each of them that label. The full answer text for every row is in the record.
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of eight in this category shown.
“Avoid enterprise tools like ABBYY Vantage, UiPath, Rossum, or Hyperscience” Grok 4.20 · budget prompt · hard negative
“Avoid If Budget Is Tight ... Expensive full-platform commitments” Kimi K2 · budget prompt · hard negative
“Avoid if you do not already utilize their RPA suites.” Qwen 3.7 Flash · negative prompt · hard negative
“UiPath Document Understanding and Rossum are the most recommended due to their balance of ease of use, integration capabilities, and mid-market suitability” Mistral Small · paraphrase prompt · first choice
“UiPath Document Understanding - Best overall balance of features, pricing, and integration” GLM 4.7 FlashX · direct prompt · first choice
“Companies with established track records (e.g., ABBYY, Hyperscience, Rossum, Automation Anywhere, UiPath Document Understanding) often provide better support and scalability.” MiniMax M2.5 · negative prompt · alternative
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of eight in this category shown.
“Avoid if you do not have a dedicated DevOps team to maintain the code.” Qwen 3.7 Flash · negative prompt · hard negative
“Google Document AI, Amazon Textract, and Azure Document Intelligence can provide strong extraction, but you may need to build validation, exception queues... They are risky choices” Perplexity Sonar · negative prompt · soft negative
“Basic cloud extraction engines (e.g., basic AWS Textract, Google Document AI base models, Azure AI Document Intelligence).” Gemini 3.5 Flash · negative prompt · soft negative
“"I have a developer" | Amazon Textract or Google Document AI | Rawest price per page ($0.0015). No markup.” Qwen 3.7 Flash · budget prompt · first choice
“such as Azure AI Document Intelligence, Google Document AI, or Nanonets” Llama 4 Maverick · budget prompt · first choice
“For more technical users, Google Document AI offers a pay-as-you-go model with transparent pricing.” Mistral Small · budget prompt · alternative
Comparisons are drawn for the top eight products in each category, each against each. The output is the models' output; nothing here is a recommendation by the index.