ERNIE 4.5 Turbo VL
ERNIE 4.5 Turbo VL is the vision-language service in the Qianfan Turbo family. Its product listing includes image understanding and a 128K context.
Architecture and access
This is a hosted endpoint, rather than the ERNIE 4.5 VL 28B downloadable model. Baidu positions it for image-based understanding, translation and related content tasks. Confirm image count and resolution limits in the selected service before preparing large batches.Baidu Qianfan current model services
Practical use
Practical guidance. For document images, compare exact field extraction and reading order, not only the fluency of a summary. Use a document-specific OCR model when you need reproducible page structure and coordinates. Keep the original image available for review.Baidu Qianfan current model services
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | ernie-4.5-turbo-vl |
| Modality | Image and text understanding; text responsesBaidu Qianfan current model services |
Related models
- PaddleOCR-VL-1.6
Baidu · PaddleOCR-VL
A 0.9-billion-parameter document vision-language model with improved training for difficult document regions.
Launch date Downloadable - ERNIE 4.5 VL 28B A3B
Baidu · ERNIE 4.5
Post-trained sparse vision-language model with 28 billion total parameters and three billion active parameters.
Launch date not confirmedDownloadable