← All models

hunyuan-large-vision

Published Publisher Tencent

Model positioning

hunyuan-large-vision is a vision-language large model trained based on Hunyuan Large, intended for image-text understanding scenarios. (main.qcloudimg.com)

Capabilities and uses

  • Supports combined input of multiple images and text, and generates text content. The official announcement states that it supports image inputs of any resolution.
  • Focuses on image-text understanding tasks; according to the official announcement, its multilingual image-text understanding capabilities have improved significantly, and it can be used in scenarios requiring understanding of both images and text. (main.qcloudimg.com)

Limitations

The output described in the announcement is text; image generation capability has not been verified. The announcement does not specify the context window, the maximum number of images per request, or a list of specific supported languages; actual integration limits and current availability are subject to the service provider's specifications. Model outputs may contain errors, and important results should be verified by a human. (main.qcloudimg.com)

Release date

2025-05-26. (main.qcloudimg.com)

Pricing

As of 2026-11-18, model-specific input and output prices in RMB or USD could not be verified on the Tencent Cloud domestic or international sites. The domestic site's billing page lists other vision models but does not list hunyuan-large-vision; prices for other models cannot be applied to this model. (cloud.tencent.com)

Official sources