Thomson Reuters Releases Thomson-1.0-Small Vision-Language Model

thomsonreuters · hf · 2026-08-26

Thomson Reuters has released the Thomson-1.0-Small vision-language model on Hugging Face. Built on the Qwen3.5-MoE architecture, it uses an image-text-to-text pipeline, supports conversational capabilities, and is compatible with various inference endpoints.

Original post →

More from Multimodal

Multimodal channel →