Apple Open-Sources LensVLM-9B to Save Tokens via Compressed Document Images

Apple has open-sourced LensVLM-9B, a Qwen3.5-9B based vision-language model that compresses long documents into small page images and decompresses relevant pages on demand to save tokens and compute.

2026-09-24 ~ 2026-09-24 · 2 related posts

• Apple open-sources LensVLM-9B, a VLM that reads compressed text images and selectively expands relevant pages — jacek2023 · 2026-09-24
• Apple Open-Sources LensVLM-9B, a Qwen3.5-9B Finetune That Shrinks Long Docs into Page Images — victormustar · 2026-09-24