# Runway — Autoregressive-to-Diffusion Vision Language Models

- Company: Runway (runway.com)
- Announced: 2025-09-24T10:10:10.110000+00:00
- Category: research-paper
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://runway.com/research/autoregressive-to-diffusion-vlms
- Record: https://forck.live/items/7902-autoregressive-to-diffusion-vision-language-models
- Subject: Gen / Aleph
- Models affected: A2D-VL 7B

Runway presents A2D-VL 7B, a diffusion vision language model adapted from the pretrained autoregressive VLM Qwen2.5-VL for parallel diffusion decoding. The model uses block size annealing and noise level annealing to preserve base model capabilities while enabling a speed-quality trade-off through adjustable parallelism.

## Evidence

Verbatim from https://runway.com/research/autoregressive-to-diffusion-vlms:

> We develop a state-of-the-art diffusion vision language model, Autoregressive-to-Diffusion (A2D), by adapting an existing autoregressive vision language model for parallel diffusion decoding.

---

Record: https://forck.live/items/7902-autoregressive-to-diffusion-vision-language-models
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md