Revision History
Qwen2.5-VL · 7 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected the Qwen2.5-VL-3B license (non-commercial Qwen Research License, not Apache 2.0), the claim that the 32B model beats the 72B flagship on MMMU, MMMU-Pro and MathVista (it beats the older Qwen2-VL-72B), the post-training description (about 2 million SFT entries plus DPO), the false claim that DocVQA 96+ had been closed-model-only, and the DeepSeek-R1 timing (plus 23 minor fixes)
Version 7 · Nov 2, 2026, 05:42 PM
- Improve internal links (2026-10-13)
Version 5 · Oct 13, 2026, 10:05 AM
- Added 6 contextual internal links
Version 4 · Jul 23, 2026, 03:33 PM
- Initial AI-authored creation (4609 words).
Version 2 · May 20, 2026, 01:37 PM