Wan
Alibaba's open-weight video generation model family, runnable on consumer GPUs.
why this verdict
Keep it — OpenAI has not replaced it
OpenAI's version overlaps, but does not finish the video job, so this one is still worth keeping open.
- A named launch, not a vibe named, unlinked
- Closed frontier video models improving faster than open weights can follow. OpenAI, January 14, 2026 — no announcement link recorded yet.
- How much of the job it covers not the job editorial call
- Parts of it. The job still needs the tool to get finished.
- Is there a free way to do it? yes
- 2 of 3 listed replacements have a usable free tier: Hunyuan Video and LTX Video.
- What the call is worth nothing to cancel
- No paid entry tier tracked, so there is no subscription to cancel.
- Threatened by
- OpenAI
- Since
- January 14, 2026
- List price
- free
- Per year
- —
The backstory
Wan is the most consequential open-weight video model line, because releasing usable weights let the entire ComfyUI and LoRA ecosystem build on top of it rather than renting access from a lab. That produced a wave of fine-tunes and community workflows no closed model can match on flexibility. Closed models lead on raw quality and probably always will, but open weights mean no per-second billing, no content filter and no vendor able to change the terms, which is decisive for a large group of creators.
Escape hatches
Tencent's competing open-weight video model
hunyuan.tencent.com open_in_newFast open video model from Lightricks
ltx.studio open_in_newHigher quality if you accept a hosted service
klingai.com open_in_new2 of 3 replacements have a usable free tier.