# A quick Minimax H3 news round-up - 28th September 2026 — retained record

> Retained reddit record from reddit:comfyui.

Canonical: https://highsignal.app/data/records/d81ce3df78bbac11

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:comfyui)  published 2026-09-28 18:25 ingested 2026-09-29 02:43

[original source ↗](https://www.reddit.com/r/comfyui/comments/1wsmjq1/a_quick_minimax_h3_news_roundup_28th_september/)

## retained text



```
Another quick Minimax H3 news and goodies round-up, for those who may have missed some items. 

-> The final 1000-step version of the MiniMax-H3-Character-Swap-LoRA is now available, along with a REF2VA workflow. Takes a source video as <Video 1> and a single-character sheet as <Picture 1>, and assists with swopping them - while keeping all other elements the same. The rendering medium for the character sheet is also transferred (e.g. comic-book style). The drawback is that even if you have close-up facial expressions they are "often not followed" - though that may not matter if you have a monster, space alien, robot/cyborg, or masked superhero. No trigger word is required. (Note than REF2VA can also natively do character swopping, with relevant prompting).

https://huggingface.co/akatz-ai/MiniMax-H3-Character-Swap-LoRA/tree/main

-> An 1980s_scifi_movies LoRA for H3, trained on the cyberpunk end of the genre. No trigger word required.  

https://huggingface.co/neph1/1980s_scifi_movies_minimax_h3

-> A new skill.md to have an LLM write prompts for... "MiniMax H3 still-frame anime illustrations, covering character feature changes, costume changes, and minor modifications."

https://github.com/tsukino4a/h3-image-edit

-> The LoRA trainer Fizgig 6.3 now makes... "sharper MiniMax H3 photo LoRAs".

https://github.com/shootthesound/Fizgig

-> The excellent open-source desktop video editor OpenShot is now at version 4.0.1, with an improved... "Razor tool, faster colour and animation editing, and a timeline that keeps clip controls within reach as you zoom and scroll".

https://www.openshot.org/blog/2026/09/27/openshot-401-improved-razor-faster-editing-linux-snap/

->  And finally... I don't normally feature the 'all-in-one H3' studios, but I see that the new ComfyUI_MiniMaxH3_Director now also has a Face Refine node for ComfyUI. This node also supports Semantic Bridge 'semantic enhancement'.

https://github.com/AIMixer/ComfyUI_MiniMaxH3_Director

https://github.com/AIMixer/ComfyUI_MiniMaxH3_Director/tree/main/director/face_refine

https://github.com/AIMixer/ComfyUI_MiniMaxH3_Director/blob/main/README_EN.md (English readme)

~ OLD POSTS ~

https://old.reddit.com/r/comfyui/comments/1wrqm3l/a_quick_minimax_h3_news_roundup_27th_september/

https://old.reddit.com/r/comfyui/comments/1wqwpah/a_quick_minimax_h3_news_roundup_26th_september/ (See 26th September post, for links to older posts)

https://old.reddit.com/r/comfyui/comments/1wgc4lj/a_quick_minimax_h3_news_roundup_15th_september/ (See 15th September post, for links to even older posts)

https://old.reddit.com/r/comfyui/comments/1w5i9iq/a_quick_minimax_h3_news_roundup_2nd_september_2026/ (See 2nd September post, for links to the starting posts)
```



document fetched 2026-09-29 00:42 · canonical https://reddit.com/r/comfyui/comments/1wsmjq1/a_quick_minimax_h3_news_roundup_28th_september
