Activity
Mon
Wed
Fri
Sun
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
What is this?
Less
More
Muse Collective
648
Free
19 contributions to Muse Collective
7d •
Muse Director V2 Update
One of the community members suggested possibly having the chunks in tabs. I have looked into and come up with this. I just wanted to get some feedback on it before I decide to integrate it. Take a look at the video below 👇 ( I promise it's a short one).
Poll
20 members have voted
4 likes • 7d
Prefer tabs so you know which chunk you're working with, to the exclusion of all others. Too many times I've hacked into a prompt, only to find I had the wrong bit selected.
27d •
Minimax H3 Director second pass workflow issue
Minimax H3 Director v1.4 second pass workflow issue Hi Andy - I'm using the prompt override box for test prompts (Prompt override from socket is checked) - without the second pass, the preview and final output all work ok. I'm setting 18 steps total; 14 steps for initial pass, then 4 for second refine stage. If I check Enable Two Stage Sampling, it crashes after generating the 14 steps first pass (no video output). I'm guessing it should do 4 more passes, then move to the latent upscaler? But the run crashes out; seems to be on the temporal chunks checkbox. Doesn't matter whether the temporal chunking is checked or not, same crash. Suggest the second refinement pass and the upscale be separated? (I might have it wrong, thought refinement passes were using the same first-pass model? Or is it trying to use the upscaler model to do the refinement and upscale at the same time?) [INFO] got prompt [INFO] [MuseMinimaxDirector] mode=Reference (Omni) — up to 7 images, 3 videos, 3 audio, 1376x768, seed=1111429277178777, seed_hunt=False, candidates=1, 1 chunk(s): ~15.0s (total 15.0s) [INFO] [MuseMinimaxDirector] chunk 1/1, seed=1111429277178777, length=362 frames, video_carry=False, audio_carry=False, hybrid=False [INFO] Requested to load MiniMaxH3VideoVAE [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] Model MiniMaxH3VideoVAE prepared for dynamic VRAM loading. 4965MB Staged. 0 patches attached. Force pre-loaded 128 weights: 348 KB. [INFO] Requested to load MiniMaxH3AudioVAE
0 likes • 26d
Hmm, not sure what is breaking it. It works fine without the second refine stage. I'm not happy with the forced Subject 1 to Picture 1, Subject 2 to Picture 2 referencing. Beth needs three images to render correctly, so I have to specify the subject definitions myself. Still playing with it.
0 likes • 10d
Unfortunately not. I'm currently running with the Minimax H3 Motion Director node - it has second pass refinement and latent upscale. I can generate quickly at 0.2mp unitl I find a seed I like, then fix the seed and move up to 0.6mp first pass, then upscale to 1.4mp. Its taking about an hour for 15 seconds, on a 5090, but the quality is pretty good. I needed more flexibility with the reference images; H3 Motion Director allows whatever 9 images, 3 videos and 3 audios you like (8 images if you're chaining segments). I have a lot of props and Beth needs more than one reference, so Andy's basic workflow node, while great, was a little too limited for me. Waiting for v2.
17d •
Muse Minimax Director V1.4 is Live on My AI Rig
After a lot of tinkering, I’ve got the full Muse MiniMax H3 Director workflow running end to end in ComfyUI. All models and nodes are installed, and the first Reference-mode test produced a 5-second video with sound in about 163 seconds. Big thanks to Muse for sharing this!
0 likes • 15d
Great work Steve!
25d •
ROTR Characters
Hi all. Here are some characters I am developing using ChatGPT for a personal project.
1 like • 21d
They look great! Somehow the ?Doc Martens boots on the third image look a tad out of character. I don't normally notice shoes, but my eyes popped straight to them on opening the image. Maybe they're too clean, too well polished? (Maybe he has a good reason?). Love the fourth character with her pale skin, dirty white tech work clothes and gadgets. Best of the lot for me. What is she, some sort of synthetic human/android? Could do with a wash! Looks great, lots of fine details.
1 like • 20d
Great aesthetic!
29d •
MiniMax Director V2 is getting there!
This is a 3+ minute music video. Pretty seemless (a little more accurate prompting needed). The only issue now is that on these long renders the face starts to degrade. Take a look.
0 likes • 25d
@Andy Whittaker that's liquid awesome, that is. Music to my ears.
1 like • 20d
Looking forward to v2 as well. Away for work right now, so having some withdrawal symptoms.
1-10 of 19
@rick-burlow-1670
Beginner AI content creator in Perth, Western Australia. 53, Geologist, Technical Lead for a mining project in outback WA.
Active 7d ago
Joined Sep 12, 2026