Activity
Mon
Wed
Fri
Sun
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
What is this?
Less
More
Muse Collective
810
Free
6 contributions to Muse Collective
2d •
Muse Character Sheet Workflows
Muse Character sheet H3 Standard and Enhanced are now available in the Classroom under Muse Workflows. Muse Workflows: https://www.skool.com/muse-collective-7594/classroom/324bbfb6?md=2c9544a96565449f9c997bf8f0ca53a4 YouTube: https://www.youtube.com/watch?v=mXCQpcZahis Take a look at the video in the classroom or view on YouTube which helps support my channel. Thanks!
1 like • 2d
@Iaroslav Istomin oooh. thanks for that. I've been fighting that one as well, though I was working around it by swapping out the node type.
0 likes • 2d
@Pukka Factor if it's like in mine, it's hidden behind the Text Preview
3d •
Director V2 (WIP + question)
Making my first attempts at a long-form lip sync video. Only about 6 chunks in to 19 total, but this raises a question. How are people avoiding video degradation over time? If you watch the singer's face and the posters on the wall behind him, degradation starts becoming noticeable around 38 to 40 seconds in, and by the 1 minute mark it's fairly pronounced. I was hoping that camera cuts would help. And ... it does seem to render fine when cutting to new faces or new angles. For example, audience closeup around 54 seconds is fine. New profile angle of the performer at 1:05 is also fine. But as soon as the camera returns to that front facing close-up view, it latches back on to the degraded image.
0 likes • 2d
@Andy Whittaker For LoRAs, I currently have them all disabled. I have previously used Realism People but I haven't been using that with this particular project. (I've had some really weird stuff happen in scenes with that LoRA enabled.)
1 like • 2d
@Philip Meadows yes, I noticed that as well, just picking my battles at the moment. :D
4d •
Character Sheets
Hey all! I am updating my Muse H3 Character Sheet node/workfow and would like to have some example sheets that you currently use. I am looking for which layouts you find works best. The new V2 of my Character Sheet workflow has a lot more options and it waaaaaayy more customizable, but I want to make sure there are layouts you would use. I usually use a panel of 5 images, Hero close-up, then full body from 4 angles. Drop your preferred layout below 👇
2 likes • 3d
@Sjt Jt Currently I'm using a multiple-pass process for my character sheets. 1st pass, I use a modified version of the Muse H3 character sheet node that uses the T=1 image VAE instead of video VAE. Comes out as a single still image character sheet with all panels rather than a 360 degree turnaround video. Faster, but a down side is that the face features on full body panels can be blurry. I'm still experimenting with determining if there's a sweet spot for the render size. 7mp still renders relatively fast as a still image with decent quality, but 2mp works fine as well for the next pass. 2nd pass, I take the H3 character sheet into the Qwen 2.1 workflow and ask it to perform a high detail resolution pass. Render this out at the same size as the original H3 sheet, then pass it through the DLSS 5 upscale. I could do everything in a single pass using the Qwen 2.1 character sheets and bypass the H3 step, but I find that the original H3 character sheet does better with maintaining character consistency across all views. And, of course, this is just what I'm using now and that's subject to change at the drop of a hat if I find something else :D
1 like • 3d
@Andy Whittaker I'll definitely take a look at the new version.
8d •
MiniMax H3 Tutorials
Ok everyone. I hope those of you that have the Director V2 node are enjoying it so far. As promised, I am going to get a load of video tutorials recorded. Are there any requests for what would be more useful first? (runpod installation is coming 1st). Let me know in the comments 👇
0 likes • 3d
Perhaps I'm just overlooking something, but is it possible to use different chunk sizes in a project? Right now if I set the chunk size to 12, for example, then all chunks in the project end up fixed at 12 seconds and I'm not seeing a way to override that. I realize that I can add cuts for transitions between chunks and I'm still experimenting with that and learning as I go. Just trying to confirm if I'm overlooking anything or if adding cuts inside the chunks is the intended way to handle scene transitions on a longer film.
1 like • 3d
Ah. Thanks. I was in Music Video mode
5d •
Muse Director V2 error
Not sure if I'm doing anything wrong, but I'm getting the following error trying to test an initial 12 second scene after installation of V2. # ComfyUI Error Report ## Error Details - **Node ID:** 6236 - **Node Type:** MuseMinimaxDirectorComboV2 - **Exception Type:** NameError - **Exception Message:** NameError: name '_muse_authored_chunk_seed' is not defined ## Stack Trace ``` File "E:\Comfy-Desktop\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 548, in execute output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfy-Desktop\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 353, in get_output_data return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\Comfy-Desktop\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 327, in _async_map_node_over_list await process_inputs(input_dict, i) File "E:\Comfy-Desktop\ComfyUI-Installs\ComfyUI\ComfyUI\execution.py", line 315, in process_inputs result = f(**inputs) File "E:\Comfy-Desktop\ComfyUI-Installs\ComfyUI\ComfyUI\custom_nodes\Muse-MiniMax-Director-Combo-V2\muse_minimax_director.py", line 7375, in execute images, audio, compiled_prompt_text, stage1_latent = _run_pass(seed, candidate_idx=0) ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^
1-6 of 6
Active 9h ago
Joined Oct 3, 2026