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.

Hi, I raised the issue of the candidate Refine step on your YouTube video comments. How can I provide you with more info to figure out the fault? Thanks

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

Building the roadmap around what's actually useful to you. Pick the one you'd get the most value from next, and I'll try to turn it into tutorials/workflows.

I have an AI character, a digital actress if you will, called Bethnimrah. Or simply, Beth. She'd been in my head for over 40 years as an outline of a <person> since my high school Dungeons and Dragons sessions; but the rules back then didn't cater for her.... exotic makeup. A couple of months ago I started mucking around with Bing Image generator. Very quickly found it limited, and moved to ChatGPT. I remembered something I wrote many years ago, back then while I was pitching the idea to a high school mate who used to run our D&D sessions. I typed it into ChatGPT as best I could remember it... what she looked like, the colour of her hair, the markings on her skin, clothes, bracers and necklace she was wearing, the background souk market in 1st-century Alexandria, Egypt (I had some detail, granted. In maybe fifteen lines, I'd described her. The image ChatGPT spat out impressed me. It was close, but not quite there. I edited my prompt, learing how to maniulate the image generator to add the detail I needed and the look of Beth in my head, her voice. Less than 10 minuites later, I had her. There, on a screen. Someone I'd been dying to meet for a very long time, but could never pencil, paint or model her in 3D. There was Beth, smiling out with a toothy grin, sitting next to a market stall, the golden-hour light shining off her golden-tan fur and leopard spots. No, Beth is not human. She's feline, taller than me at 6'2", and built like a combination of cheetah, leopard and lion, with a classically anthropomorphic body with digitigrade feet and a mane to die for. Beth was born in the 1st Centruy AD, somewhere in the upper Nile. She has a long history. She's smart, quick with a sword, and quicker to smile. I'm not usually one for cats (I'm a dog person!) - but this is the start of her story.