Workflow Venice AI Workflows and Models (DON'T POST REAL PEOPLE!!)

Me Im able to do nude scene videos right now. But i think u have to make sure u have no copyright stuff linked to your reference image and video. Cause the seedance 2.5 model itself will stop generating otherwise. In higgsfield, they do scan u uploaded files to check any id copyright objects before letting you uploaded it. In venice it doesn’t have that . Make sure what your upload is eligible.
 
Weird. I found both quite good, and the motion and interpretation of prompt to be far more naturalistic. If anything, I'd say it's more Grok-like than anything else so far. But I find with SD, it's always a weird balancing act between the prompt, the choice of image and video reference. I can use the exact same prompt with the same basic structure of reference images, and the exact same video reference, and get rather different results. One thing I will say is that the bleed-through from the video reference has been better and I think the prompt adherence is better than 2.0.
 
I can’t generate shit with my input video on seedance 2.5 even if its convert to mp4. With just input images is fine but when i add input reference video. The model rejects it halfway. Any idea what’s going on? On higgsfield it works just fine.
 
Not really, no. Try cropping or blurring your videos to hide faces and logos. It's probably Venice trying to cover it's ass.
 
Anyone know of a guide for replicating reference videos with new characters in Seedance, preferably one that gives example reference videos and prompts that can be used?
I've wasted thousand of credits trying to replicate videos I've downloaded. Either the motion doesn't match, it completely ignores the reference characters and gives me the reference video as the output, or something else equally useless. I've tried getting Grok to write the prompts but that doesn't always help. Some examples that are known to work that I could try to copy would be really useful.
 
Hi everyone, I'm new to Venice. What's the best template for editing NSFW photos? I'm coming from Grok Imagine in the app and I'm looking for something similar. I used Grok in Venice but it censored all the prompts that worked in Imagine. Thanks!
 
dude seedance 2.5 unlimited plan on higgs is literally awesome. but seedance 2.5 is censored over there! why would you apply your guardrails on an already uncensored model. I don't get it.
guys seriously there isn't any platform other than Venice that offers SD 2.5 uncensored and offers some sort of unlimited plan?
 
I'm sticking with seed 2.0 mini r2v enhanced for now. Unless I really wanna see tits bouncing...
 
Is Higgsfield's moderation really that strict for 2.5? like you can't get anything out?
 
Higgsfield bait and switched their marketing model from NSFW to Professionals, so now pretty much everything is censored out the box.
Everyone should really keep in mind that these AI companies have every incentive to get you to keep generating. It's basically gambling.
 
Gotcha thanks, so basically no point purchasing it unless you only want to generate the most SFW stuff
 
I'm on their lowest plan just for upscaling 480p outputs from Venice using Bytedance Upscale AIGC preset which specializes in AI generated videos. works great for SD 2.0 outputs. Basically what you get if you choose 1080p on Venice and it's 0.9 tokens for the Standard 1080p upscale, 3 tokens for Pro for 15 seconds of video. I can upscale locally using Topaz but I'm lazy.
 
Not since grok uncensored was yeeted. Everything else is weak and expensive
 
I remember back in January Seedance 1.5 Pro costed over a dollar (480p/12s), right now it costs like 10 cents. This is not Venice only, price is severely reduced in all platforms.
The actual question is; the compute required to run Seedance 1.5 Pro didn't change, but price for the model reduced like 90%. So there is something going wrong here. I would appreciate if somebody knows anything about this and explain the logic.
It doesn't makes sense but according to this logic, we can run Seedance 2.0 for very cheap in the near future.
 
wtf is this error???
The model returned the following error:Video generation parameters are invalid.Hint: The parameters `ratio` and `duration` specified in the request are not valid. Seedance identified your task as video editing based on your prompt. For this task type, the output ratio and duration follow the input video selected by the model for editing, and the video selected must satisfy the duration requirement of 4 to 30 seconds. Issues: [0] `ratio` must be `adaptive`. [1] `duration` must be -1.Please adjust your settings and try again. You were not charged or your credits have been refunded.
 
You must log in or register to reply here.
Back
Top