AI OmniGen
AI
image generation
SHOW HN2024
How HN took it
14 positive, 9 neutral or asking, 0 mixed, 10 critical, out of 33 comments about the project. Thread score +0.20.
Execution: 2 for, 8 againstIdea: 8 for, 0 againstPerformance: 2 for, 1 againstDocs: 1 for, 0 againstPrivacy: 0 for, 1 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Unified architecture handles many image tasks with strong instruction following
From 2 comments
Elegant architecture, trained from scratch, excels at image editing. This looks very interesting! From https://arxiv.org/html/2409.11340v1 > Unlike popular diffusion models,…
ed on HNUsing a single model to unify all image generation tasks, including many computer vision tasks and visual language reasoning, could transform future image generation models.…
sswz on HN
Image editing with reference images works well
From 2 comments
This looks promising. I love how you can reference uploaded images with markup - this is exactly what the field needs more of. After spending the last two weeks generating…
block_dagger on HNI mean, I struggle even getting Dall-E to iterate on one image without changing everything, so this is pretty cool
oatsandsugar on HN
Objections
Generation often hangs or produces no output in the hosted demo
From 4 comments
I left all the defaults as is, uploaded a small image, typed in "cafe," and 15 minutes later I am still waiting on this finishing.
lelandfe on HNSame, I left running for half an hour but nothing happened.
cubefox on HNLeft it running 1 hour nothing happens. Maybe this is a social experiment.
Citizen_Lame on HNWell, at the moment it seems it's not working at all.
freilanzer on HN
Character consistency is limited; clothing changes between images
From 1 comment
It’s not really consistent - or anymore consistent than, say, SDXL with IP adapter. Even in their example images the character they’ve input comes out wearing different clothes.
jowday on HN
Image quality is not outstanding
From 1 comment
I was able to clone the repo and run it locally, even on a Windows machine, with only minimal Python dependency grief. Takes about a minute to create or edit an image on a 4090.…
CamperBob2 on HN
Requests
Support for transparent images
From 1 comment
not yet, still can't generate transparent images
fullstackwife on HN
Better text rendering in images
From 1 comment
Anyone know how it handles Text? That's kind of my deal breaker, I like Ideogram for it's ability to do really cool fonts, etc.
gremlinsinc on HN
Questions
What is the cost per edit
From 1 comment
Curious what's the actual cost for each edit? Will this infra always be reliable?
anyi09881 on HN
How the multimodal aspect works
From 1 comment
could you elaborate on the multi modal aspect of this model?
hyuuu on HN
Alternatives named
ComfyUI
From 1 comment
Pretty cool, comfy ui and community is too cumbersome for me and still results in too much throwaway content
yieldcrv on HN
DALL-E
From 1 comment
This looks promising. I love how you can reference uploaded images with markup - this is exactly what the field needs more of. After spending the last two weeks generating…
block_dagger on HN
Ideogram
From 1 comment
Anyone know how it handles Text? That's kind of my deal breaker, I like Ideogram for it's ability to do really cool fonts, etc.
gremlinsinc on HN
Flux via Replicate or Fal.ai
From 2 comments
The simplest commercial product for finetuning your own model is probably Adobe firefly, although there’s no API access support yet. But there are cheap and only slightly more…
ncoronges on HNThe easiest flux api I’ve seen is with Fal.ai It is expensive though- Flux dev images are like $0.035/image
spaceman_2020 on HN