Viewing profile — sophrocyne
sophrocyne
HN member- Joined
- Wed, Mar 25, 2020, 11:57 AM UTC
- HN karma
- 511
- Public activity
- 93 items
- HN profile
- View on Hacker News ↗
About sophrocyne
No profile information was provided.
Recent public activity
- story
-
comment
Comment #43963540
The USCO report was flawed, biased, and hypocritical. A pre-publication of this sort is also extremely unusual. https://chatgptiseatingtheworld.com/2025/05/12/opinion-why-t...
-
comment
Comment #43763958
Time to vibe code the data into "That supa awesome database over there"
-
comment
Comment #43405142
It is. https://news.artnet.com/art-world/invoke-snags-first-ai-imag...
-
comment
Comment #43286850
For context, I'm Invoke's CEO. I'm not going to argue that they're not complicated tools - Invoke and Comfy are both actively used to push the boundaries of what can be achieved in…
-
comment
Comment #43286433
Free lunch means it doesn’t require training to achieve the results - it solves for it at run time. You can use this in local generative media tools like Invoke or Comfy, which are…
-
comment
Comment #43001918
Forgot to post some context. I'm the CEO at Invoke (i.e., the creator of this piece) -- we're one of the longest running projects for open-source image generation, originating from…
- story
-
comment
Comment #42367230
Hey all. I'm the CEO of Invoke - appreciate everyone who has mentioned us in the thread. To OP -- We work with professional artists regularly, and I'm seeing things pick up as more…
-
comment
Comment #41735102
Invoke is model agnostic, and supports Flux, including quantized versions.
- story
-
comment
Comment #40775050
The Invoke team released regional guidance using IP Adapter a few months ago, which can use color palettes + style transfer mode, along with text prompts and controlnets. Would tak…
- story
- story
-
comment
Comment #39652767
Some perspectives from someone working in the image space. These tests don't feel practical - That is, they seem intended to collapse the model, not demonstrate "in the wild" perfo…
-
comment
Comment #39641009
There is a ton you can do to help SOTA AI remain open. Join the community building the tools - Help with UI/UX, documentation, keeping up with the latest, and evangelizing whatever…
- story
-
comment
Comment #39364263
Thanks for calling us out - I'm one of the maintainers. Not entirely sure we'll be in the Stable Cascade race quite yet. Since Auto/Comfy aren't really built for businesses, they'l…
-
comment
Comment #39353370
You bet - Thanks for the feedback. Glad you're enjoying Invoke! We do have Docker packages hosted on GH, but I'll be the first to admit that we haven't prioritized ROCm. Contributo…
-
comment
Comment #39351237
Hey there - I'm a maintainer (and CEO) of Invoke. It's something we're monitoring as well. ROCm has been challenging to work with - we're actively talking to AMD to keep apprised o…
-
comment
Comment #39113311
Yes, indeed. Lincoln's still an active maintainer. Energy efficiency is key - Especially with some of these extremely inefficient (wasteful, even) features like real-time canvas. G…
-
comment
Comment #39111335
Alex - I run Invoke (one of the popular OSS SD UIs for pros) Thanks for your work - it’s been impactful since the early days of the project. Excited to see where we get to this yea…
-
comment
Comment #38493099
Yeah, Invoke's nodes/workflows backend can be hit via the API. That's how the entire front-end UI (and workflow editor/IDE) are built. I'm positive this can be done w/ Comfy too.
-
comment
Comment #38488698
https://github.com/invoke-ai/InvokeAI - runs on Mac silicon, can squeeze out SDXL images on a 16gb mac with SSD-1B or Turbo models.
-
comment
Comment #38488676
There are already a number of local, inference options that are (crucially) open-source, with more robust feature sets. And if the defense here is "but Auto1111 and Comfy don't hav…