jjcm

5 ms·

31.
by jjcm 1mo ago
What I don't understand from the blog post is what's different in this offering than github - is it just the agentic tie ins where I can ask my agent to make changes while browsing the PR? Are there any fundamental differences, ie
32.
by jjcm 1mo ago
I'm surprised OpenRouter went for 7B, while fal.ai just raised at $8b, despite fal having far less traffic (semrush reports ~5x more traffic to openrouter). They seem like very parallel businesses, just with focuses on different models
33.
by jjcm 1mo ago
Thank you for this. I indeed had it up on mine. Cloudflare has switched defaults a couple times now, which honestly is wild to me.
34.
by jjcm 1mo ago
Good that it got taken down. I'm not surprised it was made. The sleaze kinda attracts sleaze. It really feels like this headline is, "Startup known for parading young attractive women takes down website that drew too much attentio
35.
by jjcm 1mo ago
27 t/s. I suspect there will be significant speed ups in the coming weeks.
36.
by jjcm 1mo ago
Image->html test for this. Original images: https://image.non.io/neonRamenDesigns.webp Qwen 3.8 build: https://html.non.io/neonRamenQwen3.8-27b Overall I'm very impressed with how well this did. It&
37.
by jjcm 1mo ago
Same image->html test as I showed in the Gemini 3.7 flash thread. Note that GLM isn't multimodal, but it still was able to generate something similar-ish by writing a python script to inspect the image and extract elements from it.
38.
by jjcm 1mo ago
I think they're both testing very different things. The pelican test is testing if a LLM can come up with visuals on its own via writing bezier curves directly. This is testing if it can match visuals that have already been established
39.
by jjcm 1mo ago
The original prompt includes auth codes for API gen, but here's a redacted version: https://non.io/prompt-for-ramen This is the build step generated by my diffusion-based ui tool's copy-for-agent action.
40.
by jjcm 1mo ago
Oh yea, as a disclaimer the models didn't have any instructions to do a mobile version. I haven't tested them on mobile at all.
41.
by jjcm 1mo ago
Disclaimer that I haven't tried this since January, so things may have changed in the last 7mo, but this was my experience at that time: https://x.com/pwnies/status/2010523020629274723 At a high level though,
42.
by jjcm 1mo ago
I have a few custom ones (a post-trained flux 2 checkpoint for web design and a image->metalness map generator that the build step can call for more advanced lighting situations), but gpt-image-2 is better than my own for design, so it&#
43.
by jjcm 1mo ago
The prompt is a build step generated by my tool for image->html conversion, which includes APIs the model can call to generate images/patterns/svgs. https://image.non.io/12275ee8-71e9-4941-823b-e51fec157b4d.we..
44.
by jjcm 1mo ago
Images and build step were generated with my own tool ( https://news.ycombinator.com/item?id=48995754 - it's why I'm often running these img->html tests). Opus can't generate images since A\ doesn't ha
45.
by jjcm 1mo ago
Harness could be a part of it, but worth noting both the Opus and Gemini 3.7 flash tests were both ran through opencode. The grok test was ran through the cursor cli agent however.
46.
by jjcm 1mo ago
Other thoughts: I really think Google has fallen behind here. Even as a high speed offering (this build took ~7min, which is pretty good!), it wont be able to claim dominance for long with cerebras announcing the Sol preview today: https:&
47.
by jjcm 1mo ago
Here's a image->html test. Gemini has always swung above its weight class for vision work, so I'm always eager to try it with this. Original images: https://image.non.io/neonRamenDesigns.webp Gemini 3.7 build:
48.
by jjcm 2mo ago
A reasonable pushback. It originally only reached out to 5, didnt hear back, so reached out to 5 more. I personally pushed it to reach out to an additional 30 after that. One of the difficulties of sourcing this is a lot of the suppliers in
49.
by jjcm 2mo ago
It'll ask you to take over its computer to log in: https://image.non.io/4022ec77-be07-4baa-97e8-ad9d8d9aeb8a.we... After you do you just tell the bot you're done logging in and it'll keep driving. And yea, it
50.
by jjcm 2mo ago
Dupe of https://news.ycombinator.com/item?id=49261532#49263241 . Dang - might be worth merging these two.
51.
by jjcm 2mo ago
I've been using it for the last month or so. IMO in the same way we went from tab complete -> prompts -> agents, this feels like a next step on that evolution. I highly suspect others will be following suit. I was surprised with
52.
Ask HN: What new tools are you using?
3 points by jjcm 2mo ago|0 comments
53.
by jjcm 2mo ago
Sanders Calls on Tech Giants to Let China Close the Already Shortening Gap
54.
by jjcm 2mo ago
https://chat.withfudge.com/share/0gqaxigvh1/give_me_feedback...
55.
by jjcm 2mo ago
A fun golfing project, but: > smol is so smol you can understand it in an afternoon > smol is easy to adapt I would argue that golfed code is not great for these. Asking a LLM to expand out the code and comment it results in very conc
56.
by jjcm 2mo ago
Some feedback: When I enter a url to give me feedback for, it prompts signin, then returns you to the homepage instead of a review for the url you wanted a review for. Conversations on the left should be named. Right now they're just &
57.
by jjcm 2mo ago
Simple and efficient. Nicely done.
58.
by jjcm 2mo ago
A diffusion-based UI design tool I left Figma to build. My core hypothesis is that LLMs are trained to write consistent code (good for a billing function), but they also express design through code, resulting in consistent design as well. D
59.
by jjcm 2mo ago
to be fair, it's running an 8b model from like 2 years ago. Taalas just does the chip design, not the model architecture.
60.
by jjcm 2mo ago
China has caught up is the main takeaway here. The SOTA models are so close that it's really hard to compare them intelligence wise - you have to get a feel for them yourself and what works for you. What I'm really excited for is

More ›