shard isn't about generating content, it's infrastructure for distributed AI compute. instead of paying cloud providers for GPU time, it lets you tap into spare WebGPU capacity from browsers + verification nodes. think of it as shared computing resources rather than an AI assistant. The assistant is just a way to show that distributed inference works. I also plan to add access to the api.
Hi HN! I’ve been building Shard, a browser-powered distributed AI inference network designed to let users contribute compute (via WebGPU) while powerful verifier nodes finalize outputs.
What it is right now
Shard is a functioning early-stage system that lets:
• Browsers act as Scout nodes to contribute WebGPU compute
• A libp2p mesh for P2P networking
• Verifier nodes run stronger local models to validate and finalize inference
• A demo web app you can try live today
• Clients fall back gracefully if WebGPU isn’t available
• Rust daemon + Python API + web UI all wired together
It’s essentially a shared inference fabric — think distributed GPU from volunteers’ browsers + stronger hosts that stitch results into reliable responses. The repo includes tooling and builds for desktop, web, and daemon components.
Why it matters
There’s a growing gap between massive models and accessible compute. Shard aims to:
• Harness idle WebGPU in browsers (scouts)
• Validate and “finish” results on robust verifier nodes
• Enable decentralized inference without centralized cloud costs
• Explore community-driven compute networks for AI tasks
This isn’t just a demo — it’s a full stack P2P inference system with transport, networking, and workflow management.
Current limitations
• Early stage, not production hardened
• Needs more tests, documentation, and examples
• Security and incentive layers are future work
• UX around joining scheduler/mesh could improve
Come build with me
If you’re into decentralized compute, AI infrastructure, web GPU, or mesh networks — I’d love feedback, contributions, and ideas. Let’s talk about where shared inference networks could go next.
The idea for this tool came from a personal pain point: I had just finished a mobile app and needed to localize the screenshots for 10 different countries. I realized I was spending more time in Figma/Photoshop manually editing text boxes than I did coding the app. Google Lens is great for reading, but it ruins the design by slapping ugly boxes over the art.
I built LingoScreen to solve this. It uses AI to:
Detect & Segment: Identify text exactly where it sits.
Inpaint: Use AI to see what’s behind the text and restore the background perfectly.
Contextual Translate: Translate into 50+ languages with 99% accuracy.
Re-render: Put the new text back in a font and style that fits the original design.
Whether you're a solo dev trying to go global or a marketing team handling 100+ assets, LingoScreen turns a week-long design task into a 30-second automated process.
I’d love to hear your thoughts on the accuracy and the UI! I’m here all day to answer your questions.
I wanted to compare local models on my PC, so I built this for myself. I wanted to share it for others to try as well. I am open to suggestions or improvements on this project.
I also do similar work and have run tests on many models. I have listed a few here with sample images using one prompt with a single run. I know it isn't a comprehensive review like OP, but it's something. My personal preference through experience is epicRealismXL.
uBo paired with 1Blocker and Hush has my safari running at 100% blocked on all the adblock test sites. I was only at 84% with Adguard. Its a small improvement, but I'll take it.
Those "adblock test" sites misinform people, they should never be used as they lead people to make bad decisions regarding their choice of content blockers.[1]
Good read: adblock test sites can be wildly inaccurate (alerting to connections that never made it, given redirect to the local shim resource) and can easily be gamed.
You're using "whataboutism" to point fingers and say one side is worse because of this or that. I could do the same thing and say Hillary's emails don't matter because Mike Waltz is out there using Gmail to conduct official business. https://www.axios.com/2025/04/01/mike-waltz-signal-gmail-sec...
To solve this we will all have to come together and accept that nobody on either side of American Politics are on the side of the working class. Instead of pointing fingers at democrats or republicans it is beyond time for us as Americans to come together and vote in people that will work for us as a collective regardless of what political affiliations they have.
I would like to see if there is a correlation between Sleep Apnea cases and nose shape to determine if there is a link between humans living in an area they are not adapted to and how that affects their ability to breathe normally.
BMI is a contributing factor, but so is neck and throat anatomy, regardless of weight. There are plenty of non-overweight people who have sleep apnea. And treatment sucks because so many doctors don't know anything past "you should lose weight".
Ok, what about looking at nose shape of diaspora that have historically (multigenerationally) high BMI? Not an expert but might suggest Inuit, some Pacific Islanders, Māori, west Africans maybe. How are their noses shaped and anything that might help with apnea?
