Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

Diffusion PDF

A Diffusion Image Model Embedded Entirely in a PDF File

Details

External ID
49285429
Source
HN
Company
—
Product
Diffusion PDF
Website domain
alexvd.dev
Launched
Aug. 13, 2026
Cohort
—
Upvotes
5
Upvotes percentile
0.12634408602150538
Tags
—
Fetched at
Sept. 10, 2026, 5:32 a.m.
Updated at
Sept. 10, 2026, 5:32 a.m.

Description

Hey Hacker News! I'm Alex, I'm 19 and I work at Hack Club!The last two weeks i've been messing around with pdfs. I made slop pdf (https://slop.alexvd.dev -- notice the similar website design), a pdf that generates random ai slop stories when you open it, and I was showing that off to one of my friends who suggested I get the LLM to actually work inside the pdf, without an API call to a relay.I laughed at the idea, but then we did some research together, and we managed to find EvanZhouDev's llm.pdf (https://github.com/EvanZhouDev/llm.pdf). Because it existed, I didn't want to waste my time or my tokens building something like it. So I pivoted, eventually deciding to make a diffusion image model work fully inside a pdf.I forked EvanZhouDev's llm.pdf repo, and eventually managed to get it working (with the help of claude). Here's the finished version.It took a while to train, but was worth it so i can show it to my future classmates (I'm an incoming freshman in college this year).

Enrichment

Theme
document processing and generation tools
Vertical
Horizontal
Function
Hardware & robotics
Audience
Developer
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
image generation model packaged in a pdf
Manually corrected
False

Could you build this?

No Compiling or implementing an entire diffusion model runtime into an isolated PDF file using embedded PostScript/JavaScript within strict PDF viewer constraints requires esoteric low-level knowledge of PDF specifications and model quantization.

What it would actually take: The architecture involves heavily quantized neural network weights (INT2/INT4 or tiny UNet/diffusion architectures) embedded as PDF stream objects, combined with custom Acrobat JavaScript or PostScript rendering loops that manually compute matrix multiplications without standard WebGL/WASM runtimes. The hard part is overcoming PDF sandboxing, memory ceilings, and the total lack of standard compute acceleration to execute iterative denoising steps inside PDF readers.

Discussion

No comments on this launch.

Competitors

Other products that read as similar to this one — 79 launches clear the similarity bar, closest 8 shown.

Attention rank: #63 of 80 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 286 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a hardware & robotics tool for Fintech yet.