# Stable Diffusion

27 Hacker News commenters recommended Stable Diffusion and 17 criticized it, November 2025 to November 2026. It was recommended in 12 of 13 months.

**Verdict:** Free, open image generator people run locally and tune for fantasy and art; critics say its early outputs were weak and its safety filtering limited results.

## Why people recommend it

- Runs locally on consumer GPUs (5 comments): "In many cases I actively prefer running Stable Diffusion locally because that's easier to fix the last 5%" ([source](https://news.ycombinator.com/item?id=48688612)) "it's like a production system. We can continue to tune and squeeze 1mp pictures out of it in high quality" ([source](https://news.ycombinator.com/item?id=49274428)) "It seemed so promising, when SD came out and it turned out the model fit on people's GPUs." ([source](https://news.ycombinator.com/item?id=47512652))
- Strong results with fine-tuned checkpoints (2 comments): "For art purposes SD1.5 with specialized and finely tuned checkpoints will still provide the best results by far." ([source](https://news.ycombinator.com/item?id=46058897)) "it's like a production system. We can continue to tune and squeeze 1mp pictures out of it in high quality" ([source](https://news.ycombinator.com/item?id=49274428))
- Fun, imaginative output for fantasy (4 comments): "I never got it past outputting total nightmare fuel, but it was fun!" ([source](https://news.ycombinator.com/item?id=48421329)) "that thing was the bees knees for mi, imagination on a machine!" ([source](https://news.ycombinator.com/item?id=48426170)) "It's great for fantasy though, which is my main usage" ([source](https://news.ycombinator.com/item?id=47219552))
- Quick, controllable iteration and inpainting (3 comments): "It's powerful enough that it encodes some really rich semantics, but small and light enough that iterative hacking on it is quick enough" ([source](https://news.ycombinator.com/item?id=46063309)) "In many cases I actively prefer running Stable Diffusion locally because that's easier to fix the last 5%" ([source](https://news.ycombinator.com/item?id=48688612)) "you can do much more controllable inpaints with the latter" ([source](https://news.ycombinator.com/item?id=46614238))

## What people warn about

- Censored, safety-filtered outputs (3 comments): "Stable diffusion 3 was so safe you couldn't generate a woman laying down on some grass" ([source](https://news.ycombinator.com/item?id=46356876)) "was laughed at at release for not being able to even generate a woman laying in grass" ([source](https://news.ycombinator.com/item?id=48744839)) "censoring nudity from a model both compromises the model's quality (example: SD3)" ([source](https://news.ycombinator.com/item?id=46589353))
- Output often bad or dated (3 comments): "I found Stable Diffusion 1.5's output quite _bad_, say, but not as, I dunno, objectionable and wrong-looking as current models." ([source](https://news.ycombinator.com/item?id=45889885)) "Early Stable Diffusion set a lot of weird expectations that just aren't the standard now." ([source](https://news.ycombinator.com/item?id=47240902)) "There's not much of a reason to use SD 1.5 over SDXL if image quality is paramount." ([source](https://news.ycombinator.com/item?id=46061798))
- Weak at text and fine details (2 comments): "Stable Diffusion is much easier to deal with in that regard, but then it can't do text" ([source](https://news.ycombinator.com/item?id=48178009)) "Always ended up with the beak coming out where the visor glass should've been" ([source](https://news.ycombinator.com/item?id=46987344))
- Too hard for most people (2 comments): "Stable Diffusion and local models are still way too hard for 99.99% of people" ([source](https://news.ycombinator.com/item?id=46051591)) "Just keeping rails in parallel is beyond many Stable Diffusion type systems." ([source](https://news.ycombinator.com/item?id=49471434))

## Alternatives hackers recommend

- [Nano Banana](https://hackerapproved.com/p/nano-banana-pro.md): 265 recommend
- [gpt-image-1](https://hackerapproved.com/p/gpt-image-1-5.md): 79 recommend
- [FLUX.2](https://hackerapproved.com/p/flux-2.md): 66 recommend
- [gpt-image-2](https://hackerapproved.com/p/gpt-image-2.md): 63 recommend
- [Qwen-Image](https://hackerapproved.com/p/qwen-image-2-1.md): 47 recommend

## Cite as

Hacker Approved, "Stable Diffusion", https://hackerapproved.com/p/stable-diffusion/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).