Stable Diffusion SDXL
Shares rank 18 of 22 image generation models by recommenders
People recommend it for LoRA-based stylizing and inpainting and fast local runs; critics say it is weak at composition and has been surpassed.
People recommend it for LoRA-based stylizing and inpainting and fast local runs; critics say it is weak at composition and has been surpassed.
12 people gave an opinion
- recommend it
- 8
- criticize it
- 4
- approval share
- 67%
Recommended in 5 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: not recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: recommended
- June 2026: not recommended
- July 2026: not recommended
- August 2026: not recommended
- September 2026: not recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 8 people recommend itNeeds 10 or more, not met.
- 2 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 5 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Strong results with LoRAs and inpainting
“SDXL and FLUX models with LoRAs can and do vastly outperform at tons of things singular big models can't or won't do now”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” “SDXL is crap at composition. It's really good with LoRAs to stylize/inpaint though.”
Comment from 7 Dec 2025 in “Z-Image: Powerful and highly efficient image generation model with 6B parameters” “post-training on SDXL and Flux have given me good results here in the past”
Comment from 20 Nov 2025 in “Nano Banana Pro” Still competitive on image quality
“There's not much of a reason to use SD 1.5 over SDXL if image quality is paramount.”
Comment from 26 Nov 2025 in “FLUX.2: Frontier Visual Intelligence” “Even SDXL looks better, and that's a literal dinosaur.”
Comment from 26 Nov 2025 in “FLUX.2: Frontier Visual Intelligence” “dreaded SDXL is still unmatched”
Comment from 24 Oct 2026 in “A practical guide to AI image generation, from diffusion to agents”
What people warn about
Each mark is one comment
Outclassed by newer models
“SDXL has been outclassed for a while”
Comment from 6 Dec 2025 in “Z-Image: Powerful and highly efficient image generation model with 6B parameters” “SDXL has long been surpassed”
Comment from 7 Dec 2025 in “Z-Image: Powerful and highly efficient image generation model with 6B parameters”
Other image generation models people recommend
The four most recommended other image generation models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Nano Banana | 265 | 159 | 63% | 12 | 6 mo5 owners | — |
| gpt-image-1 | 79 | 79 | 50% | 12 | — | — |
| FLUX.2 | 66 | 51 | 56% | 10 | — | — |
| gpt-image-2 | 63 | 26 | 71% | 8 | — | Approved |
| Stable Diffusion SDXLThis page | 8 | 4 | 67% | 5 | — | — |
What people said
All 14 comments that mention it, newest first. Each links to the original comment.
“dreaded SDXL is still unmatched”
“From a prompt adherency perspective, absolutely not.”
“will likely destroy a lot of small details”
“byzantine syntax requirements of the SD/SDXL days”
“the average person doing the work has been very happy with SDXL based models”
“I also run SDXL it runs rather quickly”
“SDXL is crap at composition. It's really good with LoRAs to stylize/inpaint though.”
“SDXL has long been surpassed”
“SDXL continues to shine here.”
“SDXL has been outclassed for a while”
“There's not much of a reason to use SD 1.5 over SDXL if image quality is paramount.”
“Even SDXL looks better, and that's a literal dinosaur.”
“post-training on SDXL and Flux have given me good results here in the past”
“SDXL and FLUX models with LoRAs can and do vastly outperform at tons of things singular big models can't or won't do now”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.