OpenAI image generation
Shares rank 10 of 22 image generation models by recommenders
Recommended for realistic images, text inside images, and quick one-off edits that often match the prompt on the first try. Critics say edits can change faces and scene details.
Recommended for realistic images, text inside images, and quick one-off edits that often match the prompt on the first try. Critics say edits can change faces and scene details.
48 people gave an opinion
- recommend it
- 25
- criticize it
- 23
- approval share
- 52%
Recommended in 6 of 13 months
- November 2025: recommended
- December 2025: not recommended
- January 2026: not recommended
- February 2026: not recommended
- March 2026: recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: not recommended
- July 2026: not recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 25 people recommend itNeeds 10 or more, met.
- 1 recommender for every criticNeeds 2 to 1 or better, not met.
- Recommended in 6 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 on first try
“With OpenAI I usually got what I wanted on the first or second try”
Comment from 10 Mar 2026 in “We Tried Paying Artists Royalties on AI-Generated Work – Learnings” “I don't recall ANY examples where it completely failed to do anything.”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” “It's not super deterministic but it didn't fail once on my attempts.”
Comment from 21 Apr 2026 in “OpenAI Livestream” Better than rival image models
“I find the OpenAI results head-and-shoulders above the others”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” “this thing is like 5x better than flash at fine grain detail”
Comment from 22 Apr 2026 in “OpenAI Livestream” “it's just as good as OpenAI's image create, if not better”
Comment from 4 Nov 2025 in “Studio Ghibli, Bandai Namco, Square Enix Demand OpenAI to Stop Using Their IP” Realistic images good for pro use
“Much of what can be produced by eg. the openai's image 2.5 model is more-or-less indistinguishable from a person's work.”
Comment from 12 Sep 2026 in “Fuck it, make it anyway” “they truly look real and good enough for pro use”
Comment from 22 Apr 2026 in “OpenAI Livestream” Text inside images looks good
“the outputs including text are astonishing”
Comment from 18 May 2026 in “Every AI Subscription Is a Ticking Time Bomb for Enterprise” “The text work inside the images are nice”
Comment from 22 Apr 2026 in “OpenAI Livestream”
What people warn about
Each mark is one comment
Often misses prompt or details
“OpenAI strait up failed to do the prompt a large fraction of the time”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” “Images 2 really struggles, and has yet to complete one I was satisfied with”
Comment from 22 Apr 2026 in “OpenAI Livestream” “the actual details are always wrong, such as showing pawns in wrong locations, missing pawns”
Comment from 21 Apr 2026 in “OpenAI Livestream” - Comment from 20 Nov 2025 in “Nano Banana Pro”
Alters faces and scene details
“OpenAI would alter the structure of their head (making it rounder), eyes (making them rounder)”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” “I've noticed that OpenAI modifies faces on a regular basis.”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” “It's also changing scene features too. Like removed background trees.”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” Color cast and blurring
“intentionally-crappy orange-cyan palette on everything the model generates”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” “the 'piss filter' of openAI image generation hasn't been fixed yet”
Comment from 11 Nov 2025 in “We ran over 600 image generations to compare AI image models” “OpenAI blurs the entire image car and ground fairly uniformly”
Comment from 12 Nov 2025 in “We ran over 600 image generations to compare AI image models” Guardrails block public domain subjects
“OpenAI wouldn't make me a Looney Tunes Roadrunner Martin Scorsese "Absolute Cinema" parody”
Comment from 22 Apr 2026 in “OpenAI Livestream” “OpenAI has ridiculous guardrails for illustrations covering any public domain subject”
Comment from 4 Nov 2025 in “Studio Ghibli, Bandai Namco, Square Enix Demand OpenAI to Stop Using Their IP”
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 |
| OpenAI image generationThis page | 25 | 23 | 52% | 6 | — | — |
What people said
40 of the 54 comments that mention it, newest first. Each links to the original comment.
“Customers lave the Ghibli style image-gen.”
“Much of what can be produced by eg. the openai's image 2.5 model is more-or-less indistinguishable from a person's work.”
“Their use-cases for the recent 'Images' release were also terrible.”
“the outputs including text are astonishing”
“Google and OpenAI are good for one-offs”
“The recent GPT upgrade might actually be the most helpful tool in the space.”
“producing really nice results for turning sprites into sprite animations”
“OpenAI wouldn't make me a Looney Tunes Roadrunner Martin Scorsese "Absolute Cinema" parody”
“editing what it creates still feels underbaked”
“it made a super cool picture for me. It made me smile.”
“Images 2 really struggles, and has yet to complete one I was satisfied with”
“they truly look real and good enough for pro use”
“The text work inside the images are nice”
“I wouldn't say it's a huge step forward in terms of quality compared to what I've seen before from AI”
“this thing is like 5x better than flash at fine grain detail”
“It almost nailed it for me (two squares have both white and black color).”
“the actual details are always wrong, such as showing pawns in wrong locations, missing pawns”
“This is the first time i've found value in an image model for stuff I actually do.”
“It's not super deterministic but it didn't fail once on my attempts.”
“OpenAI is happy to edit and change every time I tried”
“With OpenAI I usually got what I wanted on the first or second try”
“struggled with some OOD images like elephants doing backflips”
“it was ground-breaking when it was released”
“OpenAI blurs the entire image car and ground fairly uniformly”
“it's the only one that seems to know what bokeh is at all”
“OpenAI had some very good results but also some missing the mark.”
“IMO it just failed ukiyo-e Japanese woodblock print style”
“I don't recall ANY examples where it completely failed to do anything.”
“OpenAI strait up failed to do the prompt a large fraction of the time”
“intentionally-crappy orange-cyan palette on everything the model generates”
“OpenAI's image pipeline takes an image as input, derives the semantic details, and then essentially regenerates an entirely new image”
“It's also changing scene features too. Like removed background trees.”
“I've noticed that OpenAI modifies faces on a regular basis.”
“I find the OpenAI results head-and-shoulders above the others”
“prone to taking eight times longer to generate an image before showing it add a third cat over and over and over again”
“the 'piss filter' of openAI image generation hasn't been fixed yet”
“OpenAI would alter the structure of their head (making it rounder), eyes (making them rounder)”
“OpenAI has ridiculous guardrails for illustrations covering any public domain subject”
“openai generated the app icon”
“it's just as good as OpenAI's image create, if not better”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.