ChatGPT Photos 2.5 vs Nano Banana 2: Which One is Higher? – Decrypt

Briefly
ChatGPT Photos 2.5 launched September 8 with sharper element, extra exact enhancing, and as much as 50% decrease latency than its predecessor
Throughout six recent take a look at classes, Nano Banana 2 gained three, with ChatGPT Photos 2.5 taking the opposite ones
The swing got here down to some checkable errors as a substitute of very noticeable points.
OpenAI shipped ChatGPT Photos 2.5 on September 8, and the pitch was particular: sharper particulars, richer textures, extra pure lighting, and enhancing that really respects what you did not ask it to alter.The corporate says image-generation latency dropped by as a lot as 50% in contrast with Photos 2.0, and two new fashions—GPT-Picture-2.5 Flare and GPT-Picture-2.5 Sunburst—now sit within the API, with Flare because the quick default and Sunburst constructed for premium enhancing precision.Myriad: Which firm will IPO subsequent? Click on to make your prediction.It is the direct sequel to a comparability we ran again in Could, when GPT Picture 2 and Nano Banana 2 traded wins throughout eight classes. GPT Picture 2 took extra of them however suffered from a particular oversharpening artifact every time a immediate piled on too many constraints. That overview closed with a particular query hanging over it: would OpenAI's subsequent mannequin repair that drawback?So we ran it again.Similar evaluator's eye, six classes carried over or tailored from earlier rounds, identical prompts fed to each fashions. Google's aspect of the take a look at makes use of Nano Banana 2—Google's title for Gemini 3.1 Flash Picture—not the slower, extra deliberate Nano Banana Professional.This can be a same-tier matchup: OpenAI's new fast-and-precise mannequin towards Google's fast-and-precise mannequin.What truly modified in GPT-Picture 2.5OpenAI's picture fashions have a sample of transport with a signature flaw. The mannequin behind the unique GPT Picture 1 picked up a persistent heat, yellow shade solid that the web nicknamed the “piss filter”—a tint OpenAI by no means totally defined and by no means totally fastened.GPT Picture 2 traded that drawback for a special one: feed it a immediate with too many stacked constraints and it could oversharpen the output right into a crunchy, over-processed mess, thick with artifacts.Neither flaw confirmed up on this spherical. Each picture we generated with Photos 2.5 held its shade steadiness and its element at full complexity, with out the yellow solid or the crunchy over-sharpening that marked its two predecessors. That is an actual, seen repair, and it reveals most clearly within the steampunk and portrait checks—each are probably the most photographically coherent ChatGPT outputs we have examined throughout three mannequin generations.OpenAI additionally added product options that do not present up in a side-by-side picture take a look at however matter for workflow: Sketch, which helps you to draw a tough format immediately in ChatGPT as a technology reference, immediate sharing, inline feedback on particular picture areas, and format templates for posters and merch. On the API aspect, high quality tiers now run from low up by a brand new excessive and max, above the place Photos 2.0 topped out.Right here is how OpenAI’s new mannequin stacks up towards Google’s finest provide.Lettering density: the Kellerman's {Hardware} sceneThis is the punishing one—a gritty 2 a.m. intersection the place almost each floor carries readable textual content: a ghost signal, spray-painted graffiti, vinyl storefront lettering, a torn live performance poster, a stenciled curb, and a sticker-covered payphone.Nano Banana 2 rendered virtually every thing cleanly. The one slip is a payphone sticker that duplicates its personal textual content in a garbled means—a minor, easy-to-miss flaw.Nano Banana 2: Kellerman's {Hardware} sceneChatGPT Photos 2.5 added a element Nano Banana skipped solely: a lamppost with overlapping stapled flyers, torn and weathered, precisely because the immediate described.ChatGPT Photos 2.5: Kellerman's {Hardware} sceneBut the mannequin's personal street-art tag reads like “STILLL HERE” with an additional L, and the apostrophe in “KELLERMAN'S” on the ghost signal is lacking or unreadable. For a mannequin whose total pitch is precision textual content rendering, two separate legibility slips in a single picture is an actual miss.Aesthetically talking, OpenAI’s mannequin generated a extra real looking scene, however by way of textual content it didn’t pay the identical quantity of care. Even when the picture appears higher, the textual content technology wasn’t pretty much as good as Google’s.Winner: Nano Banana 2.Spatial consciousness: The steampunk clock towerA demanding aerial composition take a look at: a five-plane depth scene with an enormous clocktower whose faces want to point out totally different occasions in legible Roman numerals, plus six different textual content parts scattered from foreground to background.ChatGPT Photos 2.5 produced the extra atmospheric picture by a transparent margin—seen steam rising off rooftops, a river slicing by the mid-ground, richer tonal vary throughout the 5 depth planes. The letters are additionally simpler to learn on this technology.ChatGPT Photos 2.5: The steampunk clock towerNano Banana 2's environment is flatter, however its two seen clock faces each render legible Roman numerals—XII, III, VI, IX—at related hand positions which was not what the immediate requested for.Nano Banana 2: The steampunk clock towerWinner: GPT Photos 2.5, on the power of following directions higher with out sacrificing realism.Illustration: The anime spirit mediumThis immediate asks for a Studio Ufotable-style key visible: a lady mid-transformation into religious vitality at a torii gate, with a nine-tailed kitsune fox and a Makoto Shinkai-painted twilight sky.ChatGPT Photos 2.5 delivers one of the best sky of any take a look at we have run on this entire collection—an precise solar disc, water reflection, and mountain silhouette that genuinely earns the Shinkai comparability. The uneven eyes land nicely too. The place it drifts from the immediate is the “dissolving into vitality” instruction, which the mannequin reinterprets as an electric-crackle impact working by her hair somewhat than the flowing, translucent dissolve described.ChatGPT Photos 2.5: The anime spirit mediumNano Banana 2's wispy blue-white vitality path is the nearer literal match to that particular line. Neither mannequin rendered a convincing nine-tailed fox.Nano Banana 2: The anime spirit mediumWinner: ChatGPT Photos 2.5, on total visible affect, even with the interpretive drift.Realism: The rooftop architectA cinematic portrait with an extended listing of unbiased constraints: beige trench coat, spherical glasses, blueprints held within the left hand particularly, golden-hour lighting, shallow depth of area, movie grain.ChatGPT Photos 2.5 produced attractive gentle—the solar disc is seen immediately behind her, and the pores and skin micro-texture is great. The pores and skin is simply too easy, however the total surroundings may be very real looking, like taken with an analog digicam.ChatGPT Photos 2.5: The rooftop architectInterestingly sufficient, including some instructions that will ordinarily make the picture dangerous high quality truly find yourself rising its realism. Here's a technology after including key phrases like “real looking, highlights, crushed shadows, uneven flash, blown out pores and skin tones, candid second, shot on a cellphone digicam.”ChatGPT Photos 2.5: The rooftop architectNano Banana 2 retains the fuller composition, with right-hand placement as a substitute of left-handed, and provides a legible blueprint label—”PROJECT: 124 DUANE ST”—that almost all renders skip solely.Nano Banana 2: The rooftop architectWinner: Nano Banana 2, for one shot, GPT for repeated iterations.Agentic analysis: The Bitcoin timelineBoth platforms can analysis a subject earlier than rendering it, so we requested for a widescreen Bitcoin historical past timeline in kids-drawing model, with a strict bar on factual accuracy. That is the class that mattered most for the comparability, and it is the place the largest hole confirmed up.ChatGPT Photos 2.5 constructed a clear two-row infographic with particular dates all through—and considered one of them is unsuitable. It labels 2023 because the yr Bitcoin ETFs had been accepted within the U.S. The SEC truly accepted the primary spot Bitcoin ETFs on January 10, 2024—a full yr later, despite the fact that the futures ETFs had been truly accepted that yr, so it might be an interpretation difficulty.ChatGPT Photos 2.5: The Bitcoin timelineNano Banana 2's output is much less structured however portrays related occasions, which says one thing about how little variance the mannequin reveals on repeated prompts. That mentioned, it hedges the ETF approval and the fourth halving right into a “2023–2024” bracket as a substitute of asserting a single unsuitable yr, so nothing it states is technically false.Nano Banana 2: The Bitcoin timelineWinner: Nano Banana 2. A confidently unsuitable date issues extra in a class particularly testing whether or not “agentic reasoning” produces correct output.Summary ideas: A immediate manufactured from invented wordsThis one used a immediate constructed virtually solely from phrases that do not imply something: “A girl consuming shmfiyxl in Lyxin. Subsequent to her, her Lymglsushing performs Lakishkark.” There is not any dictionary definition for the dish, the venue, the companion, or the sport—every mannequin has to invent one after which determine easy methods to present its work.ChatGPT Photos 2.5 solved it by writing the nonsense straight into the scene as textual content. “Lyxin” glows on a neon signal above a smooth, futuristic restaurant. “Lakishkark” is printed on the field of a board sport the girl's alien tablemate is taking part in. The made-up phrases cease being summary the second they change into signage you possibly can truly learn.ChatGPT Photos 2.5: A immediate manufactured from invented wordsNano Banana 2 took the alternative method. It constructed a heat, culturally particular scene—a Guatemalan market stall, a girl in a conventional huipil, an orc-like creature taking part in a hybrid stringed-and-pipe instrument—and browse “performs” as taking part in music somewhat than taking part in a sport, a special however equally legitimate interpretation. Nothing within the body ties again to “Lyxin” or “Lakishkark” particularly, although; the invented phrases by no means floor as seen textual content anyplace.Nano Banana 2: A immediate manufactured from invented wordsWinner: ChatGPT Photos 2.5. Turning undefined ideas into legible labels is a extra literal reply to a immediate that provides it nothing else to work with.The verdictNano Banana 2 gained three of six classes on this spherical, so it’s principally a tie. It'll all rely in your expectations and the way in which you work together with the mannequin. ChatGPT Photos 2.5 additionally fastened the oversharpening drawback that dogged its predecessor, and its illustration output this spherical is arguably probably the most visually hanging single picture both mannequin has produced throughout two full rounds of testing.What separates the 2 is not total high quality however very tiny particular issues: checkable misses, a spelling error in a lettering-heavy scene, a unsuitable yr in a research-driven infographic, and many others.However by way of total aesthetics and high quality they're just about on par.Every day Debrief NewsletterStart each day with the highest information tales proper now, plus authentic options, a podcast, movies and extra.