--- title: Nano Banana 2.1 costs 4x less than Pro, and people are... description: Google released Nano Banana 2.1, pitched as matching or beating its previous Pro image model while costing roughly a quarter as much ($0.034 vs $0.134 per 1K... canonical: https://daily.dev/posts/nano-banana-2-1-costs-4x-less-than-pro-and-people-are-checking-uscegblwk twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Nano Banana 2.1 costs 4x less than Pro, and people are checking | daily.dev og:description: Google released Nano Banana 2.1, pitched as matching or beating its previous Pro image model while costing roughly a quarter as much ($0.034 vs $0.134 per 1K... og:url: https://daily.dev/posts/nano-banana-2-1-costs-4x-less-than-pro-and-people-are-checking-uscegblwk og:image: https://api.daily.dev/og/posts/uScegbLwK.png og:image:alt: Nano Banana 2.1 costs 4x less than Pro, and people are checking og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Nano Banana 2.1 costs 4x less than Pro, and people are checking **[Trends](https://daily.dev/sources/trends)** · 2 min read · 0 upvotes · 0 comments ## Summary Google released Nano Banana 2.1, pitched as matching or beating its previous Pro image model while costing roughly a quarter as much ($0.034 vs $0.134 per 1K image). It's live across AI Studio, the Gemini API, Google Cloud, the Gemini app, and Flow, with claimed improvements in instruction following, text rendering, Google Image Search grounding, and multi-character consistency. However, observers questioned whether the pricing implies it's secretly a Pro-tier model, and noted that output in Flow looks similar to the old Nano Banana Pro, raising doubts about whether requests are actually being routed to the new model. Commentary frames the release as a competitive move by Google to reclaim ground lost to Anthropic and OpenAI in image generation. ## Content The sharpest question around Google's Nano Banana 2.1 is whether it's actually a new model. @testingcatalog noticed it showing up in Google Flow with output that looks a lot like Nano Banana Pro, and said it's unclear whether requests are being routed to anything new. Then the pricing landed and the same account put it plainly: "Based on the pricing, it looks like it is a Pro model!?" ## The pitch Google describes it as a "high-quality image generation and editing model with enhanced factuality, visual fidelity, and multi-turn consistency at Flash scale." @_philschmid says it beats the previous Pro model at roughly a quarter of the price: $0.034 per image versus $0.134. Other listed upgrades: - better instruction following - better in-image text rendering - grounding with Google Image Search - consistency for up to 5 characters and 14 reference images It's live in AI Studio, the Gemini API, Google Cloud, the Gemini app and Flow. ## The price sheet Image output costs $30 per million tokens. A 1K image uses 1,120 tokens ($0.034), 2K uses 1,680 ($0.050), and 4K uses 2,520 ($0.076). One report says it's built on Gemini 3.6 Flash, and that the 1K price is down from $0.067 for Nano Banana 2. The same report says there's no free tier and input costs rise to $1.50 per million tokens. So cheaper to generate, pricier to feed. ## The bigger mood The release arrives alongside a less fun change. According to that report, from October 9 free Gemini app users lose Flash and Pro and get only Flash-Lite. AI Plus keeps Flash-Lite and Flash but loses Pro. Only AI Pro and Ultra keep all three. Google also plans low, medium and high effort levels in place of the single extended-thinking option, and Deep Think moves down to AI Pro. Gemini 4 Argon, shown September 30, stays restricted to Google's Fairwind programme for governments and trusted partners. So the free tier shrinks while the image model gets cheaper for paying API users. @kloss_xyz speaks for a lot of people watching this. He used to run Gemini over everything: its design taste beat Claude, and Nano Banana beat ChatGPT Images. Then the frontier moved to Anthropic and OpenAI. But Google still owns the most valuable datasets, and if this release surprises people, "that's more competition and cheaper plans. Frankly, I'd love nothing more." ## Questions this post answers ### How much does Google's Nano Banana 2.1 image model cost per image compared to Nano Banana Pro? Nano Banana 2.1 costs about $0.034 per 1K image versus $0.134 for the previous Pro model, roughly a 4x reduction. Image output is priced at $30 per million tokens, with a 1K image using 1,120 tokens, a 2K image using 1,680 tokens ($0.050), and a 4K image using 2,520 tokens ($0.076). The model ID is gemini-nano-banana-2.1. _Developers pricing out image generation APIs can track model cost comparisons like this on daily.dev._ ### Where is Google's Nano Banana 2.1 model available? It is live in AI Studio, the Gemini API, Google Cloud, the Gemini app, and Flow. Google describes it as offering enhanced factuality, visual fidelity, and multi-turn consistency at Flash-level pricing, along with better instruction following, cleaner in-image text rendering, Google Image Search grounding, and consistency for up to five characters using as many as 14 reference images. _Teams deciding which platform to integrate for image generation follow availability details like these on daily.dev._ ### Is Nano Banana 2.1 actually a different model from Nano Banana Pro or just relabeled pricing? It's unconfirmed. Observers noted the pricing structure looks like it could be a Pro-tier model rebranded at a lower price, and output seen in Flow resembles Nano Banana Pro's results closely enough that it's unclear whether requests are actually being routed to a genuinely new model rather than the old one. _Anyone deciding between image models benefits from watching independent side-by-side tests surface on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 4 discussions and 78 comments across x (as of 2026-10-09). **TL;DR:** Reaction is split between excitement over the 4x price cut and new capabilities (text rendering, multi-character consistency, search grounding) and frequent complaints that outputs look blurry, inconsistent, or indistinguishable from the old Pro model, with many unsure if routing is actually happening. **Sentiment:** 20% positive · 35% mixed · 45% skeptical **The case for** - The price drop ($0.034 vs $0.134) is seen as enabling new workflows like generating multiple images and picking the best one. - Better in-image text rendering is called a real unlock for diagrams, UI mockups, and slides. - Multi-character consistency with many reference images is seen as solving a long-standing pain point for storyboards and comics. - Some found it noticeably faster than Nano Banana Pro. **The pushback** - Many report blurry faces, plastic skin, and poor instruction/character-sheet following, calling the update a regression from Nano Banana 2. - Several suspect Flow and other surfaces are just quietly routing to the existing Pro model rather than a genuinely new model. - Multi-turn edits reportedly suffer from pixel drifting and inconsistent faces across revisions. - Input token costs are noted as 2.5-3x higher, raising doubts about real savings on heavy multi-turn editing. - Lack of alpha/transparent background support persists as an unaddressed request. **By community** - x (heated): Opinions are sharply divided between people impressed by the pricing and capability claims and others reporting degraded quality or skepticism that it's a real new model. **Hottest debate:** Whether 2.1 is a genuinely new model or just the existing Pro model relabeled/rerouted. **Open questions** - Does multi-turn consistency actually hold past several rounds of edits? - Is Nano Banana 2 being deprecated now that 2.1 exists at a similar or lower price? - Does higher input token cost offset the lower per-image output price in heavy multi-turn workflows? - Will transparent background/alpha support ever be added? **Highlights** > @_philschmid Shittiest update ever, zero character consistency, every model i make are skinny af. It neither follow character sheet nor instructions it sucks. scrap this shitty 2.1 completely or atleast add option to choose nb2. We don't want this shitty 2.1. It's unusable. > — [youtoobar on x · 2 points](https://x.com/youtoobar/status/2107520142335005017) > @_philschmid $0.034 vs $0.134 at better quality. that's not an upgrade, that's a price war. > — [LiuweijiaVip on x](https://x.com/LiuweijiaVip/status/2107515116149186630) > @testingcatalog Flow got Nano Banana 2 as its default the day it launched in February, so 2.1 showing up there first makes sense. With output that looks like Pro though, it could just be a relabel until Google says otherwise. #NanoBanana #Gemini > — [VisiveAI on x](https://x.com/VisiveAI/status/2107242930494898664) > @testingcatalog Pro's 4K runs roughly $0.24 per image, so this comes in at about a third of that. If the multi-turn consistency holds, long edit chains get a lot cheaper to run. > — [DoDataThings on x](https://x.com/DoDataThings/status/2107512289868017691) > @_philschmid @samsheffer The pixel drifting is insane! How is this model even usable? > — [Xxi5olc on x · 1 points](https://x.com/Xxi5olc/status/2107521651772678321) **Source threads** - [x](https://x.com/testingcatalog/status/2107215396231348558) · 1 points · 22 comments - [x](https://x.com/_philschmid/status/2107502894685749672) · 0 points · 29 comments - [x](https://x.com/testingcatalog/status/2107499443532530001) · 0 points · 27 comments - [x](https://x.com/kloss_xyz/status/2107519622274810121) · 0 points · 0 comments ## Similar posts on daily.dev - [Google releases Nano Banana Pro, its latest image generation model](https://daily.dev/posts/google-releases-nano-banana-pro-its-latest-image-generation-model-y3gpig1g0) · TechCrunch · 0 upvotes · 0 comments - [Google Launches Nano Banana 2 Lite and Gemini Omni Flash](https://daily.dev/posts/google-launches-nano-banana-2-lite-and-gemini-omni-flash-mxprhynmy) · C\# Corner · 0 upvotes · 0 comments - [Google releases Nano Banana 2 Lite, its fastest and cheapest AI image generator yet](https://daily.dev/posts/google-releases-nano-banana-2-lite-its-fastest-and-cheapest-ai-image-generator-yet-zbvuuyq2r) · The Next Web · 0 upvotes · 1 comments - [Nano Banana Pro available for enterprise](https://daily.dev/posts/nano-banana-pro-available-for-enterprise-ldnrqcvuk) · Google Cloud · 0 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#google](https://daily.dev/tags/google), [#genai](https://daily.dev/tags/genai), [#google-gemini](https://daily.dev/tags/google-gemini) [View this post on daily.dev](https://daily.dev/posts/nano-banana-2-1-costs-4x-less-than-pro-and-people-are-checking-uscegblwk) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Nano Banana 2.1 costs 4x less than Pro, and people are checking","url":"https://daily.dev/posts/nano-banana-2-1-costs-4x-less-than-pro-and-people-are-checking-uscegblwk","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/nano-banana-2-1-costs-4x-less-than-pro-and-people-are-checking-uscegblwk"},"datePublished":"2026-10-06T16:07:26.324Z","dateModified":"2026-10-09T12:25:10.691Z","description":"Google released Nano Banana 2.1, pitched as matching or beating its previous Pro image model while costing roughly a quarter as much ($0.034 vs $0.134 per 1K...","image":"https://pbs.twimg.com/media/HT9arOaXMAADozS.jpg","thumbnailUrl":"https://pbs.twimg.com/media/HT9arOaXMAADozS.jpg","isAccessibleForFree":true,"articleSection":"Trends","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Trends","logo":"https://media.daily.dev/image/upload/s--ZfSp3asX--/f_auto,q_auto/v1780996004/logos/trends?_a=BAMAMiWQ0","url":"https://daily.dev/sources/trends"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/nano-banana-2-1-costs-4x-less-than-pro-and-people-are-checking-uscegblwk","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,google,genai,google-gemini","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Trends","item":"https://daily.dev/sources/trends"},{"@type":"ListItem","position":3,"name":"Nano Banana 2.1 costs 4x less than Pro, and people are checking"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/nano-banana-2-1-costs-4x-less-than-pro-and-people-are-checking-uscegblwk#faq","mainEntity":[{"@type":"Question","name":"How much does Google's Nano Banana 2.1 image model cost per image compared to Nano Banana Pro?","acceptedAnswer":{"@type":"Answer","text":"Nano Banana 2.1 costs about $0.034 per 1K image versus $0.134 for the previous Pro model, roughly a 4x reduction. Image output is priced at $30 per million tokens, with a 1K image using 1,120 tokens, a 2K image using 1,680 tokens ($0.050), and a 4K image using 2,520 tokens ($0.076). The model ID is gemini-nano-banana-2.1. Developers pricing out image generation APIs can track model cost comparisons like this on daily.dev."}},{"@type":"Question","name":"Where is Google's Nano Banana 2.1 model available?","acceptedAnswer":{"@type":"Answer","text":"It is live in AI Studio, the Gemini API, Google Cloud, the Gemini app, and Flow. Google describes it as offering enhanced factuality, visual fidelity, and multi-turn consistency at Flash-level pricing, along with better instruction following, cleaner in-image text rendering, Google Image Search grounding, and consistency for up to five characters using as many as 14 reference images. Teams deciding which platform to integrate for image generation follow availability details like these on daily.dev."}},{"@type":"Question","name":"Is Nano Banana 2.1 actually a different model from Nano Banana Pro or just relabeled pricing?","acceptedAnswer":{"@type":"Answer","text":"It's unconfirmed. Observers noted the pricing structure looks like it could be a Pro-tier model rebranded at a lower price, and output seen in Flow resembles Nano Banana Pro's results closely enough that it's unclear whether requests are actually being routed to a genuinely new model rather than the old one. Anyone deciding between image models benefits from watching independent side-by-side tests surface on daily.dev."}}]} ```