{"slug": "nano-banana-pro", "name": "Nano Banana Pro", "category": "ai", "category_name": "AI models and assistants", "description": "Google's Gemini-based AI image generation and editing model, also called Gemini 3 Pro Image.", "recommenders": 5, "commenters": 9, "warners": 0, "score": 5.279, "rising": -0.31, "first": "2025-11-20", "latest": "2026-10-24", "why": "Liked for image quality that commenters rate above OpenAI's new image model and Gemini 2.5, and for complex edits and diagrams that follow detailed instructions.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 5, "neutral": 4, "compare": 0, "warn": 0}, "spark": [2, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1], "listed": true, "loved": ["Image quality", "Complex edits"], "caveat": "Face distortion", "short_quote": "if you want to see some legitimately impressive NB Pro outputs, check out:", "icon": false, "domain": null, "github_repo": null, "official_url": "", "pricing": "freemium", "aliases": ["Nano Banana Pro"], "verdict": [{"text": "Liked for image quality that commenters rate above OpenAI's new image model and Gemini 2.5,", "cites": [{"id": 740854178, "item_id": 46303386, "url": "https://news.ycombinator.com/item?id=46303386", "by": "Obertr", "date": "2025-12-17", "stance": "recommend", "quote": "Image model they have released is much worse than nano banana pro", "excerpt": "At this point in time I start to believe OAI is very much behind on the models race and it can't be reversed Image model they have released is much worse than nano banana pro, ghibli moment did not happen Their GPT 5.2 is obviously overfit on benchmarks as a consensus of many developers and friends I know. So Opus 4.5 is staying on top when it comes to coding The weight of the ads money from google and general direct", "context": "", "use_case": "image generation", "switched_from": "", "story": "Gemini 3 Flash: Frontier intelligence built for speed"}, {"id": 736618401, "item_id": 46038650, "url": "https://news.ycombinator.com/item?id=46038650", "by": "ruralfam", "date": "2025-11-24", "stance": "recommend", "quote": "It was leagues ahead of 2.5.", "excerpt": "I recently (last week) used Nano Banana Pro3 for some specific image generation. It was leagues ahead of 2.5. Today I used it to refine a very hard-to-write email. It made some really good suggestions. I did not take its email text verbatim. Instead I used the text and suggestions to improve my own email. Did a few drafts with Gemini3 critiqueing them. Very useful feedback. My final submission about \"..evaluate this ", "context": "", "use_case": "image generation", "switched_from": "", "story": "Three Years from GPT-3 to Gemini 3"}, {"id": 766003841, "item_id": 47875240, "url": "https://news.ycombinator.com/item?id=47875240", "by": "filearts", "date": "2026-04-23", "stance": "neutral", "quote": "Now look at GPT Image 2 and Nano Banana Pro; those original nits are mostly a thing of the past.", "excerpt": "Lot's of comments are nit-picking on how this demo misses the mark on specific technicalities of domains they're familiar with. But that's totally missing the point: this is a demo of what seems like a totally new interaction model that could not previously have existed. It's a small glimpse at a possible future where those nit-picky details get better. Remember when image models produced grotesque hands? Now look at", "context": "", "use_case": "image generation", "switched_from": "", "story": "Website streamed live directly from a model"}]}, {"text": " and for complex edits and diagrams that follow detailed instructions.", "cites": [{"id": 740682464, "item_id": 46292654, "url": "https://news.ycombinator.com/item?id=46292654", "by": "vunderba", "date": "2025-12-16", "stance": "recommend", "quote": "if you want to see some legitimately impressive NB Pro outputs, check out:", "excerpt": "I'll be running gpt-image-1.5 through my GenAI Showdown later today, but in the meantime if you want to see some legitimately impressive NB Pro outputs, check out: https://mordenstar.com/blog/edits-with-nanobanana In particular, NB Pro successfully assembled a jigsaw puzzle it had never seen before, generated semi-accurate 3D topographical extrapolations, and even swapped a window out for a mirror.", "context": "", "use_case": "image generation and editing", "switched_from": "", "story": "GPT Image 1.5"}, {"id": 736069920, "item_id": 46004370, "url": "https://news.ycombinator.com/item?id=46004370", "by": "mark_l_watson", "date": "2025-11-21", "stance": "recommend", "quote": "The results were very good because the diagram reflected what I had specified during chat.", "excerpt": "I used the new Nano Banana Pro just now, indirectly. I was brainstorming with Gemini 3 Thinking mode (now the default best thinking option on my iPadOS Gemini app) over a system design for an open source project that I hope to put a lot of effort into next year and then I asked for a detailed system level diagram. The results were very good because the diagram reflected what I had specified during chat. I probably so", "context": "paid AI practitioner since 1982", "use_case": "system design diagrams", "switched_from": "", "story": "Nano Banana Pro"}]}, {"text": " Caveats include distorted faces when editing photos of people,", "cites": [{"id": 735932368, "item_id": 45995773, "url": "https://news.ycombinator.com/item?id=45995773", "by": "bespokedevelopr", "date": "2025-11-20", "stance": "neutral", "quote": "it does that wonderfully, but in the process it is also hallucinating the images I use", "excerpt": "It\u2019s interesting, I\u2019m trying to use it to create a themed collage by providing a few images and it does that wonderfully, but in the process it is also hallucinating the images I use so I end up with weird distorted faces. Other tools can do this without issue, but something about faces in images this model just has to modify them every time. Ask it to remove background objects and the faces get distorted as well. Us", "context": "", "use_case": "image collage generation", "switched_from": "", "story": "Nano Banana Pro"}]}, {"text": " and one commenter says professional results need third-party tools with more controls.", "cites": [{"id": 742237666, "item_id": 46389854, "url": "https://news.ycombinator.com/item?id=46389854", "by": "Yash16", "date": "2025-12-25", "stance": "neutral", "quote": "especially with Nano Banana Pro \u2014you need proper tools.", "excerpt": "This works fine for casual use, but for professional, high-quality result especially with Nano Banana Pro \u2014you need proper tools. Platforms like https://picxstudio.com and Higgsfield AI deliver way better output with advanced controls. That\u2019s why I use those\u2014and honestly, image editing directly on X just doesn\u2019t cut it.", "context": "", "use_case": "image generation", "switched_from": "", "story": "Twitter (now X) added an \"Edit Image\" feature to edit any image posted with AI"}]}], "praise": [{"label": "Image quality", "summary": "Commenters say its output beats OpenAI's new image model and Gemini 2.5, and that old flaws like bad hands are mostly gone.", "quotes": [{"id": 740854178, "item_id": 46303386, "url": "https://news.ycombinator.com/item?id=46303386", "by": "Obertr", "date": "2025-12-17", "stance": "recommend", "quote": "Image model they have released is much worse than nano banana pro", "excerpt": "At this point in time I start to believe OAI is very much behind on the models race and it can't be reversed Image model they have released is much worse than nano banana pro, ghibli moment did not happen Their GPT 5.2 is obviously overfit on benchmarks as a consensus of many developers and friends I know. So Opus 4.5 is staying on top when it comes to coding The weight of the ads money from google and general direct", "context": "", "use_case": "image generation", "switched_from": "", "story": "Gemini 3 Flash: Frontier intelligence built for speed"}, {"id": 736618401, "item_id": 46038650, "url": "https://news.ycombinator.com/item?id=46038650", "by": "ruralfam", "date": "2025-11-24", "stance": "recommend", "quote": "It was leagues ahead of 2.5.", "excerpt": "I recently (last week) used Nano Banana Pro3 for some specific image generation. It was leagues ahead of 2.5. Today I used it to refine a very hard-to-write email. It made some really good suggestions. I did not take its email text verbatim. Instead I used the text and suggestions to improve my own email. Did a few drafts with Gemini3 critiqueing them. Very useful feedback. My final submission about \"..evaluate this ", "context": "", "use_case": "image generation", "switched_from": "", "story": "Three Years from GPT-3 to Gemini 3"}, {"id": 766003841, "item_id": 47875240, "url": "https://news.ycombinator.com/item?id=47875240", "by": "filearts", "date": "2026-04-23", "stance": "neutral", "quote": "Now look at GPT Image 2 and Nano Banana Pro; those original nits are mostly a thing of the past.", "excerpt": "Lot's of comments are nit-picking on how this demo misses the mark on specific technicalities of domains they're familiar with. But that's totally missing the point: this is a demo of what seems like a totally new interaction model that could not previously have existed. It's a small glimpse at a possible future where those nit-picky details get better. Remember when image models produced grotesque hands? Now look at", "context": "", "use_case": "image generation", "switched_from": "", "story": "Website streamed live directly from a model"}]}, {"label": "Complex edits", "summary": "It handled hard edits like assembling an unseen jigsaw puzzle and swapping a window for a mirror.", "quotes": [{"id": 740682464, "item_id": 46292654, "url": "https://news.ycombinator.com/item?id=46292654", "by": "vunderba", "date": "2025-12-16", "stance": "recommend", "quote": "if you want to see some legitimately impressive NB Pro outputs, check out:", "excerpt": "I'll be running gpt-image-1.5 through my GenAI Showdown later today, but in the meantime if you want to see some legitimately impressive NB Pro outputs, check out: https://mordenstar.com/blog/edits-with-nanobanana In particular, NB Pro successfully assembled a jigsaw puzzle it had never seen before, generated semi-accurate 3D topographical extrapolations, and even swapped a window out for a mirror.", "context": "", "use_case": "image generation and editing", "switched_from": "", "story": "GPT Image 1.5"}]}, {"label": "Accurate diagrams", "summary": "It produced a system diagram that matched what the user had specified in chat.", "quotes": [{"id": 736069920, "item_id": 46004370, "url": "https://news.ycombinator.com/item?id=46004370", "by": "mark_l_watson", "date": "2025-11-21", "stance": "recommend", "quote": "The results were very good because the diagram reflected what I had specified during chat.", "excerpt": "I used the new Nano Banana Pro just now, indirectly. I was brainstorming with Gemini 3 Thinking mode (now the default best thinking option on my iPadOS Gemini app) over a system design for an open source project that I hope to put a lot of effort into next year and then I asked for a detailed system level diagram. The results were very good because the diagram reflected what I had specified during chat. I probably so", "context": "paid AI practitioner since 1982", "use_case": "system design diagrams", "switched_from": "", "story": "Nano Banana Pro"}]}, {"label": "Daily use", "summary": "One former Stable Diffusion and Midjourney user now mostly uses it for image generation.", "quotes": [{"id": 803566048, "item_id": 50222878, "url": "https://news.ycombinator.com/item?id=50222878", "by": "chrisinnes", "date": "2026-10-24", "stance": "recommend", "quote": "tend to use nano banana pro now when i generate", "excerpt": "Thank you for this, I'm a bit behind the times now on local generations and tend to use nano banana pro now when i generate although i used to use stable diffusion and automatic1111 way back, dabbled in everything around and Midjourney 3 onwards. The website shows me clearly where things are at. I expect I'd still like playing with Midjourney for more artistic styled work but it was the cost that drove me away.", "context": "", "use_case": "image generation", "switched_from": "Stable Diffusion", "story": "A practical guide to AI image generation, from diffusion to agents"}]}], "caveats": [{"label": "Face distortion", "summary": "When compositing or editing images of people, it changes and distorts faces.", "quotes": [{"id": 735932368, "item_id": 45995773, "url": "https://news.ycombinator.com/item?id=45995773", "by": "bespokedevelopr", "date": "2025-11-20", "stance": "neutral", "quote": "it does that wonderfully, but in the process it is also hallucinating the images I use", "excerpt": "It\u2019s interesting, I\u2019m trying to use it to create a themed collage by providing a few images and it does that wonderfully, but in the process it is also hallucinating the images I use so I end up with weird distorted faces. Other tools can do this without issue, but something about faces in images this model just has to modify them every time. Ask it to remove background objects and the faces get distorted as well. Us", "context": "", "use_case": "image collage generation", "switched_from": "", "story": "Nano Banana Pro"}]}, {"label": "Limited controls", "summary": "One commenter says professional results need third-party platforms with more advanced controls.", "quotes": [{"id": 742237666, "item_id": 46389854, "url": "https://news.ycombinator.com/item?id=46389854", "by": "Yash16", "date": "2025-12-25", "stance": "neutral", "quote": "especially with Nano Banana Pro \u2014you need proper tools.", "excerpt": "This works fine for casual use, but for professional, high-quality result especially with Nano Banana Pro \u2014you need proper tools. Platforms like https://picxstudio.com and Higgsfield AI deliver way better output with advanced controls. That\u2019s why I use those\u2014and honestly, image editing directly on X just doesn\u2019t cut it.", "context": "", "use_case": "image generation", "switched_from": "", "story": "Twitter (now X) added an \"Edit Image\" feature to edit any image posted with AI"}]}, {"label": "Model visibility", "summary": "The Gemini chatbot UI does not show which image model is being used.", "quotes": [{"id": 787697425, "item_id": 49231089, "url": "https://news.ycombinator.com/item?id=49231089", "by": "gste", "date": "2026-08-09", "stance": "neutral", "quote": "which I think is running Nano Banana Pro for image generation", "excerpt": "I used the Pro subscription for Gemini 3 Pro, which I think is running Nano Banana Pro for image generation, but it doesn't tell you which model in the chatbot UI.", "context": "", "use_case": "image generation", "switched_from": "", "story": "Gemini 3 Pro can generate normal maps"}]}], "who": [{"by": "mark_l_watson", "context": "paid AI practitioner since 1982", "url": "https://news.ycombinator.com/item?id=46004370", "date": "2025-11-21"}], "tags": ["for image editing", "for diagrams", "for Gemini users", "Stable Diffusion alternative"], "use_cases": ["image generation", "image generation and editing", "system design diagrams", "image collage generation"], "compared": [{"slug": "gemini", "name": "Gemini", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=45993296", "title": "Nano Banana Pro", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=50171379", "title": "A practical guide to AI image generation, from diffusion to agents", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49231030", "title": "Gemini 3 Pro can generate normal maps", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47867048", "title": "Website streamed live directly from a model", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46379198", "title": "Twitter (now X) added an \"Edit Image\" feature to edit any image posted with AI", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46301851", "title": "Gemini 3 Flash: Frontier intelligence built for speed", "mentions": 1}], "sentiment_by_month": {"2025-11": {"neutral": 1, "recommend": 2}, "2025-12": {"recommend": 2, "neutral": 1}, "2026-01": {}, "2026-02": {}, "2026-03": {}, "2026-04": {"neutral": 1}, "2026-05": {}, "2026-06": {}, "2026-07": {}, "2026-08": {"neutral": 1}, "2026-09": {}, "2026-10": {"recommend": 1}}, "quotes": [{"id": 803566048, "item_id": 50222878, "url": "https://news.ycombinator.com/item?id=50222878", "by": "chrisinnes", "date": "2026-10-24", "stance": "recommend", "quote": "tend to use nano banana pro now when i generate", "excerpt": "Thank you for this, I'm a bit behind the times now on local generations and tend to use nano banana pro now when i generate although i used to use stable diffusion and automatic1111 way back, dabbled in everything around and Midjourney 3 onwards. The website shows me clearly where things are at. I expect I'd still like playing with Midjourney for more artistic styled work but it was the cost that drove me away.", "context": "", "use_case": "image generation", "switched_from": "Stable Diffusion", "story": "A practical guide to AI image generation, from diffusion to agents"}, {"id": 787697425, "item_id": 49231089, "url": "https://news.ycombinator.com/item?id=49231089", "by": "gste", "date": "2026-08-09", "stance": "neutral", "quote": "which I think is running Nano Banana Pro for image generation", "excerpt": "I used the Pro subscription for Gemini 3 Pro, which I think is running Nano Banana Pro for image generation, but it doesn't tell you which model in the chatbot UI.", "context": "", "use_case": "image generation", "switched_from": "", "story": "Gemini 3 Pro can generate normal maps"}, {"id": 766003841, "item_id": 47875240, "url": "https://news.ycombinator.com/item?id=47875240", "by": "filearts", "date": "2026-04-23", "stance": "neutral", "quote": "Now look at GPT Image 2 and Nano Banana Pro; those original nits are mostly a thing of the past.", "excerpt": "Lot's of comments are nit-picking on how this demo misses the mark on specific technicalities of domains they're familiar with. But that's totally missing the point: this is a demo of what seems like a totally new interaction model that could not previously have existed. It's a small glimpse at a possible future where those nit-picky details get better. Remember when image models produced grotesque hands? Now look at", "context": "", "use_case": "image generation", "switched_from": "", "story": "Website streamed live directly from a model"}, {"id": 742237666, "item_id": 46389854, "url": "https://news.ycombinator.com/item?id=46389854", "by": "Yash16", "date": "2025-12-25", "stance": "neutral", "quote": "especially with Nano Banana Pro \u2014you need proper tools.", "excerpt": "This works fine for casual use, but for professional, high-quality result especially with Nano Banana Pro \u2014you need proper tools. Platforms like https://picxstudio.com and Higgsfield AI deliver way better output with advanced controls. That\u2019s why I use those\u2014and honestly, image editing directly on X just doesn\u2019t cut it.", "context": "", "use_case": "image generation", "switched_from": "", "story": "Twitter (now X) added an \"Edit Image\" feature to edit any image posted with AI"}, {"id": 740854178, "item_id": 46303386, "url": "https://news.ycombinator.com/item?id=46303386", "by": "Obertr", "date": "2025-12-17", "stance": "recommend", "quote": "Image model they have released is much worse than nano banana pro", "excerpt": "At this point in time I start to believe OAI is very much behind on the models race and it can't be reversed Image model they have released is much worse than nano banana pro, ghibli moment did not happen Their GPT 5.2 is obviously overfit on benchmarks as a consensus of many developers and friends I know. So Opus 4.5 is staying on top when it comes to coding The weight of the ads money from google and general direct", "context": "", "use_case": "image generation", "switched_from": "", "story": "Gemini 3 Flash: Frontier intelligence built for speed"}, {"id": 740682464, "item_id": 46292654, "url": "https://news.ycombinator.com/item?id=46292654", "by": "vunderba", "date": "2025-12-16", "stance": "recommend", "quote": "if you want to see some legitimately impressive NB Pro outputs, check out:", "excerpt": "I'll be running gpt-image-1.5 through my GenAI Showdown later today, but in the meantime if you want to see some legitimately impressive NB Pro outputs, check out: https://mordenstar.com/blog/edits-with-nanobanana In particular, NB Pro successfully assembled a jigsaw puzzle it had never seen before, generated semi-accurate 3D topographical extrapolations, and even swapped a window out for a mirror.", "context": "", "use_case": "image generation and editing", "switched_from": "", "story": "GPT Image 1.5"}, {"id": 736618401, "item_id": 46038650, "url": "https://news.ycombinator.com/item?id=46038650", "by": "ruralfam", "date": "2025-11-24", "stance": "recommend", "quote": "It was leagues ahead of 2.5.", "excerpt": "I recently (last week) used Nano Banana Pro3 for some specific image generation. It was leagues ahead of 2.5. Today I used it to refine a very hard-to-write email. It made some really good suggestions. I did not take its email text verbatim. Instead I used the text and suggestions to improve my own email. Did a few drafts with Gemini3 critiqueing them. Very useful feedback. My final submission about \"..evaluate this ", "context": "", "use_case": "image generation", "switched_from": "", "story": "Three Years from GPT-3 to Gemini 3"}, {"id": 736069920, "item_id": 46004370, "url": "https://news.ycombinator.com/item?id=46004370", "by": "mark_l_watson", "date": "2025-11-21", "stance": "recommend", "quote": "The results were very good because the diagram reflected what I had specified during chat.", "excerpt": "I used the new Nano Banana Pro just now, indirectly. I was brainstorming with Gemini 3 Thinking mode (now the default best thinking option on my iPadOS Gemini app) over a system design for an open source project that I hope to put a lot of effort into next year and then I asked for a detailed system level diagram. The results were very good because the diagram reflected what I had specified during chat. I probably so", "context": "paid AI practitioner since 1982", "use_case": "system design diagrams", "switched_from": "", "story": "Nano Banana Pro"}, {"id": 735932368, "item_id": 45995773, "url": "https://news.ycombinator.com/item?id=45995773", "by": "bespokedevelopr", "date": "2025-11-20", "stance": "neutral", "quote": "it does that wonderfully, but in the process it is also hallucinating the images I use", "excerpt": "It\u2019s interesting, I\u2019m trying to use it to create a themed collage by providing a few images and it does that wonderfully, but in the process it is also hallucinating the images I use so I end up with weird distorted faces. Other tools can do this without issue, but something about faces in images this model just has to modify them every time. Ask it to remove background objects and the faces get distorted as well. Us", "context": "", "use_case": "image collage generation", "switched_from": "", "story": "Nano Banana Pro"}]}