{"slug": "gemini-flash-lite", "name": "Gemini Flash Lite", "category": "ai", "category_name": "AI models and assistants", "description": "Google's lightweight, low-cost Gemini model tier for high-volume, latency-sensitive tasks.", "recommenders": 5, "commenters": 15, "warners": 1, "score": 6.163, "rising": -0.31, "first": "2026-02-19", "latest": "2026-10-29", "why": "HN commenters like it as a cheap, fast model for classification, summarization and other production workloads.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 5, "neutral": 8, "compare": 1, "warn": 1}, "spark": [0, 0, 0, 0, 0, 0, 0, 2, 2, 1, 0, 0], "listed": true, "loved": ["Production workloads", "Good value"], "caveat": "Model deprecation", "short_quote": "Gemini Flash Lite 3.1 is however  surprisingly good value.", "icon": true, "domain": null, "github_repo": null, "official_url": "https://ai.google.dev/gemini-api/docs/models", "pricing": "paid", "aliases": ["Gemini Flash Lite", "Gemini 2.5 Flash Lite", "Gemini Flash Lite 3.1", "Gemini 3.5 Flash Lite", "Gemini 3.5 Flash-Lite", "Gemini 2.5 Flash-Lite"], "verdict": [{"text": "HN commenters like it as a cheap, fast model", "cites": [{"id": 783905409, "item_id": 48994088, "url": "https://news.ycombinator.com/item?id=48994088", "by": "AussieWog93", "date": "2026-07-21", "stance": "recommend", "quote": "3.5 Flash Lite sounds like it could be a strict upgrade for our use case", "excerpt": "A lot of disappointment here in the comments, but models like these aren't meant to compete with the likes of Fable or GPT 5.6. I use 3.1 Flash Lite regularly to classify listings on eCommerce websites. It's great for this task - fast, cheap and accurate. In fact, it was the single best model we tried in terms of the speed vs accuracy vs price tradeoffs - including the Chinese models. Of course, 3.5 Flash was more ac", "context": "", "use_case": "classifying eCommerce listings", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 779144514, "item_id": 48696532, "url": "https://news.ycombinator.com/item?id=48696532", "by": "sfifs", "date": "2026-06-27", "stance": "recommend", "quote": "Gemini Flash Lite 3.1 is however  surprisingly good value.", "excerpt": "So a lot depends on your specific use case but mid-sized open weight models are pretty actually good now, so this is realistic [1] The first question to ask is does your use case require handling personal or sensitive data. If you're using the LLM for OpenClaw or you want to handle sensitive or medical data, a local model generally is necessary. if it's not so sensitive - Cloud providers with some sort of user agreem", "context": "", "use_case": "", "switched_from": "", "story": "Ask HN: MacBook vs. Dedicated GPU for LLM"}]}, {"text": " for classification, summarization and other production workloads.", "cites": [{"id": 783905409, "item_id": 48994088, "url": "https://news.ycombinator.com/item?id=48994088", "by": "AussieWog93", "date": "2026-07-21", "stance": "recommend", "quote": "3.5 Flash Lite sounds like it could be a strict upgrade for our use case", "excerpt": "A lot of disappointment here in the comments, but models like these aren't meant to compete with the likes of Fable or GPT 5.6. I use 3.1 Flash Lite regularly to classify listings on eCommerce websites. It's great for this task - fast, cheap and accurate. In fact, it was the single best model we tried in terms of the speed vs accuracy vs price tradeoffs - including the Chinese models. Of course, 3.5 Flash was more ac", "context": "", "use_case": "classifying eCommerce listings", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 796255953, "item_id": 49765997, "url": "https://news.ycombinator.com/item?id=49765997", "by": "Oras", "date": "2026-09-19", "stance": "neutral", "quote": "classification tasks that I used Gemini 2.5 flash lite with", "excerpt": "I played around with Jev last night and did it for classification tasks that I used Gemini 2.5 flash lite with. It\u2019s a bit faster and bit cheaper, but this is compared to LLM. The consistency was nice to see, BUT, as someone who trained NLP models prior to LLMs, it\u2019s just BERT with more data. I can see why people would want ready made one shot classifier, and I can see the value of sending multiple classifier in one ", "context": "", "use_case": "classification", "switched_from": "", "story": "I built non-autoregressive decision models with RL a year ago"}, {"id": 787808866, "item_id": 49238054, "url": "https://news.ycombinator.com/item?id=49238054", "by": "schultetwin1", "date": "2026-08-09", "stance": "neutral", "quote": "I use Gemini Flash Lite to summarize and filter the bulletins.", "excerpt": "https://dipstickalerts.com - I've been working on this during my kid's naps. The web app allows car owners in the U.S. to see a list of all the service bulletins the manufacturer has sent out about your car over the years (called Technical Service Bulletins or TSBs). You can subscribe for new bulletins to be sent to your email as well. These service bulletins are not only recalls but also fixes for known issues with ", "context": "", "use_case": "summarizing bulletins", "switched_from": "", "story": "Ask HN: What are you working on? (August 2026)"}, {"id": 788961665, "item_id": 49310104, "url": "https://news.ycombinator.com/item?id=49310104", "by": "scirob", "date": "2026-08-15", "stance": "recommend", "quote": "in production we have been using Gemini Flash Lite as primary", "excerpt": "We maintain German Langauge index as no one publishes or reruns these sepeartly. Qwen 3.8 27B is a small improvement with some regressions in our benchmarks not a huge jump like benchmarks listed. https://dach.peerbench.ai/compare?models=qwen%2Fqwen3.8-27b,qwen%2Fqwen3.6-27b German language has never been a big focus for asian models but they still outperform Gemma models https://dach.peerbench.ai/compare?models=open", "context": "runs German-language production", "use_case": "production primary model", "switched_from": "", "story": "Qwen 3.8 27B is out: open weights, best local dense model yet"}, {"id": 796076371, "item_id": 49754773, "url": "https://news.ycombinator.com/item?id=49754773", "by": "testycool", "date": "2026-09-18", "stance": "neutral", "quote": "I had a production image processing using Gemini 2.5 Flash Lite", "excerpt": "I use models.dev's CLI tool, which I think gets data from OpenRouter, and ArtificialAnalysis so your coding agent can help you narrow it down. <sidenote> Similarly, HuggingFace has a CLI + a few skills, and they are very useful. I had a production image processing using Gemini 2.5 Flash Lite (which is getting discontinued in October), and in 20 minutes Claude Code + HF Cli recommended the best replacement small model", "context": "", "use_case": "production image processing", "switched_from": "", "story": "Qwen 3.8 Omni Flash"}]}, {"text": " The most common complaint is Google discontinuing older versions such as 2.5 Flash Lite,", "cites": [{"id": 784011089, "item_id": 49000693, "url": "https://news.ycombinator.com/item?id=49000693", "by": "do_anh_tu", "date": "2026-07-21", "stance": "recommend", "quote": "please please don't kill the old ones like gemini-2.5-flash-lite, because that all I ever need for my little apps", "excerpt": "Man I love Gemini models but these kind of pricing increase is just insanse. I have a little product and I have to keep increasing the price and reduce the limits because of this non-sense, and they did not even let us use the old models in near future, so I forced to update to the new model with basically no to little improvement because I don't even need that much. Google if you can read this, it okay to release ne", "context": "", "use_case": "small apps", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 783909088, "item_id": 48994318, "url": "https://news.ycombinator.com/item?id=48994318", "by": "zuzululu", "date": "2026-07-21", "stance": "warn", "quote": "We spent thousands just to build this workflow now Google says screw off", "excerpt": "same I just switched to OpenAI after using flash 2.5 lite for almost everything at our company. We spent thousands just to build this workflow now Google says screw off", "context": "company using it for almost everything", "use_case": "company LLM workflows", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 796076371, "item_id": 49754773, "url": "https://news.ycombinator.com/item?id=49754773", "by": "testycool", "date": "2026-09-18", "stance": "neutral", "quote": "I had a production image processing using Gemini 2.5 Flash Lite", "excerpt": "I use models.dev's CLI tool, which I think gets data from OpenRouter, and ArtificialAnalysis so your coding agent can help you narrow it down. <sidenote> Similarly, HuggingFace has a CLI + a few skills, and they are very useful. I had a production image processing using Gemini 2.5 Flash Lite (which is getting discontinued in October), and in 20 minutes Claude Code + HF Cli recommended the best replacement small model", "context": "", "use_case": "production image processing", "switched_from": "", "story": "Qwen 3.8 Omni Flash"}]}, {"text": " which forces migrations and, for some, higher prices.", "cites": [{"id": 784011089, "item_id": 49000693, "url": "https://news.ycombinator.com/item?id=49000693", "by": "do_anh_tu", "date": "2026-07-21", "stance": "recommend", "quote": "please please don't kill the old ones like gemini-2.5-flash-lite, because that all I ever need for my little apps", "excerpt": "Man I love Gemini models but these kind of pricing increase is just insanse. I have a little product and I have to keep increasing the price and reduce the limits because of this non-sense, and they did not even let us use the old models in near future, so I forced to update to the new model with basically no to little improvement because I don't even need that much. Google if you can read this, it okay to release ne", "context": "", "use_case": "small apps", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 783909088, "item_id": 48994318, "url": "https://news.ycombinator.com/item?id=48994318", "by": "zuzululu", "date": "2026-07-21", "stance": "warn", "quote": "We spent thousands just to build this workflow now Google says screw off", "excerpt": "same I just switched to OpenAI after using flash 2.5 lite for almost everything at our company. We spent thousands just to build this workflow now Google says screw off", "context": "company using it for almost everything", "use_case": "company LLM workflows", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}]}], "praise": [{"label": "Production workloads", "summary": "Used in production for classification, image processing, summarization and as a primary German-language model.", "quotes": [{"id": 788961665, "item_id": 49310104, "url": "https://news.ycombinator.com/item?id=49310104", "by": "scirob", "date": "2026-08-15", "stance": "recommend", "quote": "in production we have been using Gemini Flash Lite as primary", "excerpt": "We maintain German Langauge index as no one publishes or reruns these sepeartly. Qwen 3.8 27B is a small improvement with some regressions in our benchmarks not a huge jump like benchmarks listed. https://dach.peerbench.ai/compare?models=qwen%2Fqwen3.8-27b,qwen%2Fqwen3.6-27b German language has never been a big focus for asian models but they still outperform Gemma models https://dach.peerbench.ai/compare?models=open", "context": "runs German-language production", "use_case": "production primary model", "switched_from": "", "story": "Qwen 3.8 27B is out: open weights, best local dense model yet"}, {"id": 783905409, "item_id": 48994088, "url": "https://news.ycombinator.com/item?id=48994088", "by": "AussieWog93", "date": "2026-07-21", "stance": "recommend", "quote": "3.5 Flash Lite sounds like it could be a strict upgrade for our use case", "excerpt": "A lot of disappointment here in the comments, but models like these aren't meant to compete with the likes of Fable or GPT 5.6. I use 3.1 Flash Lite regularly to classify listings on eCommerce websites. It's great for this task - fast, cheap and accurate. In fact, it was the single best model we tried in terms of the speed vs accuracy vs price tradeoffs - including the Chinese models. Of course, 3.5 Flash was more ac", "context": "", "use_case": "classifying eCommerce listings", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 796076371, "item_id": 49754773, "url": "https://news.ycombinator.com/item?id=49754773", "by": "testycool", "date": "2026-09-18", "stance": "neutral", "quote": "I had a production image processing using Gemini 2.5 Flash Lite", "excerpt": "I use models.dev's CLI tool, which I think gets data from OpenRouter, and ArtificialAnalysis so your coding agent can help you narrow it down. <sidenote> Similarly, HuggingFace has a CLI + a few skills, and they are very useful. I had a production image processing using Gemini 2.5 Flash Lite (which is getting discontinued in October), and in 20 minutes Claude Code + HF Cli recommended the best replacement small model", "context": "", "use_case": "production image processing", "switched_from": "", "story": "Qwen 3.8 Omni Flash"}]}, {"label": "Good value", "summary": "Described as surprisingly good value with a strong speed, accuracy and price tradeoff.", "quotes": [{"id": 779144514, "item_id": 48696532, "url": "https://news.ycombinator.com/item?id=48696532", "by": "sfifs", "date": "2026-06-27", "stance": "recommend", "quote": "Gemini Flash Lite 3.1 is however  surprisingly good value.", "excerpt": "So a lot depends on your specific use case but mid-sized open weight models are pretty actually good now, so this is realistic [1] The first question to ask is does your use case require handling personal or sensitive data. If you're using the LLM for OpenClaw or you want to handle sensitive or medical data, a local model generally is necessary. if it's not so sensitive - Cloud providers with some sort of user agreem", "context": "", "use_case": "", "switched_from": "", "story": "Ask HN: MacBook vs. Dedicated GPU for LLM"}, {"id": 783905409, "item_id": 48994088, "url": "https://news.ycombinator.com/item?id=48994088", "by": "AussieWog93", "date": "2026-07-21", "stance": "recommend", "quote": "3.5 Flash Lite sounds like it could be a strict upgrade for our use case", "excerpt": "A lot of disappointment here in the comments, but models like these aren't meant to compete with the likes of Fable or GPT 5.6. I use 3.1 Flash Lite regularly to classify listings on eCommerce websites. It's great for this task - fast, cheap and accurate. In fact, it was the single best model we tried in terms of the speed vs accuracy vs price tradeoffs - including the Chinese models. Of course, 3.5 Flash was more ac", "context": "", "use_case": "classifying eCommerce listings", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}]}, {"label": "Capable enough", "summary": "Commenters say it handles a lot and is all they need for small apps.", "quotes": [{"id": 776320390, "item_id": 48520024, "url": "https://news.ycombinator.com/item?id=48520024", "by": "bachmeier", "date": "2026-06-13", "stance": "recommend", "quote": "You can do a lot of things with Gemini Flash Lite 3.1.", "excerpt": "> The upfront cost is steep and the models you can actually run at home are weaker than what the frontier labs ship, so this only pays off if you can keep the rig busy with long running tasks where a slower, cheaper model grinds away overnight. Most people can\u2019t keep a home machine that loaded, and the hardware you buy today may look like a bad bet in a year. Oh, so this is not a post about AI coding at home. It's ab", "context": "", "use_case": "cheap AI coding", "switched_from": "", "story": "AI Coding at Home Without Going Broke"}, {"id": 784011089, "item_id": 49000693, "url": "https://news.ycombinator.com/item?id=49000693", "by": "do_anh_tu", "date": "2026-07-21", "stance": "recommend", "quote": "please please don't kill the old ones like gemini-2.5-flash-lite, because that all I ever need for my little apps", "excerpt": "Man I love Gemini models but these kind of pricing increase is just insanse. I have a little product and I have to keep increasing the price and reduce the limits because of this non-sense, and they did not even let us use the old models in near future, so I forced to update to the new model with basically no to little improvement because I don't even need that much. Google if you can read this, it okay to release ne", "context": "", "use_case": "small apps", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}]}, {"label": "Classification", "summary": "Called fast, cheap and accurate for classifying eCommerce listings.", "quotes": [{"id": 783905409, "item_id": 48994088, "url": "https://news.ycombinator.com/item?id=48994088", "by": "AussieWog93", "date": "2026-07-21", "stance": "recommend", "quote": "3.5 Flash Lite sounds like it could be a strict upgrade for our use case", "excerpt": "A lot of disappointment here in the comments, but models like these aren't meant to compete with the likes of Fable or GPT 5.6. I use 3.1 Flash Lite regularly to classify listings on eCommerce websites. It's great for this task - fast, cheap and accurate. In fact, it was the single best model we tried in terms of the speed vs accuracy vs price tradeoffs - including the Chinese models. Of course, 3.5 Flash was more ac", "context": "", "use_case": "classifying eCommerce listings", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 796255953, "item_id": 49765997, "url": "https://news.ycombinator.com/item?id=49765997", "by": "Oras", "date": "2026-09-19", "stance": "neutral", "quote": "classification tasks that I used Gemini 2.5 flash lite with", "excerpt": "I played around with Jev last night and did it for classification tasks that I used Gemini 2.5 flash lite with. It\u2019s a bit faster and bit cheaper, but this is compared to LLM. The consistency was nice to see, BUT, as someone who trained NLP models prior to LLMs, it\u2019s just BERT with more data. I can see why people would want ready made one shot classifier, and I can see the value of sending multiple classifier in one ", "context": "", "use_case": "classification", "switched_from": "", "story": "I built non-autoregressive decision models with RL a year ago"}]}], "caveats": [{"label": "Model deprecation", "summary": "Older versions like 2.5 Flash Lite are being discontinued, forcing users to migrate or switch providers.", "quotes": [{"id": 784011089, "item_id": 49000693, "url": "https://news.ycombinator.com/item?id=49000693", "by": "do_anh_tu", "date": "2026-07-21", "stance": "recommend", "quote": "please please don't kill the old ones like gemini-2.5-flash-lite, because that all I ever need for my little apps", "excerpt": "Man I love Gemini models but these kind of pricing increase is just insanse. I have a little product and I have to keep increasing the price and reduce the limits because of this non-sense, and they did not even let us use the old models in near future, so I forced to update to the new model with basically no to little improvement because I don't even need that much. Google if you can read this, it okay to release ne", "context": "", "use_case": "small apps", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 783909088, "item_id": 48994318, "url": "https://news.ycombinator.com/item?id=48994318", "by": "zuzululu", "date": "2026-07-21", "stance": "warn", "quote": "We spent thousands just to build this workflow now Google says screw off", "excerpt": "same I just switched to OpenAI after using flash 2.5 lite for almost everything at our company. We spent thousands just to build this workflow now Google says screw off", "context": "company using it for almost everything", "use_case": "company LLM workflows", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 796076371, "item_id": 49754773, "url": "https://news.ycombinator.com/item?id=49754773", "by": "testycool", "date": "2026-09-18", "stance": "neutral", "quote": "I had a production image processing using Gemini 2.5 Flash Lite", "excerpt": "I use models.dev's CLI tool, which I think gets data from OpenRouter, and ArtificialAnalysis so your coding agent can help you narrow it down. <sidenote> Similarly, HuggingFace has a CLI + a few skills, and they are very useful. I had a production image processing using Gemini 2.5 Flash Lite (which is getting discontinued in October), and in 20 minutes Claude Code + HF Cli recommended the best replacement small model", "context": "", "use_case": "production image processing", "switched_from": "", "story": "Qwen 3.8 Omni Flash"}]}, {"label": "Price increases", "summary": "One developer says pricing increases forced them to raise prices and cut limits.", "quotes": [{"id": 784011089, "item_id": 49000693, "url": "https://news.ycombinator.com/item?id=49000693", "by": "do_anh_tu", "date": "2026-07-21", "stance": "recommend", "quote": "please please don't kill the old ones like gemini-2.5-flash-lite, because that all I ever need for my little apps", "excerpt": "Man I love Gemini models but these kind of pricing increase is just insanse. I have a little product and I have to keep increasing the price and reduce the limits because of this non-sense, and they did not even let us use the old models in near future, so I forced to update to the new model with basically no to little improvement because I don't even need that much. Google if you can read this, it okay to release ne", "context": "", "use_case": "small apps", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}]}, {"label": "Free quota removed", "summary": "A commenter shut down a public instance after the generous free quota was removed.", "quotes": [{"id": 768645872, "item_id": 48040367, "url": "https://news.ycombinator.com/item?id=48040367", "by": "lxgr", "date": "2026-05-06", "stance": "neutral", "quote": "that's what I used for by LLM-backed HTTP server", "excerpt": "Nice, that's what I used for by LLM-backed HTTP server [1] a while ago as well :) It's a shame they got rid of the generous free quota a while ago, which is why I had to shut my public instance down. [1] https://github.com/lxgr/vibeserver/", "context": "", "use_case": "LLM-backed HTTP server", "switched_from": "", "story": "Show HN: Hallucinopedia"}]}, {"label": "Slow releases", "summary": "One commenter noted the Flash Lite line lagged behind schedule at version 2.5.", "quotes": [{"id": 753218128, "item_id": 47076133, "url": "https://news.ycombinator.com/item?id=47076133", "by": "seizethecheese", "date": "2026-02-19", "stance": "neutral", "quote": "I use Gemini flash lite in a side project, and it\u2019s stuck on 2.5.", "excerpt": "I use Gemini flash lite in a side project, and it\u2019s stuck on 2.5. It\u2019s now well behind schedule. Any speculation as to what\u2019s going on?", "context": "", "use_case": "side project", "switched_from": "", "story": "Gemini 3.1 Pro"}]}], "who": [{"by": "scirob", "context": "runs German-language production", "url": "https://news.ycombinator.com/item?id=49310104", "date": "2026-08-15"}], "tags": ["for classification", "for side projects", "for production pipelines", "for German-language tasks"], "use_cases": ["agentic video analysis", "phone app model", "classification", "production image processing", "production primary model", "summarizing bulletins"], "compared": [{"slug": "jev", "name": "Jev", "co": 2, "switched_from": 0, "switched_to": 1}, {"slug": "gemini-flash", "name": "Gemini Flash", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "gemma", "name": "Gemma", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "openrouter", "name": "OpenRouter", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "gemini", "name": "Gemini", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=48993414", "title": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=50295027", "title": "Different Gemini model setups for agentic video analysis: Deep vs. Lite", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49913571", "title": "Gemini 4 Argon", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49765348", "title": "I built non-autoregressive decision models with RL a year ago", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49747925", "title": "Qwen 3.8 Omni Flash", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49299605", "title": "Qwen 3.8 27B is out: open weights, best local dense model yet", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {}, "2026-02": {"neutral": 1}, "2026-03": {}, "2026-04": {}, "2026-05": {"neutral": 1}, "2026-06": {"neutral": 2, "recommend": 2}, "2026-07": {"recommend": 2, "warn": 1}, "2026-08": {"neutral": 1, "recommend": 1}, "2026-09": {"neutral": 3}, "2026-10": {"compare": 1}}, "quotes": [{"id": 804720449, "item_id": 50295028, "url": "https://news.ycombinator.com/item?id=50295028", "by": "2upmedia", "date": "2026-10-29", "stance": "compare", "quote": "a full analysis of Gemini 3.8 Flash vs Gemini 3.5 Flash lite using agentic video understanding", "excerpt": "For some time I've been interested in analyzing videos. Here's a full analysis of Gemini 3.8 Flash vs Gemini 3.5 Flash lite using agentic video understanding and how you could take advantage of both. I also created an open source Claude Code and Codex plugin you could use https://github.com/jorgecolonconsulting/clipwitness.", "context": "", "use_case": "agentic video analysis", "switched_from": "", "story": "Different Gemini model setups for agentic video analysis: Deep vs. Lite"}, {"id": 798636402, "item_id": 49914775, "url": "https://news.ycombinator.com/item?id=49914775", "by": "p_l", "date": "2026-09-30", "stance": "neutral", "quote": "mine has 3.1 Pro, 3.5 Flash-Lite, and 3.8 Flash (with optional extra effort)", "excerpt": "Not sure what's the problem with phone app rollout, mine has 3.1 Pro, 3.5 Flash-Lite, and 3.8 Flash (with optional extra effort), and pretty much it switched to latest Flash series as main option since 3.5 just after release", "context": "", "use_case": "phone app model", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 796255953, "item_id": 49765997, "url": "https://news.ycombinator.com/item?id=49765997", "by": "Oras", "date": "2026-09-19", "stance": "neutral", "quote": "classification tasks that I used Gemini 2.5 flash lite with", "excerpt": "I played around with Jev last night and did it for classification tasks that I used Gemini 2.5 flash lite with. It\u2019s a bit faster and bit cheaper, but this is compared to LLM. The consistency was nice to see, BUT, as someone who trained NLP models prior to LLMs, it\u2019s just BERT with more data. I can see why people would want ready made one shot classifier, and I can see the value of sending multiple classifier in one ", "context": "", "use_case": "classification", "switched_from": "", "story": "I built non-autoregressive decision models with RL a year ago"}, {"id": 796076371, "item_id": 49754773, "url": "https://news.ycombinator.com/item?id=49754773", "by": "testycool", "date": "2026-09-18", "stance": "neutral", "quote": "I had a production image processing using Gemini 2.5 Flash Lite", "excerpt": "I use models.dev's CLI tool, which I think gets data from OpenRouter, and ArtificialAnalysis so your coding agent can help you narrow it down. <sidenote> Similarly, HuggingFace has a CLI + a few skills, and they are very useful. I had a production image processing using Gemini 2.5 Flash Lite (which is getting discontinued in October), and in 20 minutes Claude Code + HF Cli recommended the best replacement small model", "context": "", "use_case": "production image processing", "switched_from": "", "story": "Qwen 3.8 Omni Flash"}, {"id": 788961665, "item_id": 49310104, "url": "https://news.ycombinator.com/item?id=49310104", "by": "scirob", "date": "2026-08-15", "stance": "recommend", "quote": "in production we have been using Gemini Flash Lite as primary", "excerpt": "We maintain German Langauge index as no one publishes or reruns these sepeartly. Qwen 3.8 27B is a small improvement with some regressions in our benchmarks not a huge jump like benchmarks listed. https://dach.peerbench.ai/compare?models=qwen%2Fqwen3.8-27b,qwen%2Fqwen3.6-27b German language has never been a big focus for asian models but they still outperform Gemma models https://dach.peerbench.ai/compare?models=open", "context": "runs German-language production", "use_case": "production primary model", "switched_from": "", "story": "Qwen 3.8 27B is out: open weights, best local dense model yet"}, {"id": 787808866, "item_id": 49238054, "url": "https://news.ycombinator.com/item?id=49238054", "by": "schultetwin1", "date": "2026-08-09", "stance": "neutral", "quote": "I use Gemini Flash Lite to summarize and filter the bulletins.", "excerpt": "https://dipstickalerts.com - I've been working on this during my kid's naps. The web app allows car owners in the U.S. to see a list of all the service bulletins the manufacturer has sent out about your car over the years (called Technical Service Bulletins or TSBs). You can subscribe for new bulletins to be sent to your email as well. These service bulletins are not only recalls but also fixes for known issues with ", "context": "", "use_case": "summarizing bulletins", "switched_from": "", "story": "Ask HN: What are you working on? (August 2026)"}, {"id": 784011089, "item_id": 49000693, "url": "https://news.ycombinator.com/item?id=49000693", "by": "do_anh_tu", "date": "2026-07-21", "stance": "recommend", "quote": "please please don't kill the old ones like gemini-2.5-flash-lite, because that all I ever need for my little apps", "excerpt": "Man I love Gemini models but these kind of pricing increase is just insanse. I have a little product and I have to keep increasing the price and reduce the limits because of this non-sense, and they did not even let us use the old models in near future, so I forced to update to the new model with basically no to little improvement because I don't even need that much. Google if you can read this, it okay to release ne", "context": "", "use_case": "small apps", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 783909088, "item_id": 48994318, "url": "https://news.ycombinator.com/item?id=48994318", "by": "zuzululu", "date": "2026-07-21", "stance": "warn", "quote": "We spent thousands just to build this workflow now Google says screw off", "excerpt": "same I just switched to OpenAI after using flash 2.5 lite for almost everything at our company. We spent thousands just to build this workflow now Google says screw off", "context": "company using it for almost everything", "use_case": "company LLM workflows", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 783905409, "item_id": 48994088, "url": "https://news.ycombinator.com/item?id=48994088", "by": "AussieWog93", "date": "2026-07-21", "stance": "recommend", "quote": "3.5 Flash Lite sounds like it could be a strict upgrade for our use case", "excerpt": "A lot of disappointment here in the comments, but models like these aren't meant to compete with the likes of Fable or GPT 5.6. I use 3.1 Flash Lite regularly to classify listings on eCommerce websites. It's great for this task - fast, cheap and accurate. In fact, it was the single best model we tried in terms of the speed vs accuracy vs price tradeoffs - including the Chinese models. Of course, 3.5 Flash was more ac", "context": "", "use_case": "classifying eCommerce listings", "switched_from": "", "story": "Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber"}, {"id": 779144514, "item_id": 48696532, "url": "https://news.ycombinator.com/item?id=48696532", "by": "sfifs", "date": "2026-06-27", "stance": "recommend", "quote": "Gemini Flash Lite 3.1 is however  surprisingly good value.", "excerpt": "So a lot depends on your specific use case but mid-sized open weight models are pretty actually good now, so this is realistic [1] The first question to ask is does your use case require handling personal or sensitive data. If you're using the LLM for OpenClaw or you want to handle sensitive or medical data, a local model generally is necessary. if it's not so sensitive - Cloud providers with some sort of user agreem", "context": "", "use_case": "", "switched_from": "", "story": "Ask HN: MacBook vs. Dedicated GPU for LLM"}, {"id": 776320390, "item_id": 48520024, "url": "https://news.ycombinator.com/item?id=48520024", "by": "bachmeier", "date": "2026-06-13", "stance": "recommend", "quote": "You can do a lot of things with Gemini Flash Lite 3.1.", "excerpt": "> The upfront cost is steep and the models you can actually run at home are weaker than what the frontier labs ship, so this only pays off if you can keep the rig busy with long running tasks where a slower, cheaper model grinds away overnight. Most people can\u2019t keep a home machine that loaded, and the hardware you buy today may look like a bad bet in a year. Oh, so this is not a post about AI coding at home. It's ab", "context": "", "use_case": "cheap AI coding", "switched_from": "", "story": "AI Coding at Home Without Going Broke"}, {"id": 776153489, "item_id": 48509593, "url": "https://news.ycombinator.com/item?id=48509593", "by": "dennis16384", "date": "2026-06-12", "stance": "neutral", "quote": "It's mostly tested on small analysis and question answering with Gemini Flash Lite", "excerpt": "Nice, I recently did a similar but much simpler thing and open-sourced it under MIT, maybe some bits and pieced will be useful https://github.com/eatmydata-org/eatmydata For example, MIT-licensed sqlite vector search extension. Overall, I have a orchestrator - sql coder - js coder - dashboards, all without backend, running locally in the browser. It's mostly tested on small analysis and question answering with Gemini", "context": "", "use_case": "question answering", "switched_from": "", "story": "Launch HN: BitBoard (YC P25) \u2013 Analytics Workspace for Agents"}, {"id": 775669346, "item_id": 48479334, "url": "https://news.ycombinator.com/item?id=48479334", "by": "vineyardmike", "date": "2026-06-10", "stance": "neutral", "quote": "It made me even use the tiny models like Gemini Flash Lite and GPT Mini/Nano more too.", "excerpt": "Recently I had switched to OpenCode to try out many of the Non-US-Frontier-Labs models. My unexpected favorite model to use was Mercury (a diffusion model). Not because it was \u201csmart\u201d but because it was stupid fast. It was more of a pair-programming experience instead of the SOTA agentic experience of prompting and waiting. Honestly, it was also way more fun and brought back some of the pre-AI coding experience while", "context": "", "use_case": "fast coding assistance", "switched_from": "", "story": "DiffusionGemma: 4x Faster Text Generation"}, {"id": 768645872, "item_id": 48040367, "url": "https://news.ycombinator.com/item?id=48040367", "by": "lxgr", "date": "2026-05-06", "stance": "neutral", "quote": "that's what I used for by LLM-backed HTTP server", "excerpt": "Nice, that's what I used for by LLM-backed HTTP server [1] a while ago as well :) It's a shame they got rid of the generous free quota a while ago, which is why I had to shut my public instance down. [1] https://github.com/lxgr/vibeserver/", "context": "", "use_case": "LLM-backed HTTP server", "switched_from": "", "story": "Show HN: Hallucinopedia"}, {"id": 753218128, "item_id": 47076133, "url": "https://news.ycombinator.com/item?id=47076133", "by": "seizethecheese", "date": "2026-02-19", "stance": "neutral", "quote": "I use Gemini flash lite in a side project, and it\u2019s stuck on 2.5.", "excerpt": "I use Gemini flash lite in a side project, and it\u2019s stuck on 2.5. It\u2019s now well behind schedule. Any speculation as to what\u2019s going on?", "context": "", "use_case": "side project", "switched_from": "", "story": "Gemini 3.1 Pro"}]}