# Google Gemini 35 Hacker News commenters recommended Google Gemini and 33 criticized it, November 2025 to November 2026. It was recommended in 11 of 13 months. **Verdict:** Recommended for strong general knowledge, image and video understanding, multilingual text and cheap fast models. Caveat: many say they are weak at coding and do poorly with agent harnesses. ## Why people recommend it - Strong general knowledge and intelligence (4 comments): "Gemini models still have the best intelligence (when asked any questions, most likely to get it right), but in production they still have many failure modes" ([source](https://news.ycombinator.com/item?id=48313290)) "Gemini models have the best all-around knowledge, they are way above other models in general knowledge" ([source](https://news.ycombinator.com/item?id=49902687)) "Gemini models are still good for knowledge as per omniscience benchmarks on artificial analysis" ([source](https://news.ycombinator.com/item?id=49315049)) - Strong spatial reasoning (3 comments): "where really the only ones fullt grasping spatial reasoning even compared to opus" ([source](https://news.ycombinator.com/item?id=49292844)) "Spatial understanding seems to be the best for the price and speed." ([source](https://news.ycombinator.com/item?id=49692903)) "Gemini models are a little bit better about spatial reasoning" ([source](https://news.ycombinator.com/item?id=46659190)) - Strong vision and video understanding (4 comments): "The frontier model can extremely "ask" a cheap Gemini vision model to give it data from datasheets or specs." ([source](https://news.ycombinator.com/item?id=49574053)) "Gemini models can process very long videos and provide description with timestamps if asked for." ([source](https://news.ycombinator.com/item?id=46126050)) "Gemini models are the least jagged in my experience and have the best image understanding" ([source](https://news.ycombinator.com/item?id=48236044)) - Cheap, fast and good value (4 comments): "Gemini models are actually fine for me as an amateur+ I found especially since they have regional pricing" ([source](https://news.ycombinator.com/item?id=50256647)) "it's performed very well, even the cheaper Gemini models" ([source](https://news.ycombinator.com/item?id=49490208)) "Their new models are a pretty good value and crazy fast" ([source](https://news.ycombinator.com/item?id=49518061)) ## What people warn about - Weak at coding (4 comments): "I've found that Gemini models often produce pseudocode that seems good at first glance but is typically wrong or incomplete" ([source](https://news.ycombinator.com/item?id=46889729)) "The gemini models are basically unusable for coding right now, they seem to have gone backwards in capability." ([source](https://news.ycombinator.com/item?id=48463553)) "fail to call tools or explore correctly which limits their usefulness for agentic coding" ([source](https://news.ycombinator.com/item?id=48632801)) - Inaccurate or unreliable in production (4 comments): "Gemini models still have the best intelligence (when asked any questions, most likely to get it right), but in production they still have many failure modes" ([source](https://news.ycombinator.com/item?id=48313290)) "they still felt very late 2024 in terms of hallucination rate" ([source](https://news.ycombinator.com/item?id=47076248)) "Generally, they are always less precise and less practically useful than benchmarks suggest." ([source](https://news.ycombinator.com/item?id=49919810)) - Poor agent harness and tool use (3 comments): "Gemini models are really smart, unfortunately they don't play well with any harness" ([source](https://news.ycombinator.com/item?id=49263805)) "Trying to use Gemini models "locally" resulted in a similar "short shrift" of effort resulting in mistakes and lots of turns." ([source](https://news.ycombinator.com/item?id=48504789)) "fail to call tools or explore correctly which limits their usefulness for agentic coding" ([source](https://news.ycombinator.com/item?id=48632801)) - Struggles with long complex tasks (2 comments): "they couldn't reliably identify specific scenes in a 50K word novel unless I trimmed down the context" ([source](https://news.ycombinator.com/item?id=46902944)) "they simply couldn't do the long complex tasks I'd being doing with Anthropic models" ([source](https://news.ycombinator.com/item?id=49539263)) ## Alternatives hackers recommend - [Claude Opus](https://hackerapproved.com/p/claude-opus.md): 3331 recommend - [OpenAI GPT](https://hackerapproved.com/p/gpt-5-5.md): 1205 recommend - [Claude Sonnet](https://hackerapproved.com/p/claude-sonnet.md): 854 recommend - [DeepSeek V4 Flash](https://hackerapproved.com/p/deepseek-v4-flash.md): 828 recommend - [Fable](https://hackerapproved.com/p/fable.md): 823 recommend ## Where to get it - [Visit the official site](https://gemini.google.com/) ## Cite as Hacker Approved, "Google Gemini", https://hackerapproved.com/p/gemini-models/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).