# DeepSeek V4 Pro 302 Hacker News commenters recommended DeepSeek V4 Pro and 118 criticized it, November 2025 to November 2026. It was recommended in 7 of 13 months; owners who said how long report a median of 0.06 years (16 owners). It meets the Hacker Approved badge rule. **Verdict:** A near-frontier coding model at a fraction of rivals' cost, helped by strong context caching, though it trails Opus at planning and can misread intent or make mistakes. ## Why people recommend it - Far cheaper than rival models (13 comments): "It's quality of work is not bad at all, it is fairly fast and given the fraction of the cost compared to Claude, I can keep going which makes it a very compelling option." ([source](https://news.ycombinator.com/item?id=48147588)) "Its just as capable, and makes the same kinds of dumb mistakes and the other two have been making since March. For the price, I'm more than happy with it." ([source](https://news.ycombinator.com/item?id=48441517)) "the model is REALLY good. I try Claude at work and Deepseek personally and this is the only model that works without trying to actively bankcrypt me." ([source](https://news.ycombinator.com/item?id=48239523)) - Excellent context caching (6 comments): "DeepSeek has consistently been among the best performers and its extreme caching performance makes it cheaper than just about anything" ([source](https://news.ycombinator.com/item?id=48713609)) "it's costing the same or less than the typical coding subscription, due to the discount and the power of context caching" ([source](https://news.ycombinator.com/item?id=48259195)) "the cache-hit rate is extremely high, and the cache invalidation window is much longer than every other provider's" ([source](https://news.ycombinator.com/item?id=48044296)) - Strong implementation from detailed plans (5 comments): "solid at implementation when given a specification, but falls a good bit short of Opus 4.7 max thinking when planning out larger changes" ([source](https://news.ycombinator.com/item?id=47985027)) "I can't trust it to write it's own plans from a spec, but if I give it a detailed execution plan written by Opus, it's fast and cheap" ([source](https://news.ycombinator.com/item?id=49276406)) "Results for kimi and deepseek are pretty much indistinguishable from opus for coding if opus writes the plan" ([source](https://news.ycombinator.com/item?id=48509102)) - Close to frontier model quality (4 comments): "We have lots of open-weight models like DeepSeek V4 Pro that are very close to SOTA and we know the cost of running them." ([source](https://news.ycombinator.com/item?id=48279652)) "I really do feel like DeepSeek V4 Pro is often better than current Sonnet is, in the general case." ([source](https://news.ycombinator.com/item?id=48512324)) "DeepSeek v4 Pro is a far more capable model and still cheaper than anything that it competes with" ([source](https://news.ycombinator.com/item?id=49286589)) ## What people warn about - Weaker than Opus at planning (4 comments): "solid at implementation when given a specification, but falls a good bit short of Opus 4.7 max thinking when planning out larger changes" ([source](https://news.ycombinator.com/item?id=47985027)) "I can't trust it to write it's own plans from a spec, but if I give it a detailed execution plan written by Opus, it's fast and cheap" ([source](https://news.ycombinator.com/item?id=49276406)) "DeepSeek V4-Pro Max shines in competitive coding benchmarks. However, it trails both Opus models on software engineering." ([source](https://news.ycombinator.com/item?id=47887846)) - Mistakes and misunderstood intent (5 comments): "V4-pro in particular seems very capable, but will just dramatically completely misunderstand user intent" ([source](https://news.ycombinator.com/item?id=49309087)) "it would always start off promising, just to quickly get sidetracked, start hallucinating problems in the code" ([source](https://news.ycombinator.com/item?id=49625945)) "DeepSeek V4 Pro is really affordable per-token but regularly kept making mistakes in the tasks I gave it" ([source](https://news.ycombinator.com/item?id=48940626)) - Privacy and training-data concerns (2 comments): "Their privacy policy doesn't forbid them from just straight up publishing your raw prompts as training data." ([source](https://news.ycombinator.com/item?id=49276217)) "requires enabling "Allow paid endpoints that train on request data" in the OpenRouter privacy settings" ([source](https://news.ycombinator.com/item?id=49276060)) ## 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://www.deepseek.com/) ## Cite as Hacker Approved, "DeepSeek V4 Pro", https://hackerapproved.com/p/deepseek-v4-pro/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).