# NumPy 86 Hacker News commenters recommended NumPy and 32 criticized it, November 2025 to November 2026. It was recommended in 12 of 13 months. It meets the Hacker Approved badge rule. **Verdict:** People recommend it as the standard fast array math library for Python; critics note overhead on small data and a limited interface. ## Why people recommend it - Fast, backed by compiled C code (4 comments): "Numpy is wrapping around C data structures that the loading process loads into" ([source](https://news.ycombinator.com/item?id=46832501)) "If you need speed and efficiency on your local machine, you use NumPy outright" ([source](https://news.ycombinator.com/item?id=46124552)) "calling out to external code that isn't written in Python (e.g., NumPy, etc)" ([source](https://news.ycombinator.com/item?id=45837496)) - Foundation for data science and ML (3 comments): "libraries, and specifically NumPy, are the reason Python is still in the picture today" ([source](https://news.ycombinator.com/item?id=47419246)) "Without NumPy and Pandas, ML in Python would never have been a thing." ([source](https://news.ycombinator.com/item?id=46797521)) "numpy is the the de-facto foundation for data science in python" ([source](https://news.ycombinator.com/item?id=46400201)) - Strong ecosystem of scientific packages (3 comments): "Python's combination of readable syntax and powerful libraries (NumPy, SymPy, Matplotlib)" ([source](https://news.ycombinator.com/item?id=46383455)) "Numpy is a really great library making all of those scientific packages" ([source](https://news.ycombinator.com/item?id=45902667)) "extremely easy to use, fast and extensive in the whole ecosystem" ([source](https://news.ycombinator.com/item?id=46055802)) - Readable, close to whiteboard math (3 comments): "writing matrix maths in numpy is much easier thanks to not having to make this arbitrary distinction" ([source](https://news.ycombinator.com/item?id=46285200)) "For linear algebra and ML with Jax/Numpy I find it quite readable." ([source](https://news.ycombinator.com/item?id=49376137)) "Numpy is a bit more verbose but still 1-to-1 with the whiteboard." ([source](https://news.ycombinator.com/item?id=46280881)) ## What people warn about - Limited interface and fewer language features (4 comments): "not having zero overhead structs that one can in an array handicaps types of performance codes one can write" ([source](https://news.ycombinator.com/item?id=47178638)) "While Numpy has many similar operations, it lacks the terseness, concepts like trains and forks etc." ([source](https://news.ycombinator.com/item?id=45951748)) "numpy has the same problem as blas basically, right? The limited interface." ([source](https://news.ycombinator.com/item?id=47383824)) - Unopinionated, awkward ergonomics (3 comments): "numpy tries to be this highly unopinionated tool that can do everything, and ends up annoying to use anywhere" ([source](https://news.ycombinator.com/item?id=46281225)) "it's a shame to have to bring in a bazooka to get decent ergonomics and performances" ([source](https://news.ycombinator.com/item?id=46265603)) "it's much less convenient and less self explanatory than the apl system" ([source](https://news.ycombinator.com/item?id=48140290)) - Overhead, can be slower than Python (2 comments): "NumPy does make the mechanism clearer, but aside from learning purposes, it might be a bit slow." ([source](https://news.ycombinator.com/item?id=49872888)) "NumPy, PyTorch, JAX have non-trivial overhead, and might even be slower than vanilla Python" ([source](https://news.ycombinator.com/item?id=47387367)) - Heavy dependencies, overkill for simple tasks (2 comments): "adding Pandas and NumPy which have numerous (really: squillions of) dependencies outside the foundation library" ([source](https://news.ycombinator.com/item?id=46052593)) "NumPy solves this but is heavyweight and often overkill for 1D data." ([source](https://news.ycombinator.com/item?id=46456459)) ## Alternatives hackers recommend - [Polars](https://hackerapproved.com/p/polars.md): 152 recommend - [pandas](https://hackerapproved.com/p/pandas.md): 93 recommend - [SciPy](https://hackerapproved.com/p/scipy.md): 12 recommend - [sh Python library](https://hackerapproved.com/p/sh.md): 10 recommend - [Tacopy](https://hackerapproved.com/p/tacopy.md): 8 recommend ## Cite as Hacker Approved, "NumPy", https://hackerapproved.com/p/numpy/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).