Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Training a 4B model to produce 81% faster query plans than Postgres(rohanbansal.com ↗)
    55comments
  2. Breaking the 1.58-bit Barrier for Ternary LLMs(arxiv.org ↗)
    5comments
  3. Xiaomi Mimo 2.6 live post-training dashboard(xiaomi.com ↗)
    44comments
  4. Nvidia announces native GPU programming in Rust(nvidia.com ↗)
    8comments
  5. Small programming tricks(will-keleher.com ↗)
    166comments
  6. AWS says it can't restore some data from mideast facilities struck by Iran(wsj.com ↗)
    102comments
  7. Reversing Factorio's RNG(gegell.github.io ↗)
    10comments
  8. Performance Improvements in .NET 11(devblogs.microsoft.com/dotnet ↗)
    8comments
  9. Backups Aren't Simple(filipovski.net ↗)
    discuss
  10. Accurate Models of AMD Matrix Cores(arxiv.org ↗)
    5comments
  11. The engineering behind the US Strategic Petroleum Reserve(johnjwang.com ↗)
    discuss
  12. Japan's book scene is moving from bookstores to libraries(untranslatedjp.substack.com ↗)
    16comments
  13. How good are frontier models at physics?(arxiv.org ↗)
    18comments
  14. Dream-RSI: Recursive Self-Improvement through Evolving Worlds(arxiv.org ↗)
    49comments
  15. Anatomy of a Texture(agentlien.github.io ↗)
    10comments
  16. Mistral X Mozilla: Private, Multilingual AI Browsing(mistral.ai ↗)
    182comments
  17. WalShadow: Sub-second Postgres replication to ClickHouse from physical WAL(clickhouse.com ↗)
    5comments
  18. Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations(github.com/arnegiacomo ↗)
    236comments
  19. Vectorized and performance-portable Quicksort (2022)(googleblog.com ↗)
    24comments
  20. Anecdotally, programmers dislike "reduce"(evanhahn.com ↗)
    107comments
  21. Why Does the Universe Expand?(cosmicave.org ↗)
    1comments
  22. I replaced my brown-noise browser tab with a menu bar app(oldmanrahul.com ↗)
    discuss
  23. Reverse-engineered Jev-like model(github.com/vinnylarouge ↗)
    5comments
  24. Hackers Got Inside a Flock Camera(wired.com ↗)
    200comments
  25. Training Text-to-Image Models 3.6× Faster(linum.ai ↗)
    1comments
  26. macOS 27 Golden Gate – Review(arstechnica.com ↗)
    93comments
  27. The Siberian Ice Maiden and the Scythian World(patrickwyman.substack.com ↗)
    4comments
  28. Kyber (YC W23) Is Hiring a Forward Deployed Engineer(ycombinator.com ↗)
    discuss
  29. The DeepMind Institute(deepmind.com ↗)
    39comments
  30. Tell the speakers that you liked their talks(ohhelloana.blog ↗)
    74comments

Nvidia announces native GPU programming in Rust

51 pointsby 11h agodeveloper.nvidia.com
8 comments
30m agoHN ↗

I'm looking forward to trying these when they stabilize! I currently use WGPU for graphics, and cudarc for CUDA.

Note: Cuda-oxide is similar to Cudarc's host component, but uses a rust-style kernel dialect. Advantage: Share structs between host and device. Disadvantage: Trading standard Cuda kernels for a new, WIP dialect.

I haven't tried the tile API yet; looking forward to it.

The last time I checked, Cuda Oxide was Linux only, and required Async; these are why I haven't tried it yet.

15m agoHN ↗

cudarc been great for me, because it's easy to look up existing examples and references, and it maps 1-to-1 with what I see. I'm already having a tough time with CUDA itself, a dialect of it makes a tad harder to rely on previous work.

Seems more ergonomic in general though, both approaches they share, compared to cudarc, and less build infrastructure and fiddling with environments, which is great.

23m agoHN ↗

The launch is checked rather than trusted.

Damn even Nvidia is putting out fully Claude-written articles.

20m agoHN ↗

First of all, this is a pre-1.0 release that requires a nightly Rust compiler (if you choose the SIMT track with cuda-oxide) so that one is going to be unstable software.

Secondly, When an issue occurs with a kernel or you want to write your own custom kernel in Rust, now we need to diagnose if the problem came from either cuda-oxide (SIMT), Rust's side, CUDA or Tile (If you decide to choose the Tile track).

Another dependency into the list and course everything is open source except CUDA itself. So any issue that happens on the CUDA level, you are forced to wait for them to fix it.

7m agoHN ↗

In this age of LLM written everything which has softly killed my motivation for learning Rust somewhat, this has revived my interest if not only for the fact the LLMs haven't yet been trained on this yet!

1m agoHN ↗

LLM don't need to be trained in a library to use it well. It's just Rust which they know well.

4m agoHN ↗

what this article tells me is that no one at Nvidia actually cares about this project whatsoever. otherwise, they would have had a person actually write the announcement.