wonnx
bevy
Our great sponsors
wonnx | bevy | |
---|---|---|
18 | 573 | |
1,487 | 32,210 | |
6.8% | 3.8% | |
6.5 | 9.9 | |
26 days ago | 4 days ago | |
Rust | Rust | |
GNU General Public License v3.0 or later | MIT OR Apache-2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
wonnx
-
Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
The two I know of are IREE and Kompute[1]. I'm not sure how much momentum the latter has, I don't see it referenced much. There's also a growing body of work that uses Vulkan indirectly through WebGPU. This is currently lagging in performance due to lack of subgroups and cooperative matrix mult, but I see that gap closing. There I think wonnx[2] has the most momentum, but I am aware of other efforts.
[1]: https://kompute.cc/
[2]: https://github.com/webonnx/wonnx
-
VkFFT: Vulkan/CUDA/Hip/OpenCL/Level Zero/Metal Fast Fourier Transform Library
To a first approximation, Kompute[1] is that. It doesn't seem to be catching on, I'm seeing more buzz around WebGPU solutions, including wonnx[2] and more hand-rolled approaches, and IREE[3], the latter of which has a Vulkan back-end.
[1]: https://kompute.cc/
[2]: https://github.com/webonnx/wonnx
[3]: https://github.com/openxla/iree
-
Onnx Runtime: “Cross-Platform Accelerated Machine Learning”
There's also a third-party WebGPU implementation: https://github.com/webonnx/wonnx
-
Are there any ML crates that would compile to WASM?
By experimental I meant e.g. using WGPU to run compute shaders like wonnx, which is working fine but only on a very restricted set of devices and browsers.
- WebGPU ONNX inference runtime written in Rust
-
PyTorch Primitives in WebGPU for the Browser
https://news.ycombinator.com/item?id=35696031 ... TIL about wonnx: https://github.com/webonnx/wonnx#in-the-browser-using-webgpu...
microsoft/onnxruntime: https://github.com/microsoft/onnxruntime
Apache/arrow has language-portable Tensors for cpp: https://arrow.apache.org/docs/cpp/api/tensor.html and rust: https://docs.rs/arrow/latest/arrow/tensor/struct.Tensor.html and Python: https://arrow.apache.org/docs/python/api/tables.html#tensors https://arrow.apache.org/docs/python/generated/pyarrow.Tenso...
Fwiw it looks like the llama.cpp Tensor is from ggml, for which there are CUDA and OpenCL implementations (but not yet ROCm, or a WebGPU shim for use with emscripten transpilation to WASM): https://github.com/ggerganov/llama.cpp/blob/master/ggml.h
Are the recommendable ways to cast e.g. arrow Tensors to pytorch/tensorflow?
FWIU, Rust has a better compilation to WASM; and that's probably faster than already-compiled-to-JS/ES TensorFlow + WebGPU.
What's a fair benchmark?
-
rustformers/llm: Run inference for Large Language Models on CPU, with Rust 🦀🚀🦙
wonnx has done some fantastic work in this regard, so that's where we plan to start once we get there. In terms of general discussion of alternate backends, see this issue.
-
I want to talk about WebGPU
> GPU in other ways, such as training ML models and then using them via an inference engine all powered by your local GPU?
Have a look at wonnix https://github.com/webonnx/wonnx
A WebGPU-accelerated ONNX inference run-time written 100% in Rust, ready for native and the web
-
Chrome Ships WebGPU
Looking forward to your WebGPU ML runtime! Also, why not contribute back to WONNX? (https://github.com/webonnx/wonnx)
-
OpenXLA Is Available Now
You can indeed perform inference using WebGPU (see e.g. [1] for GPU-accelerated inference of ONNX models on WebGPU; I am one of the authors).
The point made above is that WebGPU can only be used for GPU's and not really for other types of 'neural accelerators' (like e.g. the ANE on Apple devices).
[1] https://github.com/webonnx/wonnx
bevy
-
Web Game Engines and Libraries
Missing one of the best choices as long as "maturity" isn't on the top of your list: Bevy - https://bevyengine.org/
Game engine written in Rust, leveraging ECS in almost every place and way, with a really capable WASM export option. Wrestling ECS for the first time might take you some time, but in my experience helps you keep game code as clean and decoupled as game code could be.
-
3D and 2D: Testing out my cross-platform graphics engine
I don't see WASM/WebGPU changing anything when it comes to gaming, as an industry, personally. 3d visualizations and interactive websites? Yeah definitely a nice improvement over WebGL 2, if years late.
WebGPU is pretty far behind what AAA games are using even as of 6 years ago. There's extra overhead and security in the WebGPU spec that AAA games do not want. Browsers do not lend themselves to downloading 300gb of assets.
Additionally, indie devs aren't using Steam for the technical capabilities. It's purely about marketshare. Video games are a highly saturated market. The users are all on Steam, getting their recommendations from Steam, and buying games in Steam sales. Hence all the indie developers publish to Steam. I don't see a web browser being appealing as a platform, because there's no way for developers to advertise to users.
That's also only indie games. AAA games use their own launchers, because they don't _need_ the discoverability from being on Steam. So they don't, and avoid the fees. If anything users _want_ the Steam monopoly, because they like the platform, and hate the walled garden launchers from AAA companies.
(I work on high end rendering features for the Bevy game engine https://bevyengine.org, and have extensive experience with WebGPU)
-
What Are Const Generics and How Are They Used in Rust?
I was working through an example in the repo for the Bevy game engine recently and came across this code
-
WebAssembly Playground
That's possible. I did spend quite a bit of time tinkering with compiler flags, and followed the recommendations.
Some notes I found just now seems to agree with my results, though: https://github.com/bevyengine/bevy/issues/3978#issuecomment-...
-
Immediate Mode GUI Programming
I cannot recommend immediate mode GUI programming based on the limitations I've experienced working with egui.
egui does not support putting two widgets in the center of the screen: https://github.com/emilk/egui/issues/3211
It's really easy to get started with immediate mode, it's really easy to bust out some UI, but the second you start trying to involve dynamically resized context and responsive layouts -- abandon all hope. The fact it has to calculate everything in a single pass makes these things hard/impossible.
... that said, I'm still using it for https://ant.care/ (https://github.com/MeoMix/symbiants) because it's the best thing I've found. I'm crossing my fingers that Bevy's UI story (or Kayak https://github.com/StarArawn/kayak_ui) become significantly more fleshed out sooner rather than later. Bevy 0.13 should have lots more in this area though (https://github.com/bevyengine/bevy/discussions/9538)
- A minimal working Rust / SDL2 / WASM browser game
-
ECS, Finally
I've also been enjoying building My First Game™ in Bevy using ECS. The community around Bevy really shines, but Flecs (https://github.com/SanderMertens/flecs) is arguably a more mature, open-source ECS implementation. You don't get to write in Rust, though, which makes it less cool in my book :)
I'm not very proud of the code I've written because I've found writing a game to be much more confusing than building websites + backends, but, as the author notes, it certainly feels more elegant than OOP or globals given the context.
I'm building for WASM and Bevy's parallelism isn't supported in that context (yet? https://github.com/bevyengine/bevy/issues/4078), so the performance wins are just so-so. Sharing a thread with UI rendering suuucks.
If anyone wants to browse some code or ask questions, feel free! https://github.com/MeoMix/symbiants
-
Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
These days, some game engines have done pretty well at making compute shaders easy to use (such as Bevy [1] -- disclaimer, I contribute to that engine). But telling the scientific/financial/etc. community that they need to run their code inside a game engine to get a decent experience is a hard sell. It's not a great situation compared to how easy it is on NVIDIA's stack.
[1]: https://github.com/bevyengine/bevy/blob/main/examples/shader...
-
Trying to write a game with mods loaded at runtime
This is the API you need: https://github.com/bevyengine/bevy/pull/9774
- Not only Unity...
What are some alternatives?
stablehlo - Backward compatible ML compute opset inspired by HLO/MHLO
Amethyst - Data-oriented and data-driven game engine written in Rust
onnx - Open standard for machine learning interoperability
Godot - Godot Engine – Multi-platform 2D and 3D game engine
tract - Tiny, no-nonsense, self-contained, Tensorflow and ONNX inference
Fyrox - 3D and 2D game engine written in Rust
iree - A retargetable MLIR-based machine learning compiler and runtime toolkit.
piston - A modular game engine written in Rust
burn - Burn is a new comprehensive dynamic Deep Learning Framework built using Rust with extreme flexibility, compute efficiency and portability as its primary goals.
RG3D - 3D and 2D game engine written in Rust [Moved to: https://github.com/FyroxEngine/Fyrox]
blaze - A Rustified OpenCL Experience
specs - Specs - Parallel ECS