iree
wonnx
Our great sponsors
iree | wonnx | |
---|---|---|
10 | 18 | |
2,379 | 1,487 | |
4.4% | 6.8% | |
10.0 | 6.5 | |
5 days ago | 30 days ago | |
C++ | Rust | |
Apache License 2.0 | GNU General Public License v3.0 or later |
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.
iree
-
Calyx, a Compiler Infrastructure for Accelerator Generators
How is this different than the mlir infrastructure of llvm and xla implemented in https://iree.dev/?
-
Running pre-trained ML models in Godot
So I have been developing this GDExtension called iree.gd. It is mission to embed IREE, another cool project that compiles and runs ML models, into Godot. It took me quite a while, but finally It has reached alpha. Hope you guys could check it out the sample.
-
Nvidia H200 Tensor Core GPU
I am going to paste a cousin comment:
StableHLO[1] is an interesting project that might help AMD here:
> Our goal is to simplify and accelerate ML development by creating more interoperability between various ML frameworks (such as TensorFlow, JAX and PyTorch) and ML compilers (such as XLA and IREE).
From there, their goal would most likely be to work with XLA/OpenXLA teams on XLA[3] and IREE[2] to make RoCM a better backend.
[1] https://github.com/openxla/stablehlo
[2] https://github.com/openxla/iree
[3] https://www.tensorflow.org/xla
-
Nvidia reveals new A.I. chip, says costs of running LLMs will drop significantly
I want to promote that the Google project https://github.com/openxla/iree exists and IREE acts as a way to turn Tensorflow, Pytorch, and MLIR workflows to compute on cpu, vulkan compute, cuda, rocm, metal and others.
https://github.com/RechieKho/IREE.gd -- RechieKho and I collaborate on making this work for Godot Engine, but IREE.gd is at a proof of concept stage.
-
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
-
Requiem for Piet-GPU-Hal
In the ML section you mentioned Kompute and MediaPipe. Have you seen IREE? It has a Vulkan-like compute-only HAL. https://github.com/iree-org/iree
-
PyTorch on Apple M1 Faster Than TensorFlow-Metal
Exactly the kind of things we've been talking about! A fun and challenging tradeoff space and it's always great to connect with others!
Ahh linebender - I hadn't connected the name with your github account - piet-gpu is great, as is your blog! Also, for anyone skimming the comments this talk is fantastic and I share it with anyone new to the GPGPU space: https://www.youtube.com/watch?v=DZRn_jNZjbw
We waffled a bit with the API granularity in the beginning and it's taken building out most of the rest of the project in order to nail it down (the big refactor still pending). The biggest issue is that in simple models we'll end up emitting a single command buffer but anything with control flow (that we can't predicate), data dependencies (sparsity, thresholding, etc), or CPU work in the middle (IO, custom user code, etc) can break that up. We also hit cases where we need to flush work - such as if we run out of usable memory and need to defragment or resize our pools. We want to be able to (but aren't yet) reuse command buffers (CUDA graphs, etc) and that requires being able to both cache them and recreate them on demand (if we resize a pool we have to invalidate all cached command buffers using those resources, as update-after-bind is not universally available and if shapes change there's big ripples). Since most models beyond simple vision ones are ~thousands of dispatches it also lets us better integrate into multithreaded applications like you mention as apps can record commands for themselves in parallel without synchronization. It still would be nice to have certain operations inlined, though, and for that we want to allow custom hooks that we call into to add commands to the command buffers, turning things inside-out to make small amounts of work like image transformations in-between model layers possible (I'm really hoping we can avoid modeling the entire graphics pipeline in the compiler and this would be a way around that :). We haven't yet started on scheduling across queues but that's also very interesting especially in multi-GPU cases (with x4/x8 GPUs being common in datacenters, or NUMA CPU clusters that can be scheduled similarly).
We're fully open source (https://github.com/google/iree) but have been operating quietly while we get the groundwork in place - it's taken some time but now we're finally starting to stumble into success on certain problem categories (like transformers as in the post). Right now it's mostly just organized as a systems/compiler nerd honeypot for people looking for an ML/number crunching framework that (purposefully) doesn't look like any of the existing ones :)
Would love to chat more - even if just to commiserate over GPU APIs and such - everyone is welcome on the discord where a bunch of us nerds have gathered or we could grab virtual coffee (realized just now that this hn acct is ancient - I'm [email protected] :)
-
WONNX: Deep Learning on WebGPU using the ONNX format.
If you're interested in really pushing yourself, perhaps you can look at https://github.com/google/iree?
-
GPU computing on Apple Silicon
This doesn't answer your question, but it would be cool if we had something based on MLIR for GPU compute. From what I've read, it closes the gap between NVIDIA and other GPU vendors a lot more than pure compute shaders. e.g. ONNX-MLIR, PlaidML, and IREE.
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
What are some alternatives?
onnx-mlir - Representation and Reference Lowering of ONNX Models in MLIR Compiler Infrastructure
stablehlo - Backward compatible ML compute opset inspired by HLO/MHLO
torch-mlir - The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.
onnx - Open standard for machine learning interoperability
tract - Tiny, no-nonsense, self-contained, Tensorflow and ONNX inference
cutlass - CUDA Templates for Linear Algebra Subroutines
burn - Burn is a new comprehensive dynamic Deep Learning Framework built using Rust with extreme flexibility, compute efficiency and portability as its primary goals.
plaidml - PlaidML is a framework for making deep learning work everywhere.
blaze - A Rustified OpenCL Experience
rust-gpu - 🐉 Making Rust a first-class language and ecosystem for GPU shaders 🚧
benchmark - TorchBench is a collection of open source benchmarks used to evaluate PyTorch performance.