Alternatives
Products that do what Graphzero does
Zero-copy C++ graph engine to train PyTorch GNNs with 0 RAM.
- 1IS
2022 · github.com
- 2LF
We're excited to announce that we've open-sourced LeanRL, a lightweight PyTorch reinforcement learning library that provides recipes for fast RL training using torch.compile and CUDA graphs. By leveraging these tools, we've achieved significant speed-ups compared to the original CleanRL implementations - up to 6x faster! Reinforcement learning is notoriously CPU-bound due to the high frequency of small CPU operations. PyTorch's powerful compiler can help alleviate these issues, but comes with its own costs. LeanRL addresses this challenge by providing simple recipes to accelerate your…
2024 · github.com
- 3KY
2025 · github.com
- 4HI
I found that duplicating a specific block of 7 middle layers in Qwen2-72B, without modifying any weights, improved performance across all Open LLM Leaderboard benchmarks and took #1. As of 2026, the top 4 models on that leaderboard are still descendants. The weird finding: single-layer duplication does nothing. Too few layers, nothing. Too many, it gets worse. Only circuit-sized blocks of ~7 layers work. This suggests pretraining carves out discrete functional circuits in the layer stack that only work when preserved whole. The whole thing was developed on 2x RTX 4090s in my basement. I'm…
Mar 2026 · dnhkng.github.io
- 5TL
Hello HN! For the past 6 months I've been working on an open source python library that implements differentiable geometric optics in PyTorch. It's very experimental still, but eventually the goal is to use it to design optical systems with a state of the art optimization framework and a beautiful code based API. Think OpenSCAD, but for optical systems. Not only is PyTorch's autograd an amazing general purpose optimizer, but torch.nn (the neural network building blocks) can be used pretty much out of the box to model an optical system. This is because there is a strong analogy to be made…
2025 · victorpoughon.github.io
- 6GF
Hello HN, We are pleased to introduce you graphlearn-for-pytorch (https://github.com/alibaba/graphlearn-for-pytorch), an open-source distributed graph neural network library based on PyTorch and compatible with PyG. Our library is designed to make it easy for developers to build and train large-scale graph models in a distributed environment. With graphlearn-for-pytorch, you can leverage GPUs to accelerate graph sampling and utilize UVA to reduce the overheads of feature collection. Following a scalable design, graphlearn-for-pytorch supports training GNN models on…
2023 · github.com
- 7
- 8IB
Hi HN, Over the past few months, I've been building `dsc`, a tensor library from scratch in C++/CUDA. My main focus has been on getting the basics right, prioritizing a clean API, simplicity, and clear observability for running small LLMs locally. The key features are: - C++ core with CUDA support written from scratch. - A familiar, PyTorch-like Python API. - Runs real models: it's complete enough to load a model like Qwen from HuggingFace and run inference on both CUDA and CPU with a single line change[1]. - Simple, built-in observability for both Python and C++. Next on the roadmap is…
2025 · github.com
- 9
- 10
- 11RT
I ported Microsoft's TRELLIS.2 (4B parameter image-to-3D model) to run on Apple Silicon via PyTorch MPS. The original requires CUDA with flash_attn, nvdiffrast, and custom sparse convolution kernels: none of which work on Mac. I replaced the CUDA-specific ops with pure-PyTorch alternatives: a gather-scatter sparse 3D convolution, SDPA attention for sparse transformers, and a Python-based mesh extraction replacing CUDA hashmap operations. Total changes are a few hundred lines across 9 files. Generates ~400K vertex meshes from single photos in about 3.5 minutes on M4 Pro (24GB). Not as fast as…
Apr 2026 · github.com
- 12WA
In browser PPO training demo, made possible by tinygrad: TinyJit -> WebGPU kernels. Requires WebGPU.
May 2026 · ppo.gradexp.xyz
- 13LO
Hi HN, I’m Joe. My friends Matthew, Jake and I are building Luminal (https://luminalai.com/), a GPU compiler for automatically generating fast GPU kernels for AI models. It uses search-based compilation to achieve high performance. We take high level model code, like you'd have in PyTorch, and generate very fast GPU code. We do that without using LLMs or AI - rather, we pose it as a search problem. Our compiler builds a search space, generates millions of possible kernels, and then searches through it to minimize runtime. You can try out a demo in `demos/matmul` on mac to…
2025 · github.com
- 14CA
I'm a machine learning engineer and researcher. I got fed up with how difficult it is to understand why neural networks behave the way they do, so i wrote a library to help with it. Comgra (computation graph analysis) is a library you can use with pytorch to extract all the tensor data you care about and visualize it graphically in a browser. This allows for a much more detailed analysis of what is happening than the usual approach of using tensorboard. You can go investigate tensors as training proceeds, drill down into individual neurons, inspect single data sets that are of special…
2023 · github.com
- 15
- 16
- 17
- 18SM
Hello HN, I built Syna to understand how modern ML frameworks like PyTorch actually work — from the ground up. It’s a minimal, define-by-run (dynamic graph) framework inspired by DeZero, written entirely with NumPy. Unlike most libraries, Syna includes a basic reinforcement learning module right inside the same framework — no separate packages. It’s not about speed or GPUs — it’s about clarity, simplicity, and learning the internals of machine learning. Great for students, educators, and anyone curious about what’s really happening under the hood. GitHub:…
Oct 2025 · github.com
- 19
- 20TO
Hi HN! We're Gabriel & Viraj, and we're excited to open source TensorZero. To be a little cheeky, TensorZero is an open-source platform that helps LLM applications graduate from API wrappers into defensible AI products. 1. Integrate our model gateway 2. Send metrics or feedback 3. Unlock compounding improvements in quality, cost, and latency It enables a data & learning flywheel for LLMs by unifying: • Inference: one API for all LLMs, with <1ms P99 overhead • Observability: inference & feedback → your database • Optimization: better prompts, models, inference strategies • Experimentation:…
2024 · github.com
- 21DL
Want to create a Deep Learning framework from scratch? Checkout this resource: https://koki0702.github.io/dezero-book/en/index.html In this book, you will create a deep learning framework called "DeZero" from scratch (from zero), which is the original framework of this book. With minimal code, the framework's modern features are realized. In this book, you will make this small - yet powerful enough - framework in a total of 60 steps. It will deepen your knowledge of modern frameworks such as PyTorch and TensorFlow.
2022 · koki0702.github.io
- 22OA
https://github.com/gugarosa/opytimizer Did you ever reach a bottleneck in your computational experiments? Are you tired of selecting suitable parameters for a chosen technique? If yes, Opytimizer is the real deal! This package provides an easy-to-go implementation of meta-heuristic optimizations. From agents to search space, from internal functions to external communication, we will foster all research related to optimizing stuff. Use Opytimizer if you need a library or wish to: - Create your optimization algorithm; - Design or use pre-loaded optimization tasks; -…
2021
- 23ZA
This is a low-level opensource library I developed for my own use and decided to share, as it makes it possible to process large checkpoints of neural networks without renting high-RAM instances, on a regular PC. It replaces torch.load() with a custom function that produces a dictionary that materializes tensors on the fly. Compared to other solutions it doesn't require sharding or re-encoding checkpoints and uses them completely as-is. It is a foundation to make it possible to run inference and compress language models and other large models one layer at a time - in principle, even one…
2023 · github.com
- 24IB
"Creatures" is a stretch given that the environment is Minecraft, but the idea is simple: iteratively add blocks conditionally (tensor convolution) on the current environment (blocks) to maximize some reward. In this case I use PPO RL to train creatures to touch a glowstone block but you can adapt it to use any algorithm and reward (easily, as it uses the Ray framework). What I like about this work: iteratively finding solutions has a long and colorful history of doing things well: gradient boosting, ResNets, Stable Diffusion, etc. We're after some end optimal state and usually try to get…
2024 · github.com
Ranked by how close each launch is in meaning, then by votes. Refine with a description →