From 8838279a5bfe15ce274769d5a9eb2b46b72a4c4e Mon Sep 17 00:00:00 2001 From: avosa Date: Sat, 12 Sep 2026 10:45:14 -0700 Subject: [PATCH 1/2] add gosslens to Machine Learning Framework A real-time vision engine in Zig whose ONNX inference engine is written from scratch rather than wrapping a runtime: quantization, control flow and allocation-free inference, running published models on device. --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index a825bb4..5cabba7 100644 --- a/README.md +++ b/README.md @@ -576,6 +576,7 @@ - [zml](https://github.com/zml/zml) - A high performance machine learning stack for Zig. - [Zigrad](https://github.com/Marco-Christiani/zigrad) - A deep learning framework built on an autograd engine with high level abstractions and low level control. Trains neural networks 2.5x faster than PyTorch on Apple Silicon and 1.5x faster on CPU. - [SilasMarvin/dnns-from-scratch-in-zig](https://github.com/SilasMarvin/dnns-from-scratch-in-zig) - A very simple implementation of deep neural networks written in the Zig programming language. +- [myzonerocks/gosslens](https://github.com/myzonerocks/gosslens) - Real-time visual plumbing for agents: one core behind a frozen C ABI, with its own from-scratch ONNX inference engine (quantization, control flow, allocation-free inference) running published models on device. ### Large Language Model From de7ab89b66a2e843f4316b47180a7c709be2ecd9 Mon Sep 17 00:00:00 2001 From: avosa Date: Sat, 12 Sep 2026 20:40:27 -0700 Subject: [PATCH 2/2] gosslens: one sentence, with the AI attribution the guide asks for --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 5cabba7..697697a 100644 --- a/README.md +++ b/README.md @@ -576,7 +576,7 @@ - [zml](https://github.com/zml/zml) - A high performance machine learning stack for Zig. - [Zigrad](https://github.com/Marco-Christiani/zigrad) - A deep learning framework built on an autograd engine with high level abstractions and low level control. Trains neural networks 2.5x faster than PyTorch on Apple Silicon and 1.5x faster on CPU. - [SilasMarvin/dnns-from-scratch-in-zig](https://github.com/SilasMarvin/dnns-from-scratch-in-zig) - A very simple implementation of deep neural networks written in the Zig programming language. -- [myzonerocks/gosslens](https://github.com/myzonerocks/gosslens) - Real-time visual plumbing for agents: one core behind a frozen C ABI, with its own from-scratch ONNX inference engine (quantization, control flow, allocation-free inference) running published models on device. +- [myzonerocks/gosslens](https://github.com/myzonerocks/gosslens) - Real-time visual plumbing for agents, with a from-scratch ONNX engine that runs published models on device. AI-assisted. ### Large Language Model