From d782cfbefe03c2283ec63a7217f6daefc4e0ea6a Mon Sep 17 00:00:00 2001 From: Mohammad Miadh Angkad Date: Sat, 9 Aug 2025 01:49:04 +0800 Subject: [PATCH] docs: Add Table of Contents to README Add Table of Contents after the introduction to improve document navigation and allow users to jump directly to the information they need. --- README.md | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/README.md b/README.md index b49ec6e..93abe46 100644 --- a/README.md +++ b/README.md @@ -29,6 +29,31 @@ Both models were trained using our [harmony response format][harmony] and should - **Agentic capabilities:** Use the models' native capabilities for function calling, [web browsing](#browser), [Python code execution](#python), and Structured Outputs. - **Native MXFP4 quantization:** The models are trained with native MXFP4 precision for the MoE layer, allowing `gpt-oss-120b` to run on a single 80GB GPU (like NVIDIA H100 or AMD MI300X) and `gpt-oss-20b` to run within 16GB of memory. +## Table of Contents +- [Inference Examples](#inference-examples) +- [About this Repository](#about-this-repository) +- [Setup](#setup) + - [Requirements](#requirements) + - [Installation](#installation) +- [Download the model](#download-the-model) +- [Reference Implementations](#reference-implementations) + - [PyTorch](#reference-pytorch-implementation) + - [Triton](#reference-triton-implementation-single-gpu) + - [Metal (Apple Silicon)](#reference-metal-implementation) +- [Harmony format & tools](#harmony-format--tools) +- [Clients](#clients) + - [Terminal Chat](#terminal-chat) + - [Responses API](#responses-api) + - [Codex](#codex) +- [Tools](#tools) + - [Browser](#browser) + - [Python](#python) + - [Apply Patch](#apply-patch) +- [Other details](#other-details) + - [Precision format](#precision-format) + - [Recommended Sampling Parameters](#recommended-sampling-parameters) +- [Contributing](#contributing) + ### Inference examples #### Transformers