> For the complete documentation index, see [llms.txt](https://hacktoberfest-handbook.mlh.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://hacktoberfest-handbook.mlh.com/fest-planning-guide/open-source-prize-categories.md). # Open-Source AI Prize Challenge Every `Hacktoberfest Hack Day` runs this challenge. MLH adds it to your OrganizerHQ event automatically. Include it in your project prompts, submission instructions, opening slides, and winner-selection plan. ## Best Open-Source AI Project Build an original project that uses open-source or open-weight AI as an important part of how it works. Teams could create an agent skill, build with an open-weight large or small language model, or build or adapt an open-source model harness. These are examples within one challenge, and teams may combine them. ### Requirements * Open-source or open-weight AI must be an important part of the project. * The project must be published in a public GitHub repository and use an open-source license. * An agent skill must comply with the Agent Skill Open Standard. * A model-harness entry must include an original implementation or meaningful changes to an existing open-source harness. ## What These Terms Mean * **Open source:** the project's source code is public and carries a license that allows others to use, modify, and share it. Common choices include [MIT](https://opensource.org/license/mit), [Apache-2.0](https://opensource.org/license/apache-2.0), and [GPL-3.0](https://opensource.org/license/gpl-3.0). The [OSI license list](https://opensource.org/licenses) includes more options. * **Open-weight model:** the trained model weights are available to download or run under the model provider's terms. Open-weight does not automatically mean that every part of a model is open source, so teams should name the model and link its license or terms in the README. * **Small and large language models:** for this challenge, a small language model has 10 billion parameters or fewer, while a large language model has more than 10 billion parameters. Google's [Gemma model overview](https://ai.google.dev/gemma/docs/core) includes examples on both sides of that boundary. * **Agent skill:** a reusable set of instructions and supporting files that follows the [Agent Skills open standard](https://agentskills.io/). * **Model harness:** the software around a model that manages prompts, tools, memory, context, and actions. Teams may build one from scratch or make meaningful improvements to an existing open-source project such as OpenCode, Hermes Agent, OpenClaw, or Pi. ## What a Strong Entry Looks Like A strong project: * Uses open-source or open-weight AI as an important part of how the project works * Solves a clear problem or creates a useful experience * Shows substantial technical work completed by the team * Makes its source, license, model, and key dependencies easy to verify * Has a working demonstration and a clear explanation of what the team built Use this guidance alongside [Project Submissions and Winner Selection](/project-submissions-and-judging-for-hacktoberfest-hack-day.md). Ask teams to show the relevant code and explain the contribution of the AI technology in plain language. ## Prize The challenge has one winning team. Every member receives a DEV Badge on their DEV account. Depending on availability and your event's onboarding instructions, MLH may also provide MLH+DEV branded swag bags for the winning team. These may include a branded bag, notebook, and other items. ## Hack Days Partner Prize Categories Hack Days partner prize categories are separate from the Best Open-Source AI Project challenge. They may be assigned to selected Hacktoberfest Hack Day events by the MLH Hacktoberfest team. You may indicate a preference when applying, but an assigned category appears automatically in OrganizerHQ. See the [partner challenge details](/hack-days-partner-modules.md). --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://hacktoberfest-handbook.mlh.com/fest-planning-guide/open-source-prize-categories.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.