TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: Fine-tune llama3 model to support function calling

3 pointsby michaelnny11 months ago
Few weeks ago, we were tasked to make llama3 model to do function calling and build an API similar&#x2F;compatible with OpenAI&#x27;s python client.<p>While we&#x27;re still improving our dataset and FastAPI backend server (which can&#x27;t be open-sourced), we&#x27;re happy to share with the community our open-sourced project of how to properly fine-tune llama3 model to support function calling, all following OpenAI&#x27;s best practices.<p>We hope this will help other people with similar tasks&#x2F;needs, happy to answer any questions.

3 comments

michaelnny11 months ago
In case anyone is interested, we have a very simple notebook which try to trick GPT-4o to reveal the structure of the prompt&#x2F;response related to function calling, and it&#x27;s located at:<p><a href="https:&#x2F;&#x2F;github.com&#x2F;michaelnny&#x2F;Llama3-FunctionCalling&#x2F;blob&#x2F;main&#x2F;ideas&#x2F;analyze_openai_api.ipynb">https:&#x2F;&#x2F;github.com&#x2F;michaelnny&#x2F;Llama3-FunctionCalling&#x2F;blob&#x2F;ma...</a>
K0IN11 months ago
The link seems to be broken
评论 #40667523 未加载
david154211 months ago
link is broken
评论 #40667525 未加载