# Open source LLM Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/open-source-llm
> Markdown URL: https://aitinkerers.org/technologies/open-source-llm.md
> Technology record last updated: 2026-03-09T19:24:10Z
> Generated: 2026-09-21T21:36:04Z

Open Source LLMs (Large Language Models) deliver transparent, customizable AI, challenging proprietary models with performance from 7B to 141B parameters.

Open Source LLMs are foundation models (weights and code often public) offering complete control for commercial and research applications. Key players like Meta's Llama and Mistral AI's Mixtral 8x22B (141B total parameters) have closed the performance gap on proprietary systems. Developers leverage these models for specific use cases: for example, DeepSeek Coder V2 excels at code generation, while Google's Gemma 2 (27B) targets efficient on-device deployment. The permissive Apache 2.0 license is common, enabling deep customization and deployment for Retrieval-Augmented Generation (RAG) and complex agentic workflows, bypassing vendor lock-in.

- Official technology site: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
- Public AI Tinkerers demos and talks: 3
- Result page: 1 of 1

## Recent Public Talks and Demos

### [How function calling really works](https://chicago.aitinkerers.org/talks/rsvp_7NPy7Ea_KJI)

A detailed description of how to use logit biasing to guarantee structured output, then a demonstration of an open source LLM using function calling to select tools.

- Event context: AI Tinkerers Chicago September Meetup — 2024-09-17 — Chicago
- Public talk page: https://chicago.aitinkerers.org/talks/rsvp_7NPy7Ea_KJI

### [Fine-tuning for Function Calling](https://amsterdam.aitinkerers.org/talks/rsvp_KUVM_uX6X04)

In this talk, We will take a look at fine-tuning a OSS LLM for function calling. I will share learnings we had while producing a GPT-4o level function calling model. - What is Function calling? - Learning about model objectives - Selecting proper function call syntax - Preserving base model capabilities

- Event context: AI Tinkerers Amsterdam - June — 2024-06-20 — Amsterdam
- Public talk page: https://amsterdam.aitinkerers.org/talks/rsvp_KUVM_uX6X04

### [ZenCoder - Convert your scripts into MLOps pipelines](https://munich.aitinkerers.org/talks/rsvp_d9dijtVwpU0)

One of the first jobs of somebody entering MLOps is to convert their manual scripts or notebooks into pipelines that can be deployed on the cloud. This job is tedious and can take time. Frameworks like [ZenML](https://github.com/zenml-io/zenml) go a long way in alleviating this burden by abstracting much of the complexity away. However, recent advancements in Large Language Model-based Copilots offer hope that even more repetitive aspects of this task can be automated. Unfortunately, most open source or proprietary models like GitHub Copilot are often lagging behind the most recent versions of ML libraries, therefore giving erroneous our outdated syntax when asked simple commands. This project aims to fine-tune an open-source LLM that performs better than off-the-shelf solutions in giving the right output for the latest version of ZenML. Just to make it a bit more fun, we're going to be building ZenML pipelines to achieve this task! That way we write ZenML pipelines to train a model that can produce ZenML pipelines 🐍! Sounds fun!

- Event context: AI Tinkerers Munich - January 18 — 2024-01-18 — Munich
- Public talk page: https://munich.aitinkerers.org/talks/rsvp_d9dijtVwpU0

## Related Technologies

- [function calling](https://aitinkerers.org/technologies/function-calling) ([Markdown](https://aitinkerers.org/technologies/function-calling.md)) — 11 public demos
- [Fine-tuning](https://aitinkerers.org/technologies/fine-tuning) ([Markdown](https://aitinkerers.org/technologies/fine-tuning.md)) — 20 public demos
- [GitHub Copilot](https://aitinkerers.org/technologies/github-copilot) ([Markdown](https://aitinkerers.org/technologies/github-copilot.md)) — 20 public demos
- [GPT-4o](https://aitinkerers.org/technologies/gpt-4o) ([Markdown](https://aitinkerers.org/technologies/gpt-4o.md)) — 57 public demos
- [Logit biasing](https://aitinkerers.org/technologies/logit-biasing) ([Markdown](https://aitinkerers.org/technologies/logit-biasing.md)) — 1 public demo
- [Notebooks](https://aitinkerers.org/technologies/notebooks) ([Markdown](https://aitinkerers.org/technologies/notebooks.md)) — 2 public demos
- [Structured output](https://aitinkerers.org/technologies/structured-output) ([Markdown](https://aitinkerers.org/technologies/structured-output.md)) — 3 public demos
- [Tool selection](https://aitinkerers.org/technologies/tool-selection) ([Markdown](https://aitinkerers.org/technologies/tool-selection.md)) — 1 public demo
- [ZenML](https://aitinkerers.org/technologies/zenml) ([Markdown](https://aitinkerers.org/technologies/zenml.md)) — 1 public demo