Staging environment

A gentle introduction to open weight models

Hosted by Anny He

Wed, Sep 30, 2026

7:00 PM UTC (30 minutes)

Virtual (Zoom)

Free to join

Invite your network

What you'll learn

How to evaluate open-weight models against your use case

So you can pick the best model for your use case

Which models to use for multi-modal vs. text-only

Go from overwhelm to shortlisted candidates

Trade-offs of running models locally vs. hosted remotely

We'll explore local providers (eg. ollama) and remote ones (eg. AWS)

Why this topic matters

Open-weight models have been catching up to frontier models, and coding harnesses like OpenCode and pi.dev has matured too. You can test open-weight models easily using prompts and workflows you already have. If you're building AI products, open-weight models can meaningfully lower token costs. GLM models, for example, are reportedly around 1/10 the cost of US frontier models.

You'll learn from

Anny He

Software consultant, speaker, AI educator

Anny He is the founder of Allpaca AI study group, a remote community where engineers demo AI technology to each other, and writes about open-weight models and AI experiments at AI turtle.

Anny has held Staff Engineer roles at LegalZoom and Imply and served as a Developer Evangelist at Salesforce. She has also worked at an AI-native startup building production LLM features, where she experienced the growing token costs first-hand and learned about open weight models.

Anny is a frequent speaker on AI topics, including at First Robotics and UtahJS, and works as a software consultant helping teams build and ship products in the AI era.

See all products from Anny

Sign up to join this lesson

By continuing, you agree to Maven's Terms and Privacy Policy.