Uploaded July 2025 | Updated September 2026, 2 weeks ago
In this video, we take a deep dive into Arcee Fusion — a powerful new method for model merging that enables strong domain adaptation while preserving the base model’s original capabilities and knowledge.
⭐️⭐️⭐️ Don't forget to subscribe to be notified of future videos. You can also follow me on Medium at julsimon.medium.com or Substack at https://julsimon.substack.com. ⭐️⭐️⭐️
We start with a high-level overview of what Arcee Fusion is and why it matters. Then, step by step, we break down the full merging process, using animations to clarify the complex math behind the scenes and show how it all contributes to performance gains.
What you'll learn:
- The core ideas behind Arcee Fusion
- A detailed look at the merging algorithm
- How the technique keeps the base model largely intact
Next, we jump into the code and run a real-world example using two Llama-3.1-8B models to demonstrate how it all works in practice.
We also introduce mergekit-diff, a new tool from Mergekit, to compare parameter differences between:
- The base and fine-tuned models
- The base and fused models
This final analysis highlights how Arcee Fusion updates only a small subset of the base model’s parameters, demonstrating just how efficient and elegant this approach truly is.
👉 Whether you're into model fine-tuning, LLM optimization, or just curious about the latest in AI infrastructure, this one's for you.
** Mergekit
Learn more and get expert support at arcee.ai/product/mergekit
Repository: github.com/arcee-ai/mergekit
Mergekit-Diff pull request: github.com/arcee-ai/mergekit/pull/608
⭐️⭐️⭐️ While you're here, I’ve got a great deal for you! If you care about your online security, you need Proton Pass — the ultra-secure password manager from the creators of Proton Mail. GET 60% OFF at go.getproton.me/aff_c?offer_id=42&aff_id=13055&url_id=994 ⭐️⭐️⭐️
In this video, we take a deep dive into Arcee Fusion — a powerful new method for model merging that enables strong domain adaptation while preserving the base model’s original capabilities and knowledge.
⭐️⭐️⭐️ Don't forget to subscribe to be notified of future videos. You can also follow me on Medium at julsimon.medium.com or Substack at https://julsimon.substack.com. ⭐️⭐️⭐️
We start with a high-level overview of what Arcee Fusion is and why it matters. Then, step by step, we break down the full merging process, using animations to clarify the complex math behind the scenes and show how it all contributes to performance gains.
What you'll learn:
- The core ideas behind Arcee Fusion
- A detailed look at the merging algorithm
- How the technique keeps the base model largely intact
Next, we jump into the code and run a real-world example using two Llama-3.1-8B models to demonstrate how it all works in practice.
We also introduce mergekit-diff, a new tool from Mergekit, to compare parameter differences between:
- The base and fine-tuned models
- The base and fused models
This final analysis highlights how Arcee Fusion updates only a small subset of the base model’s parameters, demonstrating just how efficient and elegant this approach truly is.
👉 Whether you're into model fine-tuning, LLM optimization, or just curious about the latest in AI infrastructure, this one's for you.
** Mergekit
Learn more and get expert support at arcee.ai/product/mergekit
Repository: github.com/arcee-ai/mergekit
Mergekit-Diff pull request: github.com/arcee-ai/mergekit/pull/608
⭐️⭐️⭐️ While you're here, I’ve got a great deal for you! If you care about your online security, you need Proton Pass — the ultra-secure password manager from the creators of Proton Mail. GET 60% OFF at go.getproton.me/aff_c?offer_id=42&aff_id=13055&url_id=994 ⭐️⭐️⭐️
