Vellum Leaderboard Tutorial: Compare Open-Source LLMs for Fine-Tuning

Ready Tensor · Intermediate ·🧠 Large Language Models ·2mo ago
In this video, we explore the Vellum platform and take a walkthrough of its LLM leaderboards. After looking at the Hugging Face leaderboard in the previous lesson, this video introduces how Vellum organizes and compares both open-weight and provider-managed large language models. We cover how to use these leaderboards to make more practical decisions when selecting a model — not just based on accuracy scores, but also based on size, speed, latency, and cost. You’ll learn how to: * Navigate Vellum’s Open LLM Leaderboard * Compare models across different datasets and tasks * Use coding benchm…
Watch on YouTube ↗ (saves to browser)

Chapters (8)

Introduction to Vellum leaderboards
0:17 Open LLM leaderboard and dataset-based comparisons
0:55 Choosing models based on size and infrastructure cost
1:37 Speed, latency, and time-to-first-token metrics
2:13 Cost comparisons and side-by-side model analysis
2:37 Vellum leaderboard vs Hugging Face leaderboard
3:18 Open vs managed LLMs on Vellum
3:54 Why numbers alone are not enough
5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems
Next Up
5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems
Dave Ebbelaar (LLM Eng)