Skip to content

Berkeley Function-Calling Leaderboard

A leaderboard for function calling accuracy

Visit official site

What Berkeley Function-Calling Leaderboard is

The platform is described as an online board that evaluates how accurately language models call functions or tools.

What you can do with it

  • Compare models on coding tasks
  • Read scores and rankings on the board
  • Study detailed error type analysis
  • Check cost and latency estimates

Getting started

  • Visit the leaderboard site
  • View the current scores and rankings
  • Open a model for its evaluation data
  • Use the error type analysis tool

Pros & cons

✓ What we like

  • Detailed error type analysis is listed
  • Model cost and latency estimates are provided

! What to watch out for

  • Comparisons follow the listed evaluation set
  • The board is described as periodically updated

FAQ

What does it measure?

The material states accuracy in calling functions or tools.

What extra analysis is listed?

The feature list names error types, cost and latency estimates.

Last reviewed: 2026-11-15

More AI coding tools tools

View all →

How we review