0

BFCL: Berkeley Function-Calling Leaderboard

Active

Evaluates LLM function/tool-calling ability on a simplified split of the Berkeley Function-Calling Leaderboard (BFCL).

Open
Publisher
University of California, Berkeley
Domain
Assistants
License
mit
Published
Feb 2025
Notable for
Benchmark for evaluating Assistants.
Canonical
github.com/UKGovernmentBEIS/inspect_evals/tree/main/src/inspect_evals/bfcl

Cite

Notes

Only stored in your browser.

Attribution

Attribution policy →

Related tools

3
View all

Implementations, trainers, datasets and scaffolds linked to this eval.

WEB Search RL Env (Arcee AI)

Arcee AI

Multi-hop web search evaluation with tool use for complex question answering

Trains towardRL EnvTool UseMulti HopQuestion Answering

BFCL V 3 RL Env (Prime Intellect)

Prime Intellect

BFCL v3 evaluation environment

Trains towardRL EnvTool Use

BFCL V 3 RL Env (Community)

BFCL (Berkeley Function Call Leaderboard) Multi-Turn Environment

Trains towardRL EnvToolsFunction Calling

FAQ

What is BFCL: Berkeley Function-Calling Leaderboard?
Evaluates LLM function/tool-calling ability on a simplified split of the Berkeley Function-Calling Leaderboard (BFCL).
How can a model improve its BFCL: Berkeley Function-Calling Leaderboard score?
Tools linked to BFCL: Berkeley Function-Calling Leaderboard on Sophon include WEB Search RL Env (Arcee AI), BFCL V 3 RL Env (Prime Intellect), BFCL V 3 RL Env (Community) - RL environments, datasets, and scaffolds that target this eval.
What license is BFCL: Berkeley Function-Calling Leaderboard under?
BFCL: Berkeley Function-Calling Leaderboard is available under mit.