Skip to Main Content
Trunchbull
Sign inSign up
PUBLIC REGISTRY
Partial portNo sandbox required

Berkeley Function Calling Leaderboard

Upstream publisher UC BerkeleyShishirPatil/gorilla

Executable evaluation of function selection, arguments, parallel calls, and multi-turn tool use.

toolsfunction-callingagents
PROVENANCEIMMUTABLE RELEASE
Upstream source
ShishirPatil/gorilla
Trunchbull port
Not packaged
Version
BFCL V4
Release digest
sha256:38f4…0ac1
License
Apache-2.0
Source commit: v4-catalog-2026-07-24Published July 27, 2026

WHAT THIS PROJECT MEASURES

README

BFCL measures whether models can select functions and construct valid arguments across simple, parallel, and multi-turn scenarios. Tool schemas are part of each case rather than incidental benchmark configuration.

Trunchbull currently ports a curated set of cases and their immutable tool definitions. Full upstream parity still requires additional function mappings and the complete AST grading pipeline.

EXECUTION CONTRACT

Requires the listed function tools, but does not require a machine sandbox.

TRUNCHBULL AVAILABILITY

Import status

Partial port

A curated Trunchbull port is available. Full BFCL v4 coverage and grader parity are still in progress.

3 CASES

Evaluation cases

3

bfcl/simple/weather-001

Current weather lookup

Chat lane

Select a weather function and construct its location argument.

weather.get_current

bfcl/parallel/flights-014

Parallel flight status

Chat lane

Issue independent status calls for two flight numbers.

flights.get_status

bfcl/multi-turn/calendar-008

Resolve and schedule

Chat lane

Collect missing details before creating a calendar event.

calendar.find_slotcalendar.create_event

3 REQUESTED TOOLS

Tools

3

weather.get_current

function

Returns current conditions for a named location.

Required

flights.get_status

function

Looks up the live status of a flight number.

Required

calendar.create_event

function

Creates an event from a resolved time range and attendees.

Required

MACHINE-READABLE PROVENANCE

Registry record

{
  "name": "Berkeley Function Calling Leaderboard",
  "catalogOwner": null,
  "source": {
    "url": "https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard",
    "version": "BFCL v4",
    "commit": "v4-catalog-2026-07-24",
    "license": "Apache-2.0"
  },
  "port": null,
  "release": {
    "digest": "sha256:38f4…0ac1",
    "lane": "chat",
    "sandboxRequired": false
  },
  "cases": [
    "bfcl/simple/weather-001",
    "bfcl/parallel/flights-014",
    "bfcl/multi-turn/calendar-008"
  ],
  "tools": [
    "weather.get_current",
    "flights.get_status",
    "calendar.create_event"
  ]
}

BUILD ON THIS PROJECT

Bring this benchmark into your workspace.

Sign up