reasoning benchmark
BFCL Leaderboard
BFCL evaluates Large Language Models' ability to invoke functions through serial and parallel calls across Python, Java, JavaScript, and REST API contexts.
Updated Sep 5, 2026
Models11
Model coverage11
MetricScore
EvidenceB
BFCL Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelME | Score88.50% | Percentile100.00% | Participants11 | EvidenceC | Evaluated |
| Rank02 | ModelME | Score84.80% | Percentile90.00% | Participants11 | EvidenceC | Evaluated |
| Rank03 | ModelME | Score76.10% | Percentile80.00% | Participants11 | EvidenceC | Evaluated |
| Rank04 | ModelAM Nova 2 SonicAmazon | Score74.50% | Percentile70.00% | Participants11 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score70.80% | Percentile60.00% | Participants11 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score70.30% | Percentile50.00% | Participants11 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score69.10% | Percentile40.00% | Participants11 | EvidenceC | Evaluated |
| Rank08 | ModelAMNova ProAmazon | Score68.40% | Percentile30.00% | Participants11 | EvidenceC | Evaluated |
| Rank09 | ModelAMNova LiteAmazon | Score66.60% | Percentile20.00% | Participants11 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score66.40% | Percentile10.00% | Participants11 | EvidenceC | Evaluated |
| Rank11 | ModelAMNova MicroAmazon | Score56.20% | Percentile0.00% | Participants11 | EvidenceC | Evaluated |
BFCL Highlights
The leading models and scores on this benchmark.
BFCL Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for BFCL
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is BFCL?
What BFCL measures and how its scores work.
The Berkeley Function Calling Leaderboard (BFCL) is a benchmark consisting of over 2,000 question-function-answer pairs covering multiple function calls, parallel function calls, and multi-turn interactions across diverse application domains.
BFCL measures function-calling ability using an Abstract Syntax Tree (AST) evaluation method, with results reported as a Score ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- BFCL
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about BFCL.
Which model scores highest on BFCL?
Llama 3.1 405B Instruct is currently ranked first with 88.50%.
What are the top three models on BFCL?
The current leaders are Llama 3.1 405B Instruct (88.50%), Llama 3.1 70B Instruct (84.80%), and Llama 3.1 8B Instruct (76.10%).
Which BFCL model has the lowest official input price?
Nova Micro has the lowest matched official input price at $0.04 input / $0.14 output per 1M tokens.
Which models are fastest among BFCL results?
The fastest matched records are Llama 3.1 8B Instruct (2,047.00 tok/s via Cerebras), Llama 3.1 70B Instruct (1,204.00 tok/s via Cerebras), and Qwen3 30B A3B (82.57 tok/s via DeepInfra).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.