reasoning benchmark
BFCL Leaderboard
The Berkeley Function Calling Leaderboard (BFCL) is the first comprehensive and executable function call evaluation dedicated to assessing Large Language Models' ability to invoke functions. It evaluates serial and parallel function calls across multiple programming languages (Python, Java, JavaScript, REST API) using a novel Abstract Syntax Tree (AST) evaluation method. The benchmark consists of over 2,000 question-function-answer pairs covering diverse application domains and complex use cases including multiple function calls, parallel function calls, and multi-turn interactions.
Updated Aug 17, 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 | Score88.5% | Percentile100.0% | Participants11 | EvidenceC | Evaluated |
| Rank02 | ModelME | Score84.8% | Percentile90.0% | Participants11 | EvidenceC | Evaluated |
| Rank03 | ModelME | Score76.1% | Percentile80.0% | Participants11 | EvidenceC | Evaluated |
| Rank04 | ModelAM Nova 2 SonicAmazon | Score74.5% | Percentile70.0% | Participants11 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score70.8% | Percentile60.0% | Participants11 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score70.3% | Percentile50.0% | Participants11 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score69.1% | Percentile40.0% | Participants11 | EvidenceC | Evaluated |
| Rank08 | ModelAMNova ProAmazon | Score68.4% | Percentile30.0% | Participants11 | EvidenceC | Evaluated |
| Rank09 | ModelAMNova LiteAmazon | Score66.6% | Percentile20.0% | Participants11 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score66.4% | Percentile10.0% | Participants11 | EvidenceC | Evaluated |
| Rank11 | ModelAMNova MicroAmazon | Score56.2% | Percentile0.0% | 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.
Ranking basisThis bfcl AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ME
- Score
- 88.5%
- Speed
- Up to 100 tok/s via Bedrock
Strengths
- Ranks #1 of 11 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures BFCL, not total model capability
- 02ME
- Score
- 84.8%
- Speed
- Up to 1,204 tok/s via Cerebras
Strengths
- Ranks #2 of 11 compared models
- 90th percentile on this benchmark
- C evidence result
Considerations
- This result measures BFCL, not total model capability
- 03ME
- Score
- 76.1%
- Speed
- Up to 2,047 tok/s via Cerebras
Strengths
- Ranks #3 of 11 compared models
- 80th percentile on this benchmark
- C evidence result
Considerations
- This result measures BFCL, not total model capability
- 04AMNova 2 SonicAmazon
- Score
- 74.5%
Strengths
- Ranks #4 of 11 compared models
- 70th percentile on this benchmark
- C evidence result
Considerations
- This result measures BFCL, not total model capability
- 05ACQwen3 235B A22BAlibaba Cloud / Qwen Team
- Score
- 70.8%
- Price
- $0.70 input / $2.8 output per 1M tokens
- Speed
- Up to 68 tok/s via Fireworks
Strengths
- Ranks #5 of 11 compared models
- 60th percentile on this benchmark
- C evidence result
Considerations
- This result measures BFCL, not total model capability
Selection summary
Best AI Models for BFCL
Llama 3.1 405B Instruct currently leads BFCL with 88.5%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is BFCL?
What BFCL measures and how its scores work.
The Berkeley Function Calling Leaderboard (BFCL) is the first comprehensive and executable function call evaluation dedicated to assessing Large Language Models' ability to invoke functions. It evaluates serial and parallel function calls across multiple programming languages (Python, Java, JavaScript, REST API) using a novel Abstract Syntax Tree (AST) evaluation method. The benchmark consists of over 2,000 question-function-answer pairs covering diverse application domains and complex use cases including multiple function calls, parallel function calls, and multi-turn interactions.
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
- Yes
- Evaluation key
- bfcl|llm-stats-current
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 88.5%.
What does BFCL measure?
The Berkeley Function Calling Leaderboard (BFCL) is the first comprehensive and executable function call evaluation dedicated to assessing Large Language Models' ability to invoke functions. It evaluates serial and parallel function calls across multiple programming languages (Python, Java, JavaScript, REST API) using a novel Abstract Syntax Tree (AST) evaluation method. The benchmark consists of over 2,000 question-function-answer pairs covering diverse application domains and complex use cases including multiple function calls, parallel function calls, and multi-turn interactions.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
11 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.