agents benchmark
BFCL-V4 Leaderboard
BFCL-V4 evaluates LLMs on their ability to accurately call functions and APIs across diverse programming scenarios.
Updated Sep 5, 2026
Models18
Model coverage18
MetricScore
EvidenceB
BFCL-V4 Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score75.00% | Percentile100.00% | Participants18 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score72.90% | Percentile94.12% | Participants18 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score72.90% | Percentile88.24% | Participants18 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score72.20% | Percentile82.35% | Participants18 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score68.50% | Percentile76.47% | Participants18 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score67.30% | Percentile70.59% | Participants18 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score66.10% | Percentile64.71% | Participants18 | EvidenceC | Evaluated |
| Rank08 | ModelAM Nova 2 ProAmazon | Score61.60% | Percentile58.82% | Participants18 | EvidenceC | Evaluated |
| Rank09 | ModelIB | Score61.39% | Percentile52.94% | Participants18 | EvidenceC | Evaluated |
| Rank10 | ModelAMNova 2 LiteAmazon | Score60.30% | Percentile47.06% | Participants18 | EvidenceC | Evaluated |
| Rank11 | ModelAMNova 2 OmniAmazon | Score58.30% | Percentile41.18% | Participants18 | EvidenceC | Evaluated |
| Rank12 | ModelLA | Score56.88% | Percentile35.29% | Participants18 | EvidenceC | Evaluated |
| Rank13 | ModelIB | Score52.41% | Percentile29.41% | Participants18 | EvidenceC | Evaluated |
| Rank14 | ModelIB | Score52.39% | Percentile23.53% | Participants18 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score50.30% | Percentile17.65% | Participants18 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score43.60% | Percentile11.76% | Participants18 | EvidenceC | Evaluated |
| Rank17 | ModelLA | Score32.50% | Percentile5.88% | Participants18 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score25.30% | Percentile0.00% | Participants18 | EvidenceC | Evaluated |
BFCL-V4 Highlights
The leading models and scores on this benchmark.
BFCL-V4 Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for BFCL-V4
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is BFCL-V4?
What BFCL-V4 measures and how its scores work.
BFCL-V4, or Berkeley Function Calling Leaderboard V4, is a benchmark in the agents category.
It measures the Score ratio for simple, multiple, parallel, and nested function calls.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- BFCL-V4
- Modality
- text
- Primary category
- agents
- 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-V4.
Which model scores highest on BFCL-V4?
Qwen3.7 Max is currently ranked first with 75.00%.
What are the top three models on BFCL-V4?
The current leaders are Qwen3.7 Max (75.00%), Qwen3.5-397B-A17B (72.90%), and Qwen3.7-Plus (72.90%).
Which BFCL-V4 model has the lowest official input price?
Qwen3.5-35B-A3B has the lowest matched official input price at $0.25 input / $2.0 output per 1M tokens.
Which models are fastest among BFCL-V4 results?
No matched runtime record is currently available.
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.