Benchmark
BFCL-V4 Benchmark
What is BFCL-V4?
Berkeley Function Calling Leaderboard V4 (BFCL-V4) evaluates LLMs on their ability to accurately call functions and APIs, including simple, multiple, parallel…
- Modality
- text
- Evaluates
- agents, tool calling
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 15
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/bfcl-v4?top_n=500
BFCL-V4 results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| LFM2.5-2.6B | Liquid AI | 0.5688 | 2026-08-04 | evidence |
| LFM2.5-VL-3B | Liquid AI | 0.325 | 2026-08-12 | evidence |
| Nova 2 Lite | Amazon | 0.603 | 2025-12-02 | evidence |
| Nova 2 Omni | Amazon | 0.583 | 2025-12-02 | evidence |
| Nova 2 Pro | Amazon | 0.616 | 2025-12-02 | evidence |
| Qwen3.5-0.8B | Qwen | 0.253 | 2026-03-02 | evidence |
| Qwen3.5-122B-A10B | Qwen | 0.722 | 2026-02-24 | evidence |
| Qwen3.5-27B | Qwen | 0.685 | 2026-02-24 | evidence |
| Qwen3.5-2B | Qwen | 0.436 | 2026-03-02 | evidence |
| Qwen3.5-35B-A3B | Qwen | 0.673 | 2026-02-24 | evidence |
| Qwen3.5-397B-A17B | Qwen | 0.729 | 2026-02-16 | evidence |
| Qwen3.5-4B | Qwen | 0.503 | 2026-03-02 | evidence |
| Qwen3.5-9B | Qwen | 0.661 | 2026-03-02 | evidence |
| Qwen3.7 Max | Qwen | 0.75 | 2026-05-19 | evidence |
| Qwen3.7-Plus | Qwen | 0.729 | 2026-05-31 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.