BFCL
AI 基准测试结果和模型表现
类型: percentageGeneral
已记录的结果
11
平均分数
71.73%
分数范围
66.40% - 97.80%
领先模型
97.80% - Drex Max 1.1
分数变化趋势
按日期绘制的单项基准测试分数。
模型结果
| 组织 | 模型 | 报告时间 | 最高分数 | 信息 | 自行报告 | 来源 |
|---|---|---|---|---|---|---|
| Drex Max 1.1 | 2026年10月28日 | 97.80% | Nace launch chart "Decision Models Benchmark" (2026-10-28); tool calling | 是 | 来源 | |
| Qwen 3 235B A22B | - | 70.80% | - | 是 | 来源 | |
| Qwen 3 32B | - | 70.30% | - | 是 | 来源 | |
| Qwen 3 30B A3B | - | 69.10% | - | 是 | 来源 | |
| Qwen 3 30B A3B Instruct 2507 | - | 69.10% | inferred version-family alias from qwen3-30b-a3b | 是 | 来源 | |
| Qwen 3 30B A3B Thinking 2507 | - | 69.10% | inferred version-family alias from qwen3-30b-a3b | 是 | 来源 | |
| Qwen 3 Coder 30B A3B Instruct | - | 69.10% | inferred high-confidence family alias from qwen3-30b-a3b (score=0.5007; benches=8) | 是 | 来源 | |
| Qwen 3 Omni 30B A3B Captioner | - | 69.10% | inferred family alias from qwen3-30b-a3b (score=0.4129; benches=8) | 是 | 来源 | |
| Qwen 3 Omni 30B A3B Instruct | - | 69.10% | inferred high-confidence family alias from qwen3-30b-a3b (score=0.4819; benches=8) | 是 | 来源 | |
| Qwen 3 Omni 30B A3B Thinking | - | 69.10% | inferred high-confidence family alias from qwen3-30b-a3b (score=0.4819; benches=8) | 是 | 来源 | |
| QwQ 32B | - | 66.40% | - | 是 | 来源 |
