Phi-4-reasoning-plus is a state-of-the-art open-weight reasoning model finetuned from Phi-4 using supervised fine-tuning and reinforcement learning. It focuses on math, science, and coding skills. This 'plus' version has higher accuracy due to additional RL training but may have higher latency.
发布日期2025年4月30日
参数规模14B
上下文长度32K
许可证MIT
知识截止2025年3月1日
Benchmarks
评测成绩
| 评测基准 | 类别 | 分数 | 来源 |
|---|---|---|---|
| FlenQA | reasoning long_context | 97.9 | 来源 |
| HumanEval+ | reasoning | 92.3 | 来源 |
| IFEval | general | 84.9 | 来源 |
| OmniMath | math reasoning | 81.9 | 来源 |
| AIME 2024 | math reasoning | 81.3 | 来源 |
| Arena Hard | general reasoning creativity | 79.0 | 来源 |
| AIME 2025 | math reasoning | 78.0 | 来源 |
| MMLU-Pro | language reasoning math general | 76.0 | 来源 |
| PhiBench | reasoning math general | 74.2 | 来源 |
| GPQA | reasoning general | 68.9 | 来源 |
| LiveCodeBench | reasoning general code | 53.1 | 来源 |
Pricing
API 价格对比
| 服务商 | 输入价 | 输出价 | 上下文 | 吞吐(tok/s) | 延迟(s) | 函数调用 | 代码执行 | 联网搜索 |
|---|---|---|---|---|---|---|---|---|
| Azure | $0.13 | $0.50 | 32K | — | — | ✗ | ✗ | ✗ |
| Azure Cognitive Services | $0.13 | $0.50 | 32K | — | — | ✗ | ✗ | ✗ |
价格单位：美元/百万 token，数据来自社区整理，仅供参考。