Qwen2.5-14B-Instruct is an instruction-tuned 14.7B parameter language model, part of the Qwen2.5 series. It features significant improvements in instruction following, long text generation (8K+ tokens), structured data understanding, and JSON output generation. The model supports a 128K token context length and multilingual capabilities across 29+ languages including Chinese, English, French, Spanish, and more.
发布日期2024年9月19日
参数规模14.7B
上下文长度—
许可证Apache 2.0
知识截止—
Benchmarks
评测成绩
| 评测基准 | 类别 | 分数 | 来源 |
|---|---|---|---|
| GSM8k | math reasoning | 94.8 | 来源 |
| HumanEval | reasoning code | 83.5 | 来源 |
| MBPP | reasoning general | 82.0 | 来源 |
| MATH | math reasoning | 80.0 | 来源 |
| MMLU-Redux | language reasoning math general | 80.0 | 来源 |
| MMLU | general reasoning language math | 79.7 | 来源 |
| BBH | reasoning math language | 78.2 | 来源 |
| MMLU-STEM | math reasoning physics chemistry | 76.4 | 来源 |
| MultiPL-E | general language | 72.8 | 来源 |
| ARC-C | reasoning general | 67.3 | 来源 |
| MMLU-Pro | language reasoning math general | 63.7 | 来源 |
| MBPP+ | reasoning general | 63.2 | 来源 |
| TruthfulQA | general reasoning legal healthcare finance | 58.4 | 来源 |
| HumanEval+ | reasoning | 51.2 | 来源 |
| GPQA | reasoning general | 45.5 | 来源 |
| TheoremQA | math reasoning physics finance | 43.0 | 来源 |
Pricing
API 价格对比
暂无 API 价格。