巴西本土大模型Sabiá-3性能媲美顶尖模型，成本低3-4倍
Sabiá-3 Technical Report
- 基于巴西语料训练，专注葡萄牙语与本地任务
- 推理能力显著优于前代Sabiá-2 Medium，平均性能达前沿水平
- 成本仅为同类模型的1/3至1/4，适合高性价比场景
本报告介绍我们的新一代旗舰语言模型Sabiá-3及其更经济的兄弟模型Sabiázinho-3。两模型均在大规模巴西语语料上训练。在多样化的专业与学术基准测试中，其在葡萄牙语及巴西相关任务上表现优异。相比此前最优模型Sabiá-2 Medium，Sabiá-3在推理密集型任务上实现显著提升。值得注意的是，Sabiá-3的平均性能达到前沿大模型水平，同时每令牌成本仅为后者的三至四分之一，凸显领域专业化带来的优势。
原文摘要 · Abstract (English)
This report presents Sabiá-3, our new flagship language model, and Sabiazinho-3, a more cost-effective sibling. The models were trained on a large brazilian-centric corpus. Evaluations across diverse professional and academic benchmarks show a strong performance on Portuguese and Brazil-related tasks. Sabiá-3 shows large improvements in comparison to our previous best of model, Sabia-2 Medium, especially in reasoning-intensive tasks. Notably, Sabiá-3's average performance matches frontier LLMs, while it is offered at a three to four times lower cost per token, reinforcing the benefits of domain specialization.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。