Sonnet 5 引发争议：编程费用高企，金融领域表现最强；与 Opus 4.8 和 GPT 5.5 对比
英文摘要
Sonnet 5's new tokenizer pushes its actual cost to levels similar to Opus 4.8. It is regarded as the best model for finance tasks like GDPeval and investment research, and it prefers tool calling to verify facts, which improves report accuracy but also raises expenses. A major pain point is that using Sonnet 5 for programming can cost more than Opus 4.8, driving most of the user complaints. Opus 4.8 excels in complex coding, planning, and HTML design but its writing trails Opus 4.6, and its tokenizer increases costs compared to the older version; overall it is considered on par with GPT 5.5. GPT 5.5 remains the preferred choice for programming. Sonnet 5, Opus 4.8, and GPT 5.5 are now available on the Cola platform.
中文摘要
Sonnet 5 更换了新 tokenizer，导致实际费用与 Opus 4.8 相近。在金融领域（如 GDPeval 和投资调研）中表现最佳，并倾向于调用工具核查事实，能提高报告准确性，但相应费用也更高。一个突出问题是用 Sonnet 5 编程时费用可能超过 Opus 4.8，这是用户吐槽最多的点。Opus 4.8 在复杂编程、规划和 HTML 设计方面非常强，但写作不如 Opus 4.6，且新 tokenizer 的花费也比 4.6 更高，整体与 GPT 5.5 各有千秋。编程方面目前首选仍是 GPT 5.5。Sonnet 5、Opus 4.8 和 GPT 5.5 现已上线 Cola 平台。
关键要点
Sonnet 5's new tokenizer makes its actual cost similar to Opus 4.8.
Sonnet 5 的新 tokenizer 使其实际费用与 Opus 4.8 相近。
Sonnet 5 is the best model for finance tasks (GDPeval, investment research) and uses tool calling to improve report accuracy, at higher cost.
Sonnet 5 在金融类任务（如 GDPeval 和投资调研）中表现最佳，并通过工具调用提高报告准确性，但费用更高。
Programming with Sonnet 5 can cost more than Opus 4.8, causing widespread user complaints.
使用 Sonnet 5 进行编程时费用可能超过 Opus 4.8，引发大量用户吐槽。
Opus 4.8 is strong in complex programming, planning, and HTML design, but its writing quality lags behind Opus 4.6 and its tokenizer increases costs.
Opus 4.8 在复杂编程、规划和 HTML 设计方面表现出色，但写作能力不如 Opus 4.6，且其 tokenizer 增加了成本。
GPT 5.5 is currently the preferred model for programming tasks.
目前编程任务的首选模型是 GPT 5.5。
Sonnet 5, Opus 4.8, and GPT 5.5 are all available on the Cola platform.
Sonnet 5、Opus 4.8 和 GPT 5.5 均已在 Cola 平台上线。