如何通过环境变量限定上下文的长度？

[HowardTangHw]

📦 Environment

• Official
• Official Preview
• Vercel / Zeabur / Sealos
• Docker
• Other

📌 Version

Latest Version

💻 Operating System

• Windows
• macOS
• Ubuntu
• Other Linux
• iOS
• Android
• Other

🌐 Browser

• Chrome
• Edge
• Safari
• Firefox
• Other

🐛 Bug Description

在Vercel 当中配置了`DEFAULT_AGENT_CONFIG` 为`params.max_tokens=32768;` 实际并未生效 Tokens还是爆掉了

📷 Recurrence Steps

No response

🚦 Expected Behavior

No response

📝 Additional Information

No response

[lobehubbot]

👀 @HowardTangHw

Thank you for raising an issue. We will investigate into the matter and get back to you as soon as possible.
Please make sure you have given us as much context as possible.
非常感谢您提交 issue。我们会尽快调查此事，并尽快回复您。 请确保您已经提供了尽可能多的背景信息。

[HowardTangHw]

看到源码当中, AgentConfig如果并没有enableMaxTokens则不可以开启max_tokens?

agentConfig.params.max_tokens = chatConfig.enableMaxTokens
      ? agentConfig.params.max_tokens
      : undefined;

看起来是需要在客户端开启enableMaxTokens 即开启单次回复限制 才会生效，
能否在Environment Variables 也加上enableMaxTokens的开关呢？

[lobehubbot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

Seeing in the source code, if AgentConfig does not have enableMaxTokens, then max_tokens cannot be enabled?

agentConfig.params.max_tokens = chatConfig.enableMaxTokens
      ? agentConfig.params.max_tokens
      : undefined;

[arvinxx]

试试 params.max_tokens=32768;chatConfig.enableMaxTokens=true

[HowardTangHw]

@arvinxx

配置上了并且重新构建了，还是没有生效@@

[arvinxx]

max_tokens 不是上下文的长度…是生成文本的最大长度

[lobehubbot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

max_token is not the length of the context... it is the maximum length of the generated text

[HowardTangHw]

那如何可以限定上下文的长度？
现在出现的问题就是上下文长度过长导致tokens消耗得很快
根据你前面的写法的话可以配置chatConfig.enableHistoryCount=true;chatConfig.historyCount=5 来配置对吗？

[arvinxx]

根据你前面的写法的话可以配置chatConfig.enableHistoryCount=true;chatConfig.historyCount=5 来配置对吗？

对