能否提供个文件用于修改默认设置

[cokice]

🥰 需求描述

就像nextweb一样，可以默认设置好压缩阈值和限制历史条数，使其类似nextweb连续对话下去，不会因为过长的对话导致api费用增加，特别是现在gpt4能上传更多的文字，普通用户没意识到这种行为，因为默认不限制条数，会导致api价格攀升。
还有作者大大我目前还是没有理解那个聊天的时候剩余token的用处是啥。

🧐 解决方案

no

📝 补充信息

No response

[lobehubbot]

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

---

🥰 Description of requirements

Just like nextweb, you can set the compression threshold and limit the number of history entries by default, so that it can continue a conversation similar to nextweb, and the api fee will not increase due to overly long conversations. Especially now that gpt4 can upload more text, ordinary users I didn't realize that this behavior, because the number of items is not limited by default, will cause the API price to rise.
Also, I still don’t understand the use of the remaining tokens during chat.

🧐 Solution

no

📝 Supplementary information

No response

[lobehubbot]

👀 @cokice

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。

[arvinxx]

就像nextweb一样，可以默认设置好压缩阈值和限制历史条数，使其类似nextweb连续对话下去，不会因为过长的对话导致api费用增加，特别是现在gpt4能上传更多的文字，普通用户没意识到这种行为，因为默认不限制条数，会导致api价格攀升。

这个倒可以变成 LobeChat 的默认设定，比如默认10条上下文这样。

需要设定默认值的话，等待 #913 实现，上一个 PR #1180 已经为此做好铺垫了，实现起来相对容易得

还有作者大大我目前还是没有理解那个聊天的时候剩余token的用处是啥。

比如你之前用 4k 上下文的 gpt-3.5，是非常容易用完的，这个是拿来做提示的。我们新版本会改成当前使用量

[lobehubbot]

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

---

Just like nextweb, you can set the compression threshold and limit the number of history entries by default, so that it can continue a conversation similar to nextweb, and the api fee will not increase due to overly long conversations, especially now that gpt4 can upload more text. Normally Users are not aware of this behavior, because the number of items is not limited by default, which will cause the API price to rise.

This can be turned into the default setting of LobeChat, such as the default 10 contexts.

If you need to set a default value, wait for #913 to be implemented. The previous PR #1180 has already paved the way for this, and it is relatively easy to implement.

Also, the author, I still don’t understand the use of the remaining tokens during chatting.

For example, the gpt-3.5 you used in the 4k context before is very easy to use up. This is used as a reminder. Our new version will be changed to the current usage

[lobehubbot]

✅ @cokice

This issue is closed, If you have any questions, you can comment and reply.
此问题已经关闭。如果您有任何问题，可以留言并回复。

[lobehubbot]

🎉 This issue has been resolved in version 0.128.0 🎉

The release is available on:

• npm package (@latest dist-tag)
• GitHub release

Your semantic-release bot 📦🚀