Can GPT-3.5 be replaced with a local large model, such as Qwen2? Are there any examples?
DebroonAir opened this issue · comments
Can GPT-3.5 be replaced with a local large model, such as Qwen2? Are there any examples?
您好，
可以的。本质来说这是个rag项目，采用的大模型是gpt3.5或者gpt4，或者开源的llm都是可以的。我们有采用最新的gpt4o进行尝试，是work的。如果采用其他模型可能需要修改prompt的格式，这里我们采用gpt匹配的langchain进行AI，system，human的message输入，你也可以采用其他message输入方式。
Yilin
Debroon ***@***.***>于2024年6月12日 周三17:32写道：
… Can GPT-3.5 be replaced with a local large model, such as Qwen2? Are there
any examples?
—
Reply to this email directly, view it on GitHub
<#9>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/APCDU63ARFDNOO7ZPMOFAJLZHAIRRAVCNFSM6AAAAABJF65DSSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGM2DQMRYGY4TSOA>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
大佬，这个好像有俩个问题，一个是没有机制复验生成结果，一个是推理速度慢、大量相似性问题
…------------------ Original ------------------
From: Yilin ***@***.***>
Date: Wed,Jun 12,2024 7:21 PM
To: wyl-willing/MindMap ***@***.***>
Cc: Debroon ***@***.***>, Author ***@***.***>
Subject: Re: [wyl-willing/MindMap] Can GPT-3.5 be replaced with a local largemodel, such as Qwen2? Are there any examples? (Issue #9)
您好，
可以的。本质来说这是个rag项目，采用的大模型是gpt3.5或者gpt4，或者开源的llm都是可以的。我们有采用最新的gpt4o进行尝试，是work的。如果采用其他模型可能需要修改prompt的格式，这里我们采用gpt匹配的langchain进行AI，system，human的message输入，你也可以采用其他message输入方式。
Yilin
Debroon ***@***.***>于2024年6月12日 周三17:32写道：
> Can GPT-3.5 be replaced with a local large model, such as Qwen2? Are there
> any examples?
>
> —
> Reply to this email directly, view it on GitHub
> <#9>, or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/APCDU63ARFDNOO7ZPMOFAJLZHAIRRAVCNFSM6AAAAABJF65DSSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGM2DQMRYGY4TSOA>
> .
> You are receiving this because you are subscribed to this thread.Message
> ID: ***@***.***>
>
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you authored the thread.Message ID: ***@***.***>
大佬，mindmap有一些应用问题，我想学习一下，我可以加钱，50块一个问题
…------------------ Original ------------------
From: Yilin ***@***.***>
Date: Wed,Jun 12,2024 7:21 PM
To: wyl-willing/MindMap ***@***.***>
Cc: Debroon ***@***.***>, Author ***@***.***>
Subject: Re: [wyl-willing/MindMap] Can GPT-3.5 be replaced with a local largemodel, such as Qwen2? Are there any examples? (Issue #9)
您好，
可以的。本质来说这是个rag项目，采用的大模型是gpt3.5或者gpt4，或者开源的llm都是可以的。我们有采用最新的gpt4o进行尝试，是work的。如果采用其他模型可能需要修改prompt的格式，这里我们采用gpt匹配的langchain进行AI，system，human的message输入，你也可以采用其他message输入方式。
Yilin
Debroon ***@***.***>于2024年6月12日 周三17:32写道：
> Can GPT-3.5 be replaced with a local large model, such as Qwen2? Are there
> any examples?
>
> —
> Reply to this email directly, view it on GitHub
> <#9>, or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/APCDU63ARFDNOO7ZPMOFAJLZHAIRRAVCNFSM6AAAAABJF65DSSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGM2DQMRYGY4TSOA>
> .
> You are receiving this because you are subscribed to this thread.Message
> ID: ***@***.***>
>
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you authored the thread.Message ID: ***@***.***>