更换了qwen2.5的模型后恢复正常使用。应该是不支持qwen3
我用的这个向量模型,但是对话大模型用的qwen3
我的是bge-m3:latest, 1024v
Originally posted by @jingbio in #2978
Comment From: hedyvhan
对话模型我用的qwen-plus模型也不行,试了bge-m3:latest和qwen2.5都不行,向量模型用的text-embedding-v3
Comment From: sunyuhan1998
@hedyvhan I checked the original issue but still don't quite understand what's happening.Can you describe a bit more overall what the current issue is?A code description would be great!
Comment From: hedyvhan
@hedyvhan I checked the original issue but still don't quite understand what's happening.Can you describe a bit more overall what the current issue is?A code description would be great!
we met the same question that when using QuestionAnswerAdvisor but not found, as figure shows blow:
some says they use alibaba model qwen2.5 or bge-m3:latest but not works, and i use model qwen-plus either。 code related as followed: `
public ChatClient pdfClient(OpenAiChatModel model, ChatMemory chatMemory, VectorStore vectorStore){
String serviceSystem = new String("""
""");
var qaAvisor = (Advisor) QuestionAnswerAdvisor.builder(vectorStore)
.searchRequest(SearchRequest.builder().similarityThreshold(0.8d).topK(6).build())
.build();
return ChatClient
.builder(model)
.defaultSystem(serviceSystem)
.defaultAdvisors(
new SimpleLoggerAdvisor(), // 日志记录,debug级别
MessageChatMemoryAdvisor.builder(chatMemory).build(), // 拦截时候,存储到内存
new QuestionAnswerAdvisor(vectorStore)
)
.build();
}`
and i read spring AI document:
but it not works for me.
spring AI document
Comment From: sunyuhan1998
@hedyvhan I tried to reproduce your problem, but in my demo, QuestionAnswerAdvisor
can be introduced normally:
to use it, the following dependencies need to be introduced:
<dependency>
<groupId>org.springframework.ai</groupId>
<artifactId>spring-ai-advisors-vector-store</artifactId>
</dependency>
My Spring AI version is 1.0.0-SNAPSHOT
Comment From: hedyvhan
i checked my code:
Comment From: hedyvhan
i checked my code:
I don't know where the problem lies 😭
Comment From: sunyuhan1998
I don't know where the problem lies 😭
I think you should adjust the version of your spring-ai-advisors-vector-store
dependency to 1.0.0