spring-ai
spring-ai copied to clipboard
An Application Framework for AI Engineering
I want to use the gemini apiKey to access my gemini Ai instead of using project-id+location. Currently, the spring ai module using gemini only supports project-id+location using google cloud. The...
**Bug description** The following exception occurred when I added the Spring AI dependency library ``` . ____ _ __ _ _ /\\ / ___'_ __ _ _(_)_ __ __ _...
Changed the filter template in PgVercorStore to replace JSONPath expressions with JSONB field access, using standard SQL operators such as '=' instead of '==', 'AND' instead of '&&', and 'OR'...
among other things it should be possible to specify a function in the OllamaChatOptions
Currently, vector store automatically calls the embedding client to generate the document embedding without checking whether the document already had an embedding. In this PR, I first check if the...
我在使用SpringWebFlux配合OllamaChatClient的stream方法返回大模型响应时如何发送重复的msg程序便会抛出异常。当我给msg添加随意无意义标点符号(与上条消息不一致)便会正常响应,以下是我的代码及异常信息。 =========================================================================== @RestController @RequestMapping("/llama") public class AIChatController { @Resource private OllamaChatClient ollamaChatClient; @GetMapping("/chat") public Flux chat(String msg) { return ollamaChatClient.stream(new Prompt(msg)); } } =========================================================================== 2024-05-09T18:33:58.966+08:00 ERROR 16072 --- [tor-http-nio-14] o.s.w.s.adapter.HttpWebHandlerAdapter...
Added in Vertex client configuration the possibility to use GoogleSearchRetrieval as tool as grounding feature https://cloud.google.com/vertex-ai/generative-ai/docs/model-reference/grounding next up (on other PR) grounding support with "retrieval" tool, more studies needed.
Reefer issue #652 For some use case after local function is called no need to pass the results to LLMs. (eg: data extraction from documents)
I find that the results improve when I add ", without any introduction." to the prompt. Example: ``` List 10 unique unicorn names suitable for IDs with just regular ASCII...
`BeanOutputConverter`works fine with records too: ``` record Profile( String nickname, int hornlength ) {} var outputParser = new BeanOutputConverter( Profile.class ); System.out.println( outputParser.getFormat() ); ``` -> ``` Your response should...