chatModel cannot be null when adding SimpleLoggerAdvisor #2876
Description
Bug description
When creating a ChatClient without explicitly specifying a ChatModel (relying on auto-configuration), the call to the LLM fails if a default advisor is set (e.g., SimpleLoggerAdvisor). The following exception is thrown:
java.lang.IllegalArgumentException: chatModel cannot be null
Environment
- Spring AI version: 1.0.0-SNAPSHOT
- Java version: GraalVM 21.0.4 - VM 23.1.4
- Impacted models: VertexAIGemini, Mistral
Steps to reproduce
- Create a ChatClient without specifying a
ChatModel, using an injectedChatClient.Builderand callingchatClient.build(). - Add a default advisor (e.g., SimpleLoggerAdvisor).
- Call the LLM, specifying the model via the prompt.
Note: Everything works as expected if no default advisor is set.
Expected behavior
The call to the LLM should succeed, and the advisor should be invoked correctly during execution.
Minimal Complete Reproducible example
package com.malt.platform.spring.test.controller
import org.springframework.ai.chat.client.ChatClient
import org.springframework.ai.chat.client.advisor.SimpleLoggerAdvisor
import org.springframework.ai.chat.prompt.ChatOptions
import org.springframework.ai.chat.prompt.Prompt
import org.springframework.web.bind.annotation.PostMapping
import org.springframework.web.bind.annotation.RequestBody
import org.springframework.web.bind.annotation.RequestMapping
import org.springframework.web.bind.annotation.RestController
@RestController
@RequestMapping("/springai-test")
class SpringAiTestController(
chatClientBuilder: ChatClient.Builder,
) {
private val chatClient: ChatClient = chatClientBuilder.defaultAdvisors(SimpleLoggerAdvisor()).build()
@PostMapping("/springai")
fun springAi(
@RequestBody prompt: String,
): String? =
chatClient
.prompt(
Prompt
.Builder()
.chatOptions(
ChatOptions.builder().model("gemini-2.0-flash-001").build(),
).build(),
).user(prompt)
.call()
.content()
}