Spring AI Error "toolInput cannot be null or empty" when calling in stream mode

Bug description Getting error "toolInput cannot be null or empty" while calling .stream() in chatClient

java.lang.IllegalArgumentException: toolInput cannot be null or empty
    at org.springframework.util.Assert.hasText(Assert.java:253) ~[spring-core-6.2.6.jar:6.2.6]
    at org.springframework.ai.tool.method.MethodToolCallback.call(MethodToolCallback.java:99) ~[spring-ai-model-1.0.0.jar:1.0.0]
    at org.springframework.ai.model.tool.DefaultToolCallingManager.lambda$executeToolCall$5(DefaultToolCallingManager.java:224) ~[spring-ai-model-1.0.0.jar:1.0.0]
    at io.micrometer.observation.Observation.observe(Observation.java:564) ~[micrometer-observation-1.14.6.jar:1.14.6]
    at org.springframework.ai.model.tool.DefaultToolCallingManager.executeToolCall(DefaultToolCallingManager.java:221) ~[spring-ai-model-1.0.0.jar:1.0.0]
    at org.springframework.ai.model.tool.DefaultToolCallingManager.executeToolCalls(DefaultToolCallingManager.java:137) ~[spring-ai-model-1.0.0.jar:1.0.0]
    at org.springframework.ai.openai.OpenAiChatModel.lambda$internalStream$10(OpenAiChatModel.java:369) ~[spring-ai-openai-1.0.0.jar:1.0.0]
    at reactor.core.publisher.FluxDefer.subscribe(FluxDefer.java:46) ~[reactor-core-3.7.5.jar:3.7.5]
    at reactor.core.publisher.FluxSubscribeOn$SubscribeOnSubscriber.run(FluxSubscribeOn.java:194) ~[reactor-core-3.7.5.jar:3.7.5]
    at reactor.core.scheduler.WorkerTask.call(WorkerTask.java:84) ~[reactor-core-3.7.5.jar:3.7.5]
    at reactor.core.scheduler.WorkerTask.call(WorkerTask.java:37) ~[reactor-core-3.7.5.jar:3.7.5]
    at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:317) ~[na:na]
    at java.base/java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:304) ~[na:na]
    at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144) ~[na:na]
    at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642) ~[na:na]
    at java.base/java.lang.Thread.run(Thread.java:1583) ~[na:na]

Environment Spring AI 1.0.0, Java version temurin-21.0.1, Vllm

Expected behavior There are no error when invoke .call() method, but have error when invoke .stream()

Minimal Complete Reproducible example Run Vllm:

docker run --runtime nvidia --gpus all --name Qwen3-8B-AWQ -p 8000:8000 --ipc=host vllm/vllm-openai:latest --model Qwen/Qwen3-8B-AWQ --gpu-memory-utilization 0.75 --max-model-len 8192 --enforce-eager --enable-auto-tool-choice --tool-call-parser hermes

application.properties

spring.ai.openai.base-url=http://localhost:8000
spring.ai.openai.chat.options.model=Qwen/Qwen3-8B-AWQ
spring.ai.openai.api-key=KEY

Create Simple Date Time calling tool

@Component
public class DateTimeTools {
  private static final Logger log = LoggerFactory.getLogger(DateTimeTools.class);

  @Tool(description = "Get date and time user")
  public String getCurrentDateTime() {
    log.info("Called tool getCurrentDateTime()");
    return LocalDateTime.now().toString();
  }
}

Create Simple controller:

@RestController
public class ChatApiController {
  private final ChatClient chatClient;

  public ChatApiController(ChatClient.Builder chatClientBuilder, DateTimeTools dateTimeTools) {
    this.chatClient = chatClientBuilder
        .defaultTools(dateTimeTools)
        .build();
  }

  @PostMapping("/api/stream/v1")
  public Flux<String> sendStreamMessage(@RequestBody MessageRequest message) {
    String userInput = message.message();

    return this.chatClient.prompt()
        .user(userInput)
        .stream()
        .content();
  }
}

Steps to reproduce Call method: POST http://localhost:8080/api/stream/v1 Content-Type: application/json

{ "message": "Give me current date" }