当前位置：首页 > news >正文

Spring AI 集成 DeepSeek 原生供应商：告别 OpenAI 兼容层，获取结构化推理过程

news 2026/6/28 5:17:35

背景

项目中一直用 Spring AI 的 OpenAI 兼容层调用 DeepSeek API（把spring.ai.openai.base-url指向https://api.deepseek.com）。这种方式能跑，但有几个痛点：

拿不到reasoning_content：DeepSeek 的推理过程（CoT）不会以结构化字段返回，只能让模型把思考过程包在<think>标签里，前端再用状态机做标签解析，极其脆弱
缺失 DeepSeek 特有 API：Prefix Completion、DeepSeek 特有的参数等都无法使用
配置语义不清晰：写着openai，实际调的却是 DeepSeek，维护成本高

Spring AI 在 1.x 版本已经官方支持了 DeepSeek，本文记录完整的迁移过程。

一、添加依赖

在pom.xml中添加 DeepSeek Starter（版本由 BOM 1.1.3 统一管理）：

<dependency> <groupId>org.springframework.ai</groupId> <artifactId>spring-ai-starter-model-deepseek</artifactId> </dependency>

二、配置供应商

在application-dev.yml中添加 DeepSeek 配置块：

spring: ai: deepseek: api-key: sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx chat: options: model: deepseek-v4-flash temperature: 1.0

API Key 建议通过环境变量注入，避免硬编码。

三、注册 Bean

在AiConfig.java中注册 DeepSeek 的 ChatModel 和 ChatClient：

import org.springframework.ai.deepseek.DeepSeekChatModel; @Bean("deepseekChatModel") public ChatModel deepseekChatModel(DeepSeekChatModel deepSeekChatModel) { return deepSeekChatModel; } @Bean("deepseekChatClient") public ChatClient deepseekChatClient( @Qualifier("deepseekChatModel") ChatModel deepseekChatModel, MessageFormatAdvisor messageFormatAdvisor, LifecycleToolCallAdvisor lifecycleToolCallAdvisor, TaskProgressAdvisor taskProgressAdvisor, RetryAdvisor retryAdvisor) { return ChatClient.builder(deepseekChatModel) .defaultToolContext(new HashMap<>(Map.of("debug", true))) .defaultAdvisors( messageFormatAdvisor, lifecycleToolCallAdvisor, taskProgressAdvisor, retryAdvisor ) .build(); }

四、Controller 改造 — 原生推理流式输出

改造前：每个 SSE Chunk 拿到的是AssistantMessage，<think>标签可能被切碎在多个 Chunk 里，需要维护复杂的状态机做拼接。

改造后：使用DeepSeekAssistantMessage，reasoningContent和text是两个独立字段：

import org.springframework.ai.deepseek.DeepSeekAssistantMessage; .concatMap(response -> { AssistantMessage output = response.getResult().getOutput(); List<ServerSentEvent<ChatChunk>> events = new ArrayList<>(); if (output.getToolCalls() != null && !output.getToolCalls().isEmpty()) { // handle tool calls return Flux.fromIterable(events); } if (output instanceof DeepSeekAssistantMessage dsMsg) { String reasoning = dsMsg.getReasoningContent(); if (reasoning != null && !reasoning.isEmpty()) { state.accumulateReasoning(reasoning, events); } } String text = output.getText(); if (text != null && !text.isEmpty()) { state.flushReasoning(events); events.add(createEvent("message", state.messageId(), text, null)); } return Flux.fromIterable(events); })

五、推理内容缓冲优化

reasoningContent以 Token 粒度到达，每个 SSE Chunk 可能只有一个字，直接推给前端会导致渲染碎片化。需要在服务端按语义边界缓冲：

private static class StreamState { private static final int REASONING_FLUSH_THRESHOLD = 50; private static final Pattern SENTENCE_BOUNDARY = Pattern.compile("[。！？.!?\n]+"); private final StringBuilder reasoningBuffer = new StringBuilder(); public void accumulateReasoning(String delta, List<ServerSentEvent<ChatChunk>> target) { reasoningBuffer.append(delta); String buf = reasoningBuffer.toString(); var matcher = SENTENCE_BOUNDARY.matcher(buf); int lastEnd = 0; while (matcher.find()) { String segment = buf.substring(lastEnd, matcher.end()).trim(); if (!segment.isEmpty()) { target.add(createEvent("thought", "reasoning", segment, null)); } lastEnd = matcher.end(); } reasoningBuffer.delete(0, lastEnd); if (reasoningBuffer.length() > REASONING_FLUSH_THRESHOLD) { target.add(createEvent("thought", "reasoning", reasoningBuffer.toString(), null)); reasoningBuffer.setLength(0); } } }

触发策略：

遇到句号/问号/感叹号/换行 → 按标点切分，整句发出
缓冲区积累超过 50 字符无标点 → 强制整块发出
切换到文本输出或工具调用 → 排空缓存

六、效果对比

改造前：前端收到逐个单词的thought事件：

event: thought data: {"content":"The","role":"thought"} event: thought data: {"content":"user","role":"thought"} event: thought data: {"content":"wants","role":"thought"}

改造后：前端收到完整的语义段落：

event: thought data: {"content":"The user wants me to add a new feature.","role":"thought"} event: thought data: {"content":"Let me think about the best approach.","role":"thought"}