@ai-sdk/deepseek 2.0.8 → 2.0.10

package/docs/30-deepseek.mdx

@@ -0,0 +1,160 @@
+---
+title: DeepSeek
+description: Learn how to use DeepSeek's models with the AI SDK.
+---
+
+# DeepSeek Provider
+
+The [DeepSeek](https://www.deepseek.com) provider offers access to powerful language models through the DeepSeek API.
+
+API keys can be obtained from the [DeepSeek Platform](https://platform.deepseek.com/api_keys).
+
+## Setup
+
+The DeepSeek provider is available via the `@ai-sdk/deepseek` module. You can install it with:
+
+<Tabs items={['pnpm', 'npm', 'yarn', 'bun']}>
+  <Tab>
+    <Snippet text="pnpm add @ai-sdk/deepseek" dark />
+  </Tab>
+  <Tab>
+    <Snippet text="npm install @ai-sdk/deepseek" dark />
+  </Tab>
+  <Tab>
+    <Snippet text="yarn add @ai-sdk/deepseek" dark />
+  </Tab>
+  <Tab>
+    <Snippet text="bun add @ai-sdk/deepseek" dark />
+  </Tab>
+</Tabs>
+
+## Provider Instance
+
+You can import the default provider instance `deepseek` from `@ai-sdk/deepseek`:
+
+```ts
+import { deepseek } from '@ai-sdk/deepseek';
+```
+
+For custom configuration, you can import `createDeepSeek` and create a provider instance with your settings:
+
+```ts
+import { createDeepSeek } from '@ai-sdk/deepseek';
+
+const deepseek = createDeepSeek({
+  apiKey: process.env.DEEPSEEK_API_KEY ?? '',
+});
+```
+
+You can use the following optional settings to customize the DeepSeek provider instance:
+
+- **baseURL** _string_
+
+  Use a different URL prefix for API calls.
+  The default prefix is `https://api.deepseek.com/v1`.
+
+- **apiKey** _string_
+
+  API key that is being sent using the `Authorization` header. It defaults to
+  the `DEEPSEEK_API_KEY` environment variable.
+
+- **headers** _Record&lt;string,string&gt;_
+
+  Custom headers to include in the requests.
+
+- **fetch** _(input: RequestInfo, init?: RequestInit) => Promise&lt;Response&gt;_
+
+  Custom [fetch](https://developer.mozilla.org/en-US/docs/Web/API/fetch) implementation.
+
+## Language Models
+
+You can create language models using a provider instance:
+
+```ts
+import { deepseek } from '@ai-sdk/deepseek';
+import { generateText } from 'ai';
+
+const { text } = await generateText({
+  model: deepseek('deepseek-chat'),
+  prompt: 'Write a vegetarian lasagna recipe for 4 people.',
+});
+```
+
+You can also use the `.chat()` or `.languageModel()` factory methods:
+
+```ts
+const model = deepseek.chat('deepseek-chat');
+// or
+const model = deepseek.languageModel('deepseek-chat');
+```
+
+DeepSeek language models can be used in the `streamText` function
+(see [AI SDK Core](/docs/ai-sdk-core)).
+
+### Reasoning
+
+DeepSeek has reasoning support for the `deepseek-reasoner` model. The reasoning is exposed through streaming:
+
+```ts
+import { deepseek } from '@ai-sdk/deepseek';
+import { streamText } from 'ai';
+
+const result = streamText({
+  model: deepseek('deepseek-reasoner'),
+  prompt: 'How many "r"s are in the word "strawberry"?',
+});
+
+for await (const part of result.fullStream) {
+  if (part.type === 'reasoning') {
+    // This is the reasoning text
+    console.log('Reasoning:', part.text);
+  } else if (part.type === 'text') {
+    // This is the final answer
+    console.log('Answer:', part.text);
+  }
+}
+```
+
+See [AI SDK UI: Chatbot](/docs/ai-sdk-ui/chatbot#reasoning) for more details
+on how to integrate reasoning into your chatbot.
+
+### Cache Token Usage
+
+DeepSeek provides context caching on disk technology that can significantly reduce token costs for repeated content. You can access the cache hit/miss metrics through the `providerMetadata` property in the response:
+
+```ts
+import { deepseek } from '@ai-sdk/deepseek';
+import { generateText } from 'ai';
+
+const result = await generateText({
+  model: deepseek('deepseek-chat'),
+  prompt: 'Your prompt here',
+});
+
+console.log(result.providerMetadata);
+// Example output: { deepseek: { promptCacheHitTokens: 1856, promptCacheMissTokens: 5 } }
+```
+
+The metrics include:
+
+- `promptCacheHitTokens`: Number of input tokens that were cached
+- `promptCacheMissTokens`: Number of input tokens that were not cached
+
+<Note>
+  For more details about DeepSeek's caching system, see the [DeepSeek caching
+  documentation](https://api-docs.deepseek.com/guides/kv_cache#checking-cache-hit-status).
+</Note>
+
+## Model Capabilities
+
+| Model               | Text Generation     | Object Generation   | Image Input         | Tool Usage          | Tool Streaming      |
+| ------------------- | ------------------- | ------------------- | ------------------- | ------------------- | ------------------- |
+| `deepseek-chat`     | <Check size={18} /> | <Check size={18} /> | <Cross size={18} /> | <Check size={18} /> | <Check size={18} /> |
+| `deepseek-reasoner` | <Check size={18} /> | <Check size={18} /> | <Cross size={18} /> | <Check size={18} /> | <Check size={18} /> |
+
+<Note>
+  Please see the [DeepSeek
+  docs](https://api-docs.deepseek.com/quick_start/pricing) for a full list of
+  available models. You can also pass any available provider model ID as a
+  string if needed.
+</Note>

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ai-sdk/deepseek",
-  "version": "2.0.8",
+  "version": "2.0.10",
   "license": "Apache-2.0",
   "sideEffects": false,
   "main": "./dist/index.js",
@@ -8,9 +8,14 @@
   "types": "./dist/index.d.ts",
   "files": [
     "dist/**/*",
+    "docs/**/*",
+    "src",
     "CHANGELOG.md",
     "README.md"
   ],
+  "directories": {
+    "doc": "./docs"
+  },
   "exports": {
     "./package.json": "./package.json",
     ".": {
@@ -28,7 +33,7 @@
     "tsup": "^8",
     "typescript": "5.8.3",
     "zod": "3.25.76",
-    "@ai-sdk/test-server": "1.0.1",
+    "@ai-sdk/test-server": "1.0.2",
     "@vercel/ai-tsconfig": "0.0.0"
   },
   "peerDependencies": {
@@ -54,7 +59,7 @@
   "scripts": {
     "build": "pnpm clean && tsup --tsconfig tsconfig.build.json",
     "build:watch": "pnpm clean && tsup --watch",
-    "clean": "del-cli dist *.tsbuildinfo",
+    "clean": "del-cli dist docs *.tsbuildinfo",
     "lint": "eslint \"./**/*.ts*\"",
     "type-check": "tsc --build",
     "prettier-check": "prettier --check \"./**/*.ts*\"",

package/src/chat/__fixtures__/deepseek-json.json

@@ -0,0 +1,32 @@
+{
+  "id": "f03bc170-b375-4561-9685-35182c8152c5",
+  "object": "chat.completion",
+  "created": 1764681341,
+  "model": "deepseek-reasoner",
+  "choices": [
+    {
+      "index": 0,
+      "message": {
+        "role": "assistant",
+        "content": "{\n  \"location\": \"San Francisco\",\n  \"condition\": \"cloudy\",\n  \"temperature\": 7\n}",
+        "reasoning_content": "I have the result from the weather tool. It returned a JSON object with location, condition, and temperature. I need to reply with JSON object ONLY. The assistant's response should be the JSON output. So I should output exactly that JSON? Or perhaps wrap it in some way? The instruction says \"Reply with JSON object ONLY.\" Probably I should output the result as is, or maybe a structured response. But the tool result is already JSON. I think I should output that JSON directly. However, I need to ensure it's a valid response. Let me output the JSON object."
+      },
+      "logprobs": null,
+      "finish_reason": "stop"
+    }
+  ],
+  "usage": {
+    "prompt_tokens": 495,
+    "completion_tokens": 144,
+    "total_tokens": 639,
+    "prompt_tokens_details": {
+      "cached_tokens": 320
+    },
+    "completion_tokens_details": {
+      "reasoning_tokens": 118
+    },
+    "prompt_cache_hit_tokens": 320,
+    "prompt_cache_miss_tokens": 175
+  },
+  "system_fingerprint": "fp_eaab8d114b_prod0820_fp8_kvcache"
+}