withcatai
diff --git a/‎.github/workflows/build.yml
Copy file name to clipboard
+2Lines changed: 2 additions & 0 deletions b/‎.github/workflows/build.yml
Copy file name to clipboard
+2Lines changed: 2 additions & 0 deletions
diff --git a/‎.vitepress/components/YouTubePlayer/YouTubePlayer.vue
Copy file name to clipboard
+53Lines changed: 53 additions & 0 deletions b/‎.vitepress/components/YouTubePlayer/YouTubePlayer.vue
Copy file name to clipboard
+53Lines changed: 53 additions & 0 deletions
diff --git a/‎.vitepress/config/apiReferenceSidebar.ts
Copy file name to clipboard
+1Lines changed: 1 addition & 0 deletions b/‎.vitepress/config/apiReferenceSidebar.ts
Copy file name to clipboard
+1Lines changed: 1 addition & 0 deletions
diff --git a/‎.vitepress/theme/LayoutContainer.vue
Copy file name to clipboard
+3-4Lines changed: 3 additions & 4 deletions b/‎.vitepress/theme/LayoutContainer.vue
Copy file name to clipboard
+3-4Lines changed: 3 additions & 4 deletions
diff --git a/‎.vitepress/theme/index.ts
Copy file name to clipboard
+5-3Lines changed: 5 additions & 3 deletions b/‎.vitepress/theme/index.ts
Copy file name to clipboard
+5-3Lines changed: 5 additions & 3 deletions
diff --git a/‎.vitepress/theme/style.css
Copy file name to clipboard
+9Lines changed: 9 additions & 0 deletions b/‎.vitepress/theme/style.css
Copy file name to clipboard
+9Lines changed: 9 additions & 0 deletions
diff --git a/‎README.md
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions b/‎README.md
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions
diff --git a/‎assets/electronTemplate.icon.png
Copy file name to clipboard
764 KB b/‎assets/electronTemplate.icon.png
Copy file name to clipboard
764 KB
diff --git a/‎docs/blog/v3.6-deepseek-r1.md
Copy file name to clipboard
+129Lines changed: 129 additions & 0 deletions b/‎docs/blog/v3.6-deepseek-r1.md
Copy file name to clipboard
+129Lines changed: 129 additions & 0 deletions
diff --git a/‎docs/blog/v3.md
Copy file name to clipboard
+3-3Lines changed: 3 additions & 3 deletions b/‎docs/blog/v3.md
Copy file name to clipboard
+3-3Lines changed: 3 additions & 3 deletions
@@ -589,6 +589,8 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
+        with:
+          lfs: true
       - uses: actions/setup-node@v4
         with:
           node-version: "20"
 
@@ -0,0 +1,53 @@
+<script setup lang="ts">
+import {computed} from "vue";
+const props = withDefaults(defineProps<{
+    id: string,
+    autoplay?: boolean,
+    controls?: boolean
+}>(), {
+    autoplay: false,
+    controls: true
+});
+
+const url = computed(() => {
+    const res = new URL(`https://www.youtube.com/embed/${props.id}`);
+
+    if (typeof location !== "undefined")
+        res.searchParams.set("origin", location.origin);
+
+    res.searchParams.set("autoplay", props.autoplay ? "1" : "0");
+    res.searchParams.set("controls", props.controls ? "1" : "0");
+    res.searchParams.set("playsinline", "1");
+    res.searchParams.set("rel", "0");
+
+    return res.href;
+});
+</script>
+
+<template>
+    <div class="youtubePlayer" v-if="url">
+        <iframe
+            class="player"
+            :src="url"
+            allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture"
+            allowfullscreen
+        />
+    </div>
+</template>
+
+<style scoped>
+.youtubePlayer {
+    margin-top: 1em;
+    overflow: hidden;
+    border-radius: 12px;
+    background-color: color-mix(in srgb, var(--vp-c-text-1) 6%, transparent);
+
+    > .player {
+        width: 100%;
+        aspect-ratio: 16 / 9;
+        min-height: 240px;
+        max-height: 600px;
+        border: none;
+    }
+}
+</style>
@@ -44,6 +44,7 @@ const chatWrappersOrder = [
     "GeneralChatWrapper",
     "TemplateChatWrapper",
     "JinjaTemplateChatWrapper",
+    "DeepSeekChatWrapper",
     "Llama3_1ChatWrapper",
     "Llama3ChatWrapper",
     "Llama2ChatWrapper",
 
@@ -16,10 +16,9 @@ if (typeof document !== "undefined")
     document?.documentElement.classList.toggle("theme-transition", themeTransitionEnabled);
 
 function updateIsBlogPage() {
-    document?.documentElement.classList.toggle(
-        "blog-page",
-        route.path !== "/blog/" && route.path.startsWith("/blog/")
-    );
+    const blogIndex = route.path === "/blog/";
+    document?.documentElement.classList.toggle("blog-page", !blogIndex && route.path.startsWith("/blog/"));
+    document?.documentElement.classList.toggle("blog-index", blogIndex);
 }
 
 watch(() => route.path, updateIsBlogPage);
 
@@ -9,6 +9,7 @@ import LatestVersionHomeBadge from "../components/LatestVersionHomeBadge/LatestV
 import CommentsSection from "../components/CommentsSection/CommentsSection.vue";
 import {NolebaseGitChangelogPlugin} from "@nolebase/vitepress-plugin-git-changelog/client";
 import LayoutContainer from "./LayoutContainer.vue";
+import YouTubePlayer from "../components/YouTubePlayer/YouTubePlayer.vue";
 
 import "./style.css";
 import "@nolebase/vitepress-plugin-git-changelog/client/style.css";
@@ -18,9 +19,9 @@ import type {EnhanceAppContext} from "vitepress";
 export default {
     extends: Theme,
     Layout: () => {
-        const text = "v3.0 is here!";
-        const link = "/blog/v3";
-        const hideDate = new Date("2025-01-01T00:00:00Z");
+        const text = "DeepSeek R1 is here!";
+        const link = "/blog/v3.6-deepseek-r1";
+        const hideDate = new Date("2025-06-01T00:00:00Z");
 
         return h(LayoutContainer, null, h(Theme.Layout, null, {
             "home-hero-info-before": () => h(LatestVersionHomeBadge, {
@@ -35,6 +36,7 @@ export default {
         }));
     },
     enhanceApp({app, router, siteData}: EnhanceAppContext) {
+        app.component("YouTubePlayer", YouTubePlayer);
         app.use(TwoslashFloatingVue);
         app.use(NolebaseGitChangelogPlugin, {
             displayAuthorsInsideCommitLine: true,
 
@@ -562,6 +562,15 @@ html.blog-page .vp-doc>div>hr:first-of-type {
     display: none;
 }
 
+@media (min-width: 960px) {
+    html.blog-index .VPDoc:not(.has-sidebar)>.container>.content,
+    html.blog-page .VPDoc:not(.has-sidebar)>.container>.content {
+        max-width: 896px;
+        padding-left: 0px;
+        padding-right: 0px;
+    }
+}
+
 /*#VPContent {*/
 /*    background-image: radial-gradient(1200px 380px at 50% 0%, color-mix(in srgb, var(--vp-c-brand-1) 32%, transparent), transparent 64%);*/
 /*}*/
 
@@ -15,7 +15,7 @@
 
 </div>
 
-✨ [`v3.0` is here!](https://node-llama-cpp.withcat.ai/blog/v3) ✨
+✨ [DeepSeek R1 is here!](https://node-llama-cpp.withcat.ai/blog/v3.6-deepseek-r1) ✨
 
 ## Features
 * Run LLMs locally on your machine
@@ -94,7 +94,7 @@ console.log("AI: " + a2);
 To contribute to `node-llama-cpp` read the [contribution guide](https://node-llama-cpp.withcat.ai/guide/contributing).
 
 ## Acknowledgements
-* llama.cpp: [ggerganov/llama.cpp](https://github.com/ggerganov/llama.cpp)
+* llama.cpp: [ggml-org/llama.cpp](https://github.com/ggml-org/llama.cpp)
 
 
 <br />
 
@@ -0,0 +1,129 @@
+---
+title: DeepSeek R1 with function calling
+date: 2025-02-21T15:00:00Z
+lastUpdated: false
+author:
+    name: Gilad S.
+    github: giladgd
+category: Release
+description: node-llama-cpp v3.6 is here, with full support for DeepSeek R1, including function calling!
+image:
+    url: https://github.com/user-attachments/assets/9ed954f8-102d-4cdd-96d8-9b6710b8a1f5
+    alt: "node-llama-cpp + DeepSeek R1"
+    width: 3072
+    height: 1536
+---
+[`node-llama-cpp`](https://node-llama-cpp.withcat.ai) v3.6 is here, with full support for [DeepSeek R1](https://github.com/deepseek-ai/DeepSeek-R1), including function calling!
+
+---
+
+## Function Calling
+`node-llama-cpp` includes [many tricks](../guide/function-calling) used to make function calling work with most models.
+This release includes special adaptations for DeepSeek R1 to improve function calling performance and stability.
+
+Here's a basic example of function calling with DeepSeek R1:
+```typescript
+import {fileURLToPath} from "url";
+import path from "path";
+import {
+    getLlama, LlamaChatSession, defineChatSessionFunction, resolveModelFile
+} from "node-llama-cpp";
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const modelsDir = path.join(__dirname, "..", "models");
+
+const modelUri = "hf:mradermacher/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M";
+
+
+const llama = await getLlama();
+const model = await llama.loadModel({
+    modelPath: await resolveModelFile(modelUri, modelsDir)
+});
+const context = await model.createContext();
+const session = new LlamaChatSession({
+    contextSequence: context.getSequence()
+});
+
+const fruitPrices: Record<string, string> = {
+    "apple": "$6",
+    "banana": "$4"
+};
+const functions = {
+    getFruitPrice: defineChatSessionFunction({
+        description: "Get the price of a fruit",
+        params: {
+            type: "object",
+            properties: {
+                name: {
+                    type: "string"
+                }
+            }
+        },
+        async handler(params) {
+            const name = params.name.toLowerCase();
+            if (Object.keys(fruitPrices).includes(name))
+                return {
+                    name: name,
+                    price: fruitPrices[name]
+                };
+
+            return `Unrecognized fruit "${params.name}"`;
+        }
+    })
+};
+
+
+const q1 = "Is an apple more expensive than a banana?";
+console.log("User: " + q1);
+
+const a1 = await session.prompt(q1, {functions});
+console.log("AI: " + a1.trim());
+```
+
+
+## Recommended Models
+Here are some recommended model URIs you can use to try out DeepSeek R1 with function calling.
+
+| Model                                                                                                   | Size   | URI                                                         |
+|---------------------------------------------------------------------------------------------------------|--------|-------------------------------------------------------------|
+| [DeepSeek R1 Distill Qwen 7B](https://huggingface.co/mradermacher/DeepSeek-R1-Distill-Qwen-7B-GGUF)     | 4.68GB | `hf:mradermacher/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M`   |
+| [DeepSeek R1 Distill Qwen 14B](https://huggingface.co/mradermacher/DeepSeek-R1-Distill-Qwen-14B-GGUF)   | 8.99GB | `hf:mradermacher/DeepSeek-R1-Distill-Qwen-14B-GGUF:Q4_K_M`  |
+| [DeepSeek R1 Distill Qwen 32B](https://huggingface.co/mradermacher/DeepSeek-R1-Distill-Qwen-32B-GGUF)   | 19.9GB | `hf:mradermacher/DeepSeek-R1-Distill-Qwen-32B-GGUF:Q4_K_M`  |
+
+::: info TIP
+Estimate the compatibility of a model with your machine before downloading it using the [`inspect estimate`](../cli/inspect/estimate.md) command:
+```shell
+npx -y node-llama-cpp inspect estimate <model URI>
+```
+:::
+
+### Try It Using the CLI
+To try out function calling with a given model using the CLI, you can use the [`chat` command](../cli/chat.md) with the `--ef` flag
+to provide the model with date and time functions:
+
+```shell
+npx -y node-llama-cpp chat --ef --prompt "What is the time?" <model URI>
+```
+
+
+## Chain of Thought Segmentation
+The thoughts generated by a reasoning model are now [separated into `thought` segments](../guide/chat-session.md#stream-response-segments) in the response,
+so you can choose whether to use them or not.
+
+By default, the [`.prompt(...)`](../api/classes/LlamaChatSession#prompt) method returns only the main response, without any `thought` segments.
+Use the [`.promptWithMeta(...)`](../api/classes/LlamaChatSession#promptwithmeta) method to get the full response.
+
+You can use the new [`onResponseChunk`](../api/type-aliases/LLamaChatPromptOptions.md#onresponsechunk) option to [stream `thought` segments as they are being generated](../guide/chat-session.md#stream-response-segments).
+
+
+## Electron App Template
+The [Electron app template](../guide/electron.md) has been updated to properly segment the thoughts in the response.
+
+Try it out by downloading the latest build [from GitHub](https://github.com/withcatai/node-llama-cpp/releases/latest),
+or by [scaffolding a new project](../guide/index.md#scaffold-new-project) based on the Electron template:
+
+```shell
+npm create node-llama-cpp@latest
+```
+
+<YouTubePlayer id="IqfMs0lfIvQ" />
@@ -15,7 +15,7 @@ image:
 ---
 [`node-llama-cpp`](https://node-llama-cpp.withcat.ai) 3.0 is finally here.
 
-With [`node-llama-cpp`](https://node-llama-cpp.withcat.ai), you can run large language models locally on your machine using the power of [`llama.cpp`](https://github.com/ggerganov/llama.cpp) with a simple and easy-to-use API.
+With [`node-llama-cpp`](https://node-llama-cpp.withcat.ai), you can run large language models locally on your machine using the power of [`llama.cpp`](https://github.com/ggml-org/llama.cpp) with a simple and easy-to-use API.
 
 It includes everything you need, from downloading models, to running them in the most optimized way for your hardware, and integrating them in your projects.
 
@@ -43,7 +43,7 @@ While `llama.cpp` is an amazing project, it's also highly technical and can be c
 `node-llama-cpp` bridge that gap, making `llama.cpp` accessible to everyone, regardless of their experience level.
 
 ### Performance
-[`node-llama-cpp`](https://node-llama-cpp.withcat.ai) is built on top of [`llama.cpp`](https://github.com/ggerganov/llama.cpp), a highly optimized C++ library for running large language models.
+[`node-llama-cpp`](https://node-llama-cpp.withcat.ai) is built on top of [`llama.cpp`](https://github.com/ggml-org/llama.cpp), a highly optimized C++ library for running large language models.
 
 `llama.cpp` supports many compute backends, including Metal, CUDA, and Vulkan. It also uses [Accelerate](https://developer.apple.com/accelerate/) on Mac.
 
@@ -116,7 +116,7 @@ npx -y node-llama-cpp chat
 Check out the [getting started guide](../guide/index.md) to learn how to use `node-llama-cpp`.
 
 ## Thank You
-`node-llama-cpp` is only possible thanks to the amazing work done on [`llama.cpp`](https://github.com/ggerganov/llama.cpp) by [Georgi Gerganov](https://github.com/ggerganov), [Slaren](https://github.com/slaren) and all the contributors from the community.
+`node-llama-cpp` is only possible thanks to the amazing work done on [`llama.cpp`](https://github.com/ggml-org/llama.cpp) by [Georgi Gerganov](https://github.com/ggerganov), [Slaren](https://github.com/slaren) and all the contributors from the community.
 
 ## What's next?
 Version 3.0 is a major milestone, but there's plenty more planned for the future.