[DRAFT] docs(genapi): add hugging face library

menu/navigation.json

@@ -987,6 +987,10 @@
                   {
                     "label": "Use function calling",
                     "slug": "use-function-calling"
+                  },
+                  {
+                    "label": "Connect through Hugging Face library",
+                    "slug": "connect-through-hugging-face-library"
                   }
                 ],
                 "label": "How to",

pages/generative-apis/how-to/connect-through-hugging-face-library.mdx

@@ -0,0 +1,108 @@
+---
+meta:
+  title: How to connect using Hugging Face library
+  description: Learn how to interact with Generative APIs using the Hugging Face library
+content:
+  h1: How to Connect using Hugging Face library
+  paragraph: Learn how to interact with Generative APIs using the Hugging Face library.
+tags: generative-apis hugging-face library
+dates:
+  validation: 2025-04-01
+  posted: 2025-04-01
+---
+<Macro id="requirements" />
+
+- A Scaleway account logged into the [console](https://console.scaleway.com)
+- [Owner](/iam/concepts/#owner) status or [IAM permissions](/iam/concepts/#permission) allowing you to perform actions in the intended Organization
+- A valid [API key](/iam/how-to/create-api-keys/) for API authentication
+- Node.js installed on your local machine
+- Scaleway credentials or Hugging Face credentials with the proper access rights (two methods of connection are available)
+
+## Steps to Connect Using Hugging Face Library
+
+1. Create a new directory on your local machine where you will store all your project files.
+
+2. Open a terminal in your project directory and run the following command to install the Hugging Face inference library:
+     ```bash
+     npm install @huggingface/inference
+     ```
+
+3. Create a new file named `main.js` in your project directory and add the following code to it:
+     ```js
+     import { InferenceClient } from '@huggingface/inference';
+
+     const client = new InferenceClient({ apiKey: process.env.SCW_SECRET_KEY });
+
+     const out = await client.chatCompletion({
+       provider: "scaleway",
+       // endpointUrl is not supported with third-party providers
+       // endpointUrl: "https://api.scaleway.ai/b409cb09-756c-430f-a8e8-748f88ef4bad",
+       // model: "meta-llama/Meta-Llama-3-8B-Instruct",
+       model: "meta-llama/Llama-3.3-70B-Instruct",
+       messages: [{ role: "user", content: "Tell me about Scaleway." }],
+       max_tokens: 512,
+       temperature: 0.1,
+     });
+
+     console.log(out.choices[0].message.content);
+     ```
+
+4. Execute the script by running the following command in your terminal:
+     ```bash
+     node main.js
+     ```
+     The model's response should be displayed in your terminal.
+
+### Using stream completion
+
+To use stream completion, you can modify your script as follows:
+```js
+import { InferenceClient } from '@huggingface/inference';
+
+const client = new InferenceClient({ apiKey: process.env.SCW_SECRET_KEY });
+
+for await (const chunk of client.chatCompletionStream({
+  model: "meta-llama/Llama-3.3-70B-Instruct",
+  provider: "scaleway",
+  messages: [{ role: "user", content: "Tell me about Scaleway." }],
+  max_tokens: 512,
+})) {
+  console.log(chunk.choices[0].delta.content);
+}
+```
+
+### Using Hugging Face tokens
+
+You can also authenticate using Hugging Face tokens. Set the `HF_TOKEN` environment variable and modify your script slightly:
+```js
+import { InferenceClient } from '@huggingface/inference';
+
+const client = new InferenceClient({ apiKey: process.env.HF_TOKEN });
+
+const out = await client.chatCompletion({
+  provider: "scaleway",
+  model: "meta-llama/Llama-3.3-70B-Instruct",
+  messages: [{ role: "user", content: "Tell me about Scaleway." }],
+  max_tokens: 512,
+  temperature: 0.1,
+});
+
+console.log(out.choices[0].message.content);
+```
+
+In some cases, providing the token directly in the `InferenceClient` constructor might not be necessary if the environment variable is set correctly:
+```js
+import { InferenceClient } from '@huggingface/inference';
+
+const client = new InferenceClient();
+
+const out = await client.chatCompletion({
+  provider: "scaleway",
+  model: "meta-llama/Llama-3.3-70B-Instruct",
+  messages: [{ role: "user", content: "Tell me about Scaleway." }],
+  max_tokens: 512,
+  temperature: 0.1,
+});
+
+console.log(out.choices[0].message.content);
+```