fix snippets

Author: hanouticelina

docs/api-inference/tasks/chat-completion.md

@@ -79,8 +79,8 @@ curl 'https://api-inference.huggingface.co/models/google/gemma-2-2b-it/v1/chat/c
 </curl>
 
 <python>
-<huggingface_hub>
 ```py
+<huggingface_hub>
 from huggingface_hub import InferenceClient
 
 client = InferenceClient(api_key="hf_***")
@@ -101,11 +101,9 @@ stream = client.chat.completions.create(
 
 for chunk in stream:
     print(chunk.choices[0].delta.content, end="")
-```
 </huggingface_hub>
 
 <openai>
-```py
 from openai import OpenAI
 
 client = OpenAI(
@@ -129,15 +127,15 @@ stream = client.chat.completions.create(
 
 for chunk in stream:
     print(chunk.choices[0].delta.content, end="")
-```
 </openai>
+```
 
 To use the Python client, see `huggingface_hub`'s [package reference](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client#huggingface_hub.InferenceClient.chat_completion).
 </python>
 
 <js>
-<huggingface_hub>
 ```js
+<huggingface_hub>
 import { HfInference } from "@huggingface/inference"
 
 const client = new HfInference("hf_***")
@@ -162,11 +160,9 @@ for await (const chunk of stream) {
 		console.log(newContent);
 	}  
 }
-```
 </huggingface_hub>
 
 <openai>
-```js
 import { OpenAI } from "openai"
 
 const client = new OpenAI({
@@ -195,8 +191,8 @@ for await (const chunk of stream) {
 		console.log(newContent);
 	}  
 }
-```
 </openai>
+```
 
 To use the JavaScript client, see `huggingface.js`'s [package reference](https://huggingface.co/docs/huggingface.js/inference/classes/HfInference#chatcompletion).
 </js>
@@ -241,8 +237,8 @@ curl 'https://api-inference.huggingface.co/models/meta-llama/Llama-3.2-11B-Visio
 </curl>
 
 <python>
-<huggingface_hub>
 ```py
+<huggingface_hub>
 from huggingface_hub import InferenceClient
 
 client = InferenceClient(api_key="hf_***")
@@ -274,11 +270,9 @@ stream = client.chat.completions.create(
 
 for chunk in stream:
     print(chunk.choices[0].delta.content, end="")
-```
 </huggingface_hub>
 
 <openai>
-```py
 from openai import OpenAI
 
 client = OpenAI(
@@ -313,15 +307,15 @@ stream = client.chat.completions.create(
 
 for chunk in stream:
     print(chunk.choices[0].delta.content, end="")
-```
 </openai>
+```
 
 To use the Python client, see `huggingface_hub`'s [package reference](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client#huggingface_hub.InferenceClient.chat_completion).
 </python>
 
 <js>
-<huggingface_hub>
 ```js
+<huggingface_hub>
 import { HfInference } from "@huggingface/inference"
 
 const client = new HfInference("hf_***")
@@ -357,11 +351,9 @@ for await (const chunk of stream) {
 		console.log(newContent);
 	}  
 }
-```
 </huggingface_hub>
 
 <openai>
-```js
 import { OpenAI } from "openai"
 
 const client = new OpenAI({
@@ -401,8 +393,8 @@ for await (const chunk of stream) {
 		console.log(newContent);
 	}  
 }
-```
 </openai>
+```
 
 To use the JavaScript client, see `huggingface.js`'s [package reference](https://huggingface.co/docs/huggingface.js/inference/classes/HfInference#chatcompletion).
 </js>

docs/api-inference/tasks/image-text-to-text.md

@@ -45,8 +45,8 @@ curl https://api-inference.huggingface.co/models/meta-llama/Llama-3.2-11B-Vision
 </curl>
 
 <python>
-<huggingface_hub>
 ```py
+<huggingface_hub>
 import requests
 
 API_URL = "https://api-inference.huggingface.co/models/meta-llama/Llama-3.2-11B-Vision-Instruct"
@@ -67,11 +67,9 @@ stream = client.chat.completions.create(
 
 for chunk in stream:
     print(chunk.choices[0].delta.content, end="")
-```
 </huggingface_hub>
 
 <openai>
-```py
 import requests
 
 API_URL = "https://api-inference.huggingface.co/models/meta-llama/Llama-3.2-11B-Vision-Instruct"
@@ -95,8 +93,8 @@ stream = client.chat.completions.create(
 
 for chunk in stream:
     print(chunk.choices[0].delta.content, end="")
-```
 </openai>
+```
 
 To use the Python client, see `huggingface_hub`'s [package reference](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client#huggingface_hub.InferenceClient.image_text-to-text).
 </python>

scripts/api-inference/scripts/generate.ts

@@ -100,16 +100,16 @@ const TASKS_DATA = (await response.json()) as any;
 ///////////////////////
 
 const formatSnippets = (result: snippets.types.InferenceSnippet | snippets.types.InferenceSnippet[], defaultClient: string, language: string): string => {
-  // For single snippet, return just the content (let the template handle the wrapping)
+  // For single snippet, return just the content
   if (!Array.isArray(result) || result.length === 1) {
     const snippet = Array.isArray(result) ? result[0] : result;
-    return `\`\`\`${language}\n${snippet.content}\n\`\`\``;
+    return snippet.content;
   }
 
-  // For multiple snippets, include the client tags
+  // For multiple snippets, return just the content with client info
   return result
     .map(snippet => 
-      `<${snippet.client || defaultClient}>\n\`\`\`${language}\n${snippet.content}\n\`\`\`\n</${snippet.client || defaultClient}>`
+      `<${snippet.client || defaultClient}>\n${snippet.content}\n</${snippet.client || defaultClient}>`
     )
     .join('\n\n');
 };

scripts/api-inference/templates/common/snippets-template.handlebars

@@ -5,14 +5,18 @@
 {{!-- cURL snippet (if exists) --}}
 {{#if taskSnippets.curl}}
 <curl>
+```bash
 {{{taskSnippets.curl}}}
+```
 </curl>
 {{/if}}
 
 {{!-- Python snippet (if exists) --}}
 {{#if taskSnippets.python}}
 <python>
+```py
 {{{taskSnippets.python}}}
+```
 
 To use the Python client, see `huggingface_hub`'s [package reference](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client#huggingface_hub.InferenceClient.{{taskSnakeCase}}).
 </python>
@@ -21,7 +25,9 @@ To use the Python client, see `huggingface_hub`'s [package reference](https://hu
 {{!-- JavaScript snippet (if exists) --}}
 {{#if taskSnippets.javascript}}
 <js>
+```js
 {{{taskSnippets.javascript}}}
+```
 
 To use the JavaScript client, see `huggingface.js`'s [package reference](https://huggingface.co/docs/huggingface.js/inference/classes/HfInference#{{taskAttached}}).
 </js>