Merge pull request #357 from mistralai/doc/v0.1.22

Update docs to v0.1.22

Author: jaccolor2

public/img/food-receipt.jpg

@@ -6,6 +6,9 @@ You specify the libraries that the agent has access to with `library_ids`, you c
 
 :::tip
 It is also possible to specify libraries created via Le Chat; these IDs are visible in the URL of the corresponding library created on Le Chat, for example: `https://chat.mistral.ai/libraries/<library_id>`; To enable the Agent to access Le Chat library, you have to be an Org admin and share it with the Organization.
+
+The opposite is also possible, you can create a library via API and share it with your team on Le Chat.  
+Find more on how to share and manage libraries [here](?tab=manage-libraries#explorer-tabs-agent_creation_and_library_management).
 :::
 
 <Tabs groupId="code">
@@ -114,3 +117,5 @@ curl --location "https://api.mistral.ai/v1/agents" \
 </Tabs>
 
 As with other agents, when creating one, you will receive an agent ID corresponding to the created agent. You can use this ID to start a conversation.
+
+Next, learn how you can manage your libraries and agents in the **[Manage Libraries](?tab=manage-libraries#explorer-tabs-agent_creation_and_library_management)** tab.

public/img/old-page.jpg

@@ -12,15 +12,15 @@ The content can be either a string, the most standard usage of llms:
 
 <Tabs groupId="code">
     <TabItem value="python" label="python" default>
-        <Tabs groupId="syn">
+        <Tabs groupId="sync">
             <TabItem value="sync" label="Synchronous" default>
 
 ```python
 import os
 from mistralai import Mistral
 
 api_key = os.environ["MISTRAL_API_KEY"]
-model = "mistral-medium-latest"
+model = "mistral-large-latest"
 
 client = Mistral(api_key=api_key)
 
@@ -29,7 +29,7 @@ chat_response = client.chat.complete(
     messages = [
         {
             "role": "user",
-            "content": "What is the best French cheese?",
+            "content": "How far is the moon from earth?",
         },
     ]
 )
@@ -47,13 +47,13 @@ from mistralai.models import UserMessage
 
 async def main():
     api_key = os.environ["MISTRAL_API_KEY"]
-    model = "mistral-medium-latest"
+    model = "mistral-large-latest"
 
     client = Mistral(api_key=api_key)
 
     chat_response = await client.chat.complete_async(
         model=model,
-        messages=[UserMessage(content="What is the best French cheese?")],
+        messages=[UserMessage(content="How far is the moon from earth?")],
     )
 
 if __name__ == "__main__":
@@ -77,8 +77,8 @@ const client = new Mistral({apiKey: apiKey});
 
 async function main() {
     const chatResponse = await client.chat.complete({
-        model: "mistral-medium-latest",
-        messages: [{role: 'user', content: 'What is the best French cheese?'}]
+        model: "mistral-large-latest",
+        messages: [{role: 'user', content: 'How far is the moon from earth?'}]
     });
 }
 
@@ -94,11 +94,11 @@ curl --location "https://api.mistral.ai/v1/chat/completions" \
      --header 'Accept: application/json' \
      --header "Authorization: Bearer $MISTRAL_API_KEY" \
      --data '{
-    "model": "mistral-medium-latest",
+    "model": "mistral-large-latest",
     "messages": [
      {
         "role": "user",
-        "content": "What is the best French cheese?"
+        "content": "How far is the moon from earth?"
       }
     ]
   }'
@@ -109,24 +109,25 @@ curl --location "https://api.mistral.ai/v1/chat/completions" \
 
 ```json
 {
-  "id": "7c163b08d98e4eebb35564ae14ffe6e4",
-  "created": 1756130273,
-  "model": "mistral-medium-latest",
+  "id": "bed9481e857e40fab7c7eea4232f294a",
+  "object": "chat.completion",
+  "model": "mistral-large-latest",
   "usage": {
-    "prompt_tokens": 10,
-    "total_tokens": 1160,
-    "completion_tokens": 1150
+    "prompt_tokens": 11,
+    "completion_tokens": 268,
+    "total_tokens": 279
   },
-  "object": "chat.completion",
+  "created": 1764257531,
   "choices": [
     {
       "index": 0,
-      "finish_reason": "stop",
       "message": {
-        "role": "assistant",
+        "content": "The distance between the **Earth and the Moon** varies because the Moon follows an **elliptical (oval-shaped) orbit** around Earth. Here are the key distances:\n\n1. **Average Distance (Semi-Major Axis):**\n   - **384,400 km (238,855 miles)**\n   - This is the most commonly cited distance.\n\n2. **Closest Approach (Perigee):**\n   - **~363,300 km (225,700 miles)**\n   - When the Moon is at its closest point to Earth.\n\n3. **Farthest Distance (Apogee):**\n   - **~405,500 km (252,000 miles)**\n   - When the Moon is at its farthest point from Earth.\n\n### Additional Fun Facts:\n- The Moon is **slowly moving away** from Earth at a rate of **~3.8 cm (1.5 inches) per year** due to tidal forces.\n- Light takes **~1.28 seconds** to travel from the Moon to Earth at the average distance.\n- The Apollo missions took **~3 days** to reach the Moon.\n\nWould you like details on how this distance is measured?",
         "tool_calls": null,
-        "content": "The \"best\" French cheese is highly subjective—it depends on personal taste, texture preferences, and how you plan to enjoy it (on a cheese board, melted, in cooking, etc.). However, here are some of the most **iconic, beloved, and award-winning French cheeses**, each celebrated for its unique qualities:\n\n### **Top Contenders for \"Best\" French Cheese**\n1. **Comté** (AOP)\n   - **Why?** A versatile, nutty, and complex **alpine cheese** aged 4 months to 3+ years. Younger Comté is mild and buttery; aged versions develop crystalline crunch and deep caramelized flavors.\n   - **Best for:** Cheese boards, fondue, grilled cheese, or eating plain.\n   - **Fun fact:** Made in the Jura Mountains, each wheel is unique based on the terroir and aging.\n\n2. **Roquefort** (AOP)\n   - **Why?** The **king of blue cheeses**, made from sheep’s milk and aged in natural caves. Creamy, tangy, and intensely flavorful with a sharp, salty bite.\n   - **Best for:** Spread on bread with honey or pears, in salads, or sauces.\n   - **Fun fact:** One of the oldest cheeses in France (mentioned by Pliny the Elder in 79 AD).\n\n3. **Camembert de Normandie** (AOP)\n   - **Why?** The **quintessential French cheese**—creamy, earthy, and slightly mushroomy when ripe. Invented in 1791 by Marie Harel.\n   - **Best for:** Baked (Camembert en croûte), with crusty bread, or apples.\n   - **Pro tip:** Eat it at room temperature when the center is soft but not runny.\n\n4. **Époisses de Bourgogne** (AOP)\n   - **Why?** A **pungent, washed-rind cheese** with a velvety texture and bold, barnyardy flavor. Napoleon reportedly loved it.\n   - **Best for:** Spreading on bread or pairing with Burgundy wine.\n   - **Warning:** So strong it’s banned on French public transport!\n\n5. **Reblochon** (AOP)\n   - **Why?** The **star of tartiflette**—creamy, nutty, and slightly smoky. Made in the Alps with raw cow’s milk.\n   - **Best for:** Melting in tartiflette, raclette, or on potatoes.\n\n6. **Brie de Meaux** (AOP)\n   - **Why?** The **\"King of Cheeses\"** (as declared at the 1815 Congress of Vienna). Rich, buttery, and mushroomy with a soft rind.\n   - **Best for:** Pairing with Champagne or fruit (grapes, figs).\n\n7. **Munster (Munster-Géromé)** (AOP)\n   - **Why?** A **stinky, washed-rind cheese** from Alsace with a creamy, tangy interior. Often served with cumin.\n   - **Best for:** Melting in tarts or eating with beer.\n\n8. **Beaufort** (AOP)\n   - **Why?** A **firm, nutty alpine cheese** similar to Gruyère but richer. Aged versions are caramelized and complex.\n   - **Best for:** Fondue, gratins, or shaving over salads.\n\n9. **Pélardon** (AOP)\n   - **Why?** A **small, tangy goat cheese** from the Cévennes or Provence. Comes in fresh (soft) or aged (firm, peppery) varieties.\n   - **Best for:** Salads, tartines, or grilled with herbs.\n\n10. **Bleu d’Auvergne** (AOP)\n   - **Why?** A **creamy, spicy blue cheese** with a sharper taste than Roquefort but less crumbly.\n   - **Best for:** Dressings, burgers, or cheese plates with walnuts.\n\n### **How to Choose?**\n- **Mild & creamy?** → Brie de Meaux, Camembert, Saint-André.\n- **Bold & stinky?** → Époisses, Munster, Pont-l’Évêque.\n- **Nutty & firm?** → Comté, Beaufort, Cantal.\n- **Blue cheese lover?** → Roquefort, Bleu d’Auvergne.\n- **Goat cheese fan?** → Crottin de Chavignol, Pélardon, Valençay.\n\n### **Pro Tips for Enjoying French Cheese**\n- **Serve at room temperature** (take it out of the fridge 1–2 hours before eating).\n- **Pair with wine**: Brie + Champagne, Roquefort + Sauternes, Comté + Jura wine.\n- **Avoid plastic wrap**—store in cheese paper or a breathable container.\n- **Visit a *fromagerie*** in France for expert recommendations!\n\n### **Final Verdict**\nIf forced to pick **one \"best\"**, **Comté** (aged 24+ months) and **Roquefort** are often crowned for their depth and versatility. But the true joy of French cheese is in the **diversity**—so try them all!\n\n**What’s your favorite?** 😊🧀"
-      }
+        "prefix": false,
+        "role": "assistant"
+      },
+      "finish_reason": "stop"
     }
   ]
 }

public/img/stadium-paris.jpeg

@@ -20,7 +20,7 @@ import os
 from mistralai import Mistral
 
 api_key = os.environ["MISTRAL_API_KEY"]
-model = "mistral-medium-latest"
+model = "mistral-large-latest"
 
 client = Mistral(api_key=api_key)
 
@@ -29,7 +29,7 @@ stream_response = client.chat.stream(
     messages = [
         {
             "role": "user",
-            "content": "What is the best French cheese?",
+            "content": "How far is the moon from earth? Answer with the distance in km only.",
         },
     ]
 )
@@ -50,7 +50,7 @@ from mistralai import Mistral
 
 async def main():
     api_key = os.environ["MISTRAL_API_KEY"]
-    model = "mistral-medium-latest"
+    model = "mistral-large-latest"
 
     client = Mistral(api_key=api_key)
 
@@ -59,7 +59,7 @@ async def main():
         messages=[
              {
                   "role": "user",
-                  "content": "Who is the best French painter? Answer in JSON.",
+                  "content": "How far is the moon from earth? Answer with the distance in km only.",
               },
         ],
     )
@@ -91,8 +91,8 @@ const client = new Mistral({ apiKey: apiKey });
 async function main() {
 
     const result = await client.chat.stream({
-        model: "mistral-medium-latest",
-        messages: [{ role: "user", content: "What is the best French cheese?" }],
+        model: "mistral-large-latest",
+        messages: [{ role: "user", content: "How far is the moon from earth? Answer with the distance in km only." }],
     });
 
     // If you want to print the stream text to the console
@@ -116,11 +116,11 @@ curl --location "https://api.mistral.ai/v1/chat/completions" \
      --header 'Accept: application/json' \
      --header "Authorization: Bearer $MISTRAL_API_KEY" \
      --data '{
-    "model": "mistral-medium-latest",
+    "model": "mistral-large-latest",
     "messages": [
      {
         "role": "user",
-        "content": "What is the best French cheese?"
+        "content": "How far is the moon from earth? Answer with the distance in km only."
       }
     ],
     "stream": true
@@ -131,105 +131,91 @@ curl --location "https://api.mistral.ai/v1/chat/completions" \
     <TabItem value="output" label="output">
 
 ```json
-data: {
-  "id":"64a7f9da74be4e06bccfcff4a2b3ed03",
-  "object":"chat.completion.chunk",
-  "created":1756223818,
-  "model":"mistral-medium-latest",
-  "choices":[
-    {
-      "index":0,
-      "delta": {
-        "role":"assistant",
-        "content":""
-      },
-      "finish_reason":null
-    }
-  ]
-}
-data: {
-  "id": "64a7f9da74be4e06bccfcff4a2b3ed03",
-  "object": "chat.completion.chunk",
-  "created": 1756223818,
-  "model": "mistral-medium-latest",
+{
+  "id": "59060ef9339a4112b2c9e57e3ee6199d",
+  "model": "mistral-large-latest",
   "choices": [
     {
       "index": 0,
       "delta": {
-        "content": "The \""
+        "role": "assistant",
+        "content": ""
       },
       "finish_reason": null
     }
-  ]
-}
-data: {
-  "id": "64a7f9da74be4e06bccfcff4a2b3ed03",
+  ],
   "object": "chat.completion.chunk",
-  "created": 1756223818,
-  "model": "mistral-medium-latest",
+  "created": 1764258570,
+  "usage": null
+}
+{
+  "id": "59060ef9339a4112b2c9e57e3ee6199d",
+  "model": "mistral-large-latest",
   "choices": [
     {
       "index": 0,
       "delta": {
-        "content": "best"
+        "content": "3"
       },
       "finish_reason": null
     }
-  ]
-}
-...
-data: {
-  "id": "64a7f9da74be4e06bccfcff4a2b3ed03",
+  ],
   "object": "chat.completion.chunk",
-  "created": 1756223818,
-  "model": "mistral-medium-latest",
+  "created": 1764258570,
+  "usage": null
+}
+{
+  "id": "59060ef9339a4112b2c9e57e3ee6199d",
+  "model": "mistral-large-latest",
   "choices": [
     {
       "index": 0,
       "delta": {
-        "content": " dish"
+        "content": "84"
       },
       "finish_reason": null
     }
-  ]
-}
-data: {
-  "id": "64a7f9da74be4e06bccfcff4a2b3ed03",
+  ],
   "object": "chat.completion.chunk",
-  "created": 1756223818,
-  "model": "mistral-medium-latest",
+  "created": 1764258570,
+  "usage": null
+}
+{
+  "id": "59060ef9339a4112b2c9e57e3ee6199d",
+  "model": "mistral-large-latest",
   "choices": [
     {
       "index": 0,
       "delta": {
-        "content": "?"
+        "content": "4"
       },
       "finish_reason": null
     }
-  ]
-}
-data: {
-  "id": "64a7f9da74be4e06bccfcff4a2b3ed03",
+  ],
   "object": "chat.completion.chunk",
-  "created": 1756223818,
-  "model": "mistral-medium-latest",
+  "created": 1764258570,
+  "usage": null
+}
+{
+  "id": "59060ef9339a4112b2c9e57e3ee6199d",
+  "model": "mistral-large-latest",
   "choices": [
     {
       "index": 0,
       "delta": {
-        "content": ""
+        "content": "00"
       },
       "finish_reason": "stop"
     }
   ],
+  "object": "chat.completion.chunk",
+  "created": 1764258570,
   "usage": {
-    "prompt_tokens": 10,
-    "total_tokens": 1254,
-    "completion_tokens": 1244
+    "prompt_tokens": 19,
+    "completion_tokens": 7,
+    "total_tokens": 26
   }
 }
-data: [DONE]
 ```
-
     </TabItem>
 </Tabs>