Fix chat API field names and structure

tpayet · tpayet · commit bfd517ffeb55 · 2025-06-11T12:17:49.000+02:00
- Change 'provider' to 'source' with correct camelCase values
- Change 'prompt' to 'prompts' object with system field
- Remove 'model' from settings (specified at runtime)
- Fix workspace object to only have 'uid' field
- Remove double quotes from Azure OpenAI in tabs
- Remove managing conversations section from guide
diff --git a/guides/ai/getting_started_with_chat.mdx b/guides/ai/getting_started_with_chat.mdx
@@ -42,69 +42,74 @@ Create a workspace with your LLM provider settings. Here are examples for differ
 
 <CodeGroup>
 
-```bash OpenAI
+```bash openAi
 curl \
   -X PUT 'http://localhost:7700/chats/my-assistant/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "openai",
-    "model": "gpt-3.5-turbo",
+    "source": "openAi",
     "apiKey": "sk-...",
-    "prompt": "You are a helpful assistant. Answer questions based only on the provided context."
+    "prompts": {
+      "system": "You are a helpful assistant. Answer questions based only on the provided context."
+    }
   }'
 ```
 
-```bash "Azure OpenAI"
+```bash azureOpenAi
 curl \
   -X PUT 'http://localhost:7700/chats/my-assistant/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "azure_openai",
-    "model": "gpt-35-turbo",
+    "source": "azureOpenAi",
     "apiKey": "your-azure-key",
     "baseUrl": "https://your-resource.openai.azure.com",
-    "prompt": "You are a helpful assistant. Answer questions based only on the provided context."
+    "prompts": {
+      "system": "You are a helpful assistant. Answer questions based only on the provided context."
+    }
   }'
 ```
 
-```bash Mistral
+```bash mistral
 curl \
   -X PUT 'http://localhost:7700/chats/my-assistant/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "mistral",
-    "model": "mistral-small-latest",
+    "source": "mistral",
     "apiKey": "your-mistral-key",
-    "prompt": "You are a helpful assistant. Answer questions based only on the provided context."
+    "prompts": {
+      "system": "You are a helpful assistant. Answer questions based only on the provided context."
+    }
   }'
 ```
 
-```bash Gemini
+```bash gemini
 curl \
   -X PUT 'http://localhost:7700/chats/my-assistant/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "gemini",
-    "model": "gemini-1.5-flash",
+    "source": "gemini",
     "apiKey": "your-gemini-key",
-    "prompt": "You are a helpful assistant. Answer questions based only on the provided context."
+    "prompts": {
+      "system": "You are a helpful assistant. Answer questions based only on the provided context."
+    }
   }'
 ```
 
-```bash vLLM
+```bash vllm
 curl \
   -X PUT 'http://localhost:7700/chats/my-assistant/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "vllm",
-    "model": "meta-llama/Llama-3-8b-chat-hf",
+    "source": "vllm",
     "baseUrl": "http://localhost:8000",
-    "prompt": "You are a helpful assistant. Answer questions based only on the provided context."
+    "prompts": {
+      "system": "You are a helpful assistant. Answer questions based only on the provided context."
+    }
   }'
 ```
 
@@ -210,96 +215,6 @@ for await (const chunk of stream) {
 
 </CodeGroup>
 
-## Managing conversations
-
-Since the chat API is stateless, maintain conversation history client-side:
-
-<CodeGroup>
-
-```javascript JavaScript
-import OpenAI from 'openai';
-
-class ChatConversation {
-  constructor(apiKey) {
-    this.client = new OpenAI({
-      baseURL: 'http://localhost:7700/chats/my-assistant',
-      apiKey: apiKey,
-    });
-    this.messages = [];
-  }
-
-  async sendMessage(content) {
-    // Add user message to history
-    this.messages.push({ role: 'user', content });
-
-    // Send conversation with history
-    const stream = await this.client.chat.completions.create({
-      model: 'gpt-3.5-turbo',
-      messages: this.messages,
-      stream: true,
-    });
-
-    // Collect assistant response
-    let assistantMessage = '';
-    for await (const chunk of stream) {
-      const content = chunk.choices[0]?.delta?.content || '';
-      assistantMessage += content;
-    }
-
-    // Add assistant response to history
-    this.messages.push({ role: 'assistant', content: assistantMessage });
-    
-    return assistantMessage;
-  }
-}
-
-// Usage
-const chat = new ChatConversation('YOUR_MEILISEARCH_API_KEY');
-const response = await chat.sendMessage('What is Meilisearch?');
-console.log(response);
-```
-
-```python Python
-from openai import OpenAI
-
-class ChatConversation:
-    def __init__(self, api_key):
-        self.client = OpenAI(
-            base_url="http://localhost:7700/chats/my-assistant",
-            api_key=api_key
-        )
-        self.messages = []
-    
-    def send_message(self, content):
-        # Add user message to history
-        self.messages.append({"role": "user", "content": content})
-        
-        # Send conversation with history
-        stream = self.client.chat.completions.create(
-            model="gpt-3.5-turbo",
-            messages=self.messages,
-            stream=True
-        )
-        
-        # Collect assistant response
-        assistant_message = ""
-        for chunk in stream:
-            if chunk.choices[0].delta.content is not None:
-                assistant_message += chunk.choices[0].delta.content
-        
-        # Add assistant response to history
-        self.messages.append({"role": "assistant", "content": assistant_message})
-        
-        return assistant_message
-
-# Usage
-chat = ChatConversation("YOUR_MEILISEARCH_API_KEY")
-response = chat.send_message("What is Meilisearch?")
-print(response)
-```
-
-</CodeGroup>
-
 ## Next steps
 
 - Explore [advanced chat API features](/reference/api/chats)
diff --git a/learn/ai_powered_search/conversational_search_with_chat.mdx b/learn/ai_powered_search/conversational_search_with_chat.mdx
@@ -69,7 +69,7 @@ Meilisearch's chatCompletions consolidates these into one streamlined process:
 
 The chatCompletions feature operates through workspaces, which are isolated configurations for different use cases or tenants. Each workspace can:
 
-- Use different LLM providers (OpenAI, Azure OpenAI, Mistral, Gemini, vLLM)
+- Use different LLM sources (openAi, azureOpenAi, mistral, gemini, vllm)
 - Apply custom prompts
 - Access specific indexes based on API keys
 - Maintain separate conversation contexts
diff --git a/reference/api/chats.mdx b/reference/api/chats.mdx
@@ -17,37 +17,33 @@ The chatCompletions feature is experimental and must be enabled through [experim
 
 ```json
 {
-  "workspace": "customer-support",
-  "createdAt": "2024-01-15T09:30:00Z",
-  "updatedAt": "2024-01-15T14:45:00Z"
+  "uid": "customer-support"
 }
 ```
 
-| Name             | Type   | Description                                                                                     |
-| :--------------- | :----- | :---------------------------------------------------------------------------------------------- |
-| **`workspace`**  | String | Unique identifier for the chatCompletions workspace                                                       |
-| **`createdAt`**  | String | Creation date of the workspace, represented in [RFC 3339](https://www.ietf.org/rfc/rfc3339.txt) format |
-| **`updatedAt`**  | String | Latest update date of the workspace, represented in [RFC 3339](https://www.ietf.org/rfc/rfc3339.txt) format |
+| Name        | Type   | Description                                         |
+| :---------- | :----- | :-------------------------------------------------- |
+| **`uid`**   | String | Unique identifier for the chatCompletions workspace |
 
 ## ChatCompletions settings object
 
 ```json
 {
-  "provider": "openai",
-  "model": "gpt-3.5-turbo",
+  "source": "openAi",
   "apiKey": "sk-...",
   "baseUrl": "https://api.openai.com/v1",
-  "prompt": "You are a helpful assistant that answers questions based on the provided context."
+  "prompts": {
+    "system": "You are a helpful assistant that answers questions based on the provided context."
+  }
 }
 ```
 
 | Name           | Type   | Description                                                                           |
 | :------------- | :----- | :------------------------------------------------------------------------------------ |
-| **`provider`** | String | LLM provider: `"openai"`, `"azure_openai"`, `"mistral"`, `"gemini"`, or `"vllm"`    |
-| **`model`**    | String | Model identifier (e.g., `"gpt-3.5-turbo"`)                                           |
-| **`apiKey`**   | String | API key for the LLM provider (write-only, optional for vLLM)                        |
-| **`baseUrl`**  | String | Base URL for the provider (required for Azure OpenAI and vLLM)                      |
-| **`prompt`**   | String | System prompt to guide the assistant's behavior                                      |
+| **`source`**   | String | LLM source: `"openAi"`, `"azureOpenAi"`, `"mistral"`, `"gemini"`, or `"vllm"`       |
+| **`apiKey`**   | String | API key for the LLM provider (write-only, optional for vllm)                        |
+| **`baseUrl`**  | String | Base URL for the provider (required for azureOpenAi and vllm)                       |
+| **`prompts`**  | Object | Prompts object containing system prompts and other configuration                     |
 
 ## Chat completions
 
@@ -179,10 +175,11 @@ Configure the LLM provider and settings for a chat workspace.
 
 ```json
 {
-  "provider": "openai",
-  "model": "gpt-3.5-turbo",
+  "source": "openAi",
   "apiKey": "sk-...",
-  "prompt": "You are a helpful assistant."
+  "prompts": {
+    "system": "You are a helpful assistant."
+  }
 }
 ```
 
@@ -196,69 +193,74 @@ Returns the updated settings object. Note that `apiKey` is write-only and will n
 
 <CodeGroup>
 
-```bash OpenAI
+```bash openAi
 curl \
   -X PUT 'http://localhost:7700/chats/customer-support/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "openai",
-    "model": "gpt-4",
+    "source": "openAi",
     "apiKey": "sk-...",
-    "prompt": "You are a helpful customer support assistant."
+    "prompts": {
+      "system": "You are a helpful customer support assistant."
+    }
   }'
 ```
 
-```bash "Azure OpenAI"
+```bash azureOpenAi
 curl \
   -X PUT 'http://localhost:7700/chats/customer-support/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "azure_openai",
-    "model": "gpt-4",
+    "source": "azureOpenAi",
     "apiKey": "your-azure-api-key",
     "baseUrl": "https://your-resource.openai.azure.com",
-    "prompt": "You are a helpful customer support assistant."
+    "prompts": {
+      "system": "You are a helpful customer support assistant."
+    }
   }'
 ```
 
-```bash Mistral
+```bash mistral
 curl \
   -X PUT 'http://localhost:7700/chats/customer-support/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "mistral",
-    "model": "mistral-large-latest",
+    "source": "mistral",
     "apiKey": "your-mistral-api-key",
-    "prompt": "You are a helpful customer support assistant."
+    "prompts": {
+      "system": "You are a helpful customer support assistant."
+    }
   }'
 ```
 
-```bash Gemini
+```bash gemini
 curl \
   -X PUT 'http://localhost:7700/chats/customer-support/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "gemini",
-    "model": "gemini-1.5-pro",
+    "source": "gemini",
     "apiKey": "your-gemini-api-key",
-    "prompt": "You are a helpful customer support assistant."
+    "prompts": {
+      "system": "You are a helpful customer support assistant."
+    }
   }'
 ```
 
-```bash vLLM
+```bash vllm
 curl \
   -X PUT 'http://localhost:7700/chats/customer-support/settings' \
   -H 'Authorization: Bearer MASTER_KEY' \
   -H 'Content-Type: application/json' \
   --data-binary '{
-    "provider": "vllm",
-    "model": "llama-3-8b",
+    "source": "vllm",
     "baseUrl": "http://your-vllm-server:8000",
-    "prompt": "You are a helpful customer support assistant."
+    "prompts": {
+      "system": "You are a helpful customer support assistant."
+    }
   }'
 ```
 
@@ -282,9 +284,10 @@ Returns the settings object without the `apiKey` field.
 
 ```json
 {
-  "provider": "openai",
-  "model": "gpt-3.5-turbo",
-  "prompt": "You are a helpful assistant."
+  "source": "openAi",
+  "prompts": {
+    "system": "You are a helpful assistant."
+  }
 }
 ```
 
@@ -340,14 +343,10 @@ curl \
 {
   "results": [
     {
-      "workspace": "customer-support",
-      "createdAt": "2024-01-15T09:30:00Z",
-      "updatedAt": "2024-01-15T14:45:00Z"
+      "uid": "customer-support"
     },
     {
-      "workspace": "internal-docs",
-      "createdAt": "2024-01-16T10:00:00Z",
-      "updatedAt": "2024-01-16T10:00:00Z"
+      "uid": "internal-docs"
     }
   ],
   "offset": 0,
