V1-34/contextualai-amendments

docs/weaviate/model-providers/_includes/provider.generative.py

@@ -468,6 +468,7 @@ def import_data():
         # top_p=0.9,
         # system_prompt="You are a helpful assistant"
         # avoid_commentary=True,
+        # knowledge=["Custom knowledge override", "Additional context"],
     )
     # highlight-end
     # Additional parameters not shown
@@ -496,6 +497,7 @@ def import_data():
         # top_p=0.9,
         # system_prompt="You are a helpful assistant"
         # avoid_commentary=True,
+        # knowledge=["Custom knowledge override", "Additional context"],
     ),
     # Additional parameters not shown
     # highlight-end

docs/weaviate/model-providers/_includes/provider.generative.ts

@@ -421,6 +421,7 @@ await client.collections.create({
     // topP: 0.9,
     // systemPrompt: 'You are a helpful assistant',
     // avoidCommentary: true,
+    // knowledge: ['Custom knowledge override', 'Additional context'],
   }),
   // highlight-end
   // Additional parameters not shown
@@ -439,6 +440,7 @@ response = await myCollection.generate.nearText("A holiday film", {
     // topP: 0.9,
     // systemPrompt: 'You are a helpful assistant',
     // avoidCommentary: true,
+    // knowledge: ['Custom knowledge override', 'Additional context'],
   }),
   // highlight-end
 }, {

docs/weaviate/model-providers/contextualai/generative.md

@@ -160,16 +160,17 @@ Configure the following generative parameters to customize the model behavior.
 
 </Tabs>
 
-For further details on model parameters, see the [Contextual AI API documentation](https://docs.contextual.ai/).
+For further details on model parameters, see the [Contextual AI API documentation](https://docs.contextual.ai/api-reference/generate/generate).
 
-If a parameter is not specified, Weaviate uses the server-side default for that parameter. They are documented below.
+If a parameter is not specified, Weaviate uses the server-side default for that parameter. They are:
 
-- `DefaultContextualAIModel`           = `"v2"`
-- `DefaultContextualAITemperature`     = `0.0`
-- `DefaultContextualAITopP`            = `0.9`
-- `DefaultContextualAIMaxNewTokens`    = `1024`
-- `DefaultContextualAISystemPrompt`    = `""`
-- `DefaultContextualAIAvoidCommentary` = `false`
+- model           = `"v2"`
+- temperature     = `0.0`
+- topP            = `0.9`
+- maxNewTokens    = `1024`
+- systemPrompt    = `""`
+- avoidCommentary = `false`
+- knowledge       = `nil`
 
 ## Select a model at runtime
 

docs/weaviate/model-providers/contextualai/reranker.md

@@ -108,9 +108,9 @@ Configure a Weaviate collection to use a Contextual AI reranker model as follows
 
 </Tabs>
 
-### Select a model
+### Reranker parameters
 
-You can specify one of the [available models](#available-models) for Weaviate to use, as shown in the following configuration example:
+Configure the reranker behavior, including the model to use, through the following parameters:
 
 <Tabs className="code" groupId="languages">
   <TabItem value="py" label="Python">
@@ -135,6 +135,8 @@ You can specify one of the [available models](#available-models) for Weaviate to
 
 The [default model](#available-models) is used if no model is specified.
 
+For further details on model parameters, see the [Contextual AI API documentation](https://docs.contextual.ai/api-reference/rerank/rerank).
+
 ## Header parameters
 
 You can provide the API key as well as some optional parameters at runtime through additional headers in the request. The following headers are available: