Remove mentions of Embeddings from Cody docs

MaedahBatool · MaedahBatool · commit 2569f6a66b74 · 2024-09-25T11:24:15.000-07:00
diff --git a/docs/cody/capabilities/chat.mdx b/docs/cody/capabilities/chat.mdx
@@ -17,7 +17,7 @@ To use Cody's chat, you'll need to have the following:
 
 ## How does chat work?
 
-Cody answers questions by searching your codebase and retrieving context relevant to your questions. Cody uses several methods to search for context, including keyword search and embeddings search. Finding and using context allows Cody to make informed responses based on your code rather than being limited to general knowledge. When Cody retrieves context to answer a question, it will tell you which code files it read to generate its response.
+Cody answers questions by searching your codebase and retrieving context relevant to your questions. Cody uses several methods to search for context, including Sourcegraph's native search and keyword search. Finding and using context allows Cody to make informed responses based on your code rather than being limited to general knowledge. When Cody retrieves context to answer a question, it will tell you which code files it read to generate its response.
 
 Cody can assist you with various use cases such as:
 
diff --git a/docs/cody/clients/install-jetbrains.mdx b/docs/cody/clients/install-jetbrains.mdx
@@ -168,7 +168,7 @@ If Cody's answer isn't helpful, you can try asking again with different context:
 
 ## Context fetching mechanism
 
-JetBrains users on the Free or Pro plan can leverage [local search](/cody/core-concepts/context#context-selection) as the primary context source for Cody chat. Local or remote embeddings will no longer be produced or used as a context source.
+JetBrains users on the Free or Pro plan can leverage [local search](/cody/core-concepts/context#context-selection) as the primary context source for Cody chat.
 
 Enterprise users can leverage the full power of the Sourcegraph search engine as the primary context provider to Cody.
 
diff --git a/docs/cody/core-concepts/context.mdx b/docs/cody/core-concepts/context.mdx
@@ -18,7 +18,6 @@ Context and [methods of retrieving context](#context-selection) are crucial to t
 Cody employs various methods to gather context relevant to user input, ensuring the quality of the information provided. These methods include:
 
 - **Keyword Search**: A traditional text search approach that finds keywords matching the user input. When needed, queries are automatically rewritten to include more relevant terms.
-- **Embeddings**: Dense vector search in which both documents (code) and queries (questions) are converted to an embeddings vector via an embeddings model. Retrieval is then done by computing the similarity between the query and document vectors
 - **Sourcegraph Search**: Sourcegraph Search API. Queries are sent to the SG instance (managed or self-hosted), and search is done using the SG search stack. Relevant documents are returned to the user IDE for use by the LLM
 - **Code Graph**: Analyzing the structure of the code, Cody examines how components are interconnected and used, finding context based on code elements' relationships
 
@@ -32,13 +31,13 @@ Users on Cody Free or Pro plans use context depending on the type of [clients](/
 
 Here's a detailed breakdown of the context fetching mechanism for each client for Cody Free, Pro, and Enterprise users:
 
-|    **Tier**    | **Client** | **Repositories** | **Notes** |
-| -------------- | ---------- | ---------------- | --------- |
-| **Free/Pro**   | VS Code    | 1        | Includes embeddings (beta) |
-|                | JetBrains  | 1        |  |
-| **Enterprise** | Cody Web   | Multi    |  |
-|                | VS Code    | Multi    |  |
-|                | JetBrains  | Multi    |  |
+|    **Tier**    | **Client** | **Repositories** |
+| -------------- | ---------- | ---------------- |
+| **Free/Pro**   | VS Code    | 1                |
+|                | JetBrains  | 1                |
+| **Enterprise** | Cody Web   | Multi            |
+|                | VS Code    | Multi            |
+|                | JetBrains  | Multi            |
 
 ## How does context work with Cody prompts?
 
@@ -66,7 +65,7 @@ Cody's usage and limits policies help optimize its performance and ensure cost-e
 
 ### Cody context window size
 
-While Cody aims to provide maximum context for each prompt, there are limits to ensure efficiency. For more details, see our [documentation on token limits](/cody/core-concepts/token-limits.mdx).  
+While Cody aims to provide maximum context for each prompt, there are limits to ensure efficiency. For more details, see our [documentation on token limits](/cody/core-concepts/token-limits.mdx).
 
 ### Manage context window size
 
diff --git a/docs/cody/core-concepts/embeddings.mdx b/docs/cody/core-concepts/embeddings.mdx
@@ -2,6 +2,8 @@
 
 <p className="subtitle">Learn how you can use embeddings with Cody for better code understanding.</p>
 
+<Callout type="info">Embeddings are no longer supported on Cody Free, Pro or Enterprise.</Callout>
+
 ## What are embeddings?
 
 Embeddings are a semantic representation of text that allows you to create a search index over your codebase. Cody splits your codebase into searchable chunks and sends them to an external service specified in your site's configuration for embedding. The resulting embedding index is stored in your local hard drive.
@@ -11,4 +13,3 @@ Embeddings are a semantic representation of text that allows you to create a sea
 <Callout type= "info">As of Sourcegraph version 5.3, embeddings are only available to Cody Free and Cody Pro users on the VS Code extension.</Callout>
 
 Cody Free and Pro users can leverage local embeddings for context fetching. Embeddings are created automatically and are used to provide context to Cody in chat, edits, and more.
-
diff --git a/docs/cody/core-concepts/index.mdx b/docs/cody/core-concepts/index.mdx
@@ -14,6 +14,4 @@
 
 <QuickLink title="Token Limits" icon='languagemodels' href="/cody/core-concepts/token-limits" description="Understand the constraints of Cody's context window." />
 
-<QuickLink title="Embeddings" icon='plugins' href="/cody/core-concepts/embeddings" description="Learn how you can use embeddings with Cody for better code understanding." />
-
 </QuickLinks>
diff --git a/docs/cody/core-concepts/keyword-search.mdx b/docs/cody/core-concepts/keyword-search.mdx
@@ -4,16 +4,6 @@
 
 Keyword search is the traditional approach to text search. It splits content into terms and builds a mapping from terms to documents. At query time, it extracts terms from the query and uses the mapping to retrieve your documents.
 
-Both Cody chat and completions use Keyword Search. It comes out of the box without any additional setup. While powerful, this method can be used as a fallback solution when a codebase lacks embeddings with enhanced context quality and decreased latency over time.
+Both Cody chat and completions use Keyword Search. It comes out of the box without any additional setup. Cody with Keyword Search searches your local VS Code workspace and is a cost-effective and time-saving solution.
 
-## Keyword Search vs Embeddings
-
-Embeddings search over your entire repo set. While Cody with Keyword Search only searches your local VS Code workspace.
-
-While setting things up, Keyword Search as the default experience is a cost-effective and time-saving solution. Embeddings need to be produced and managed. It's a time-consuming process.
-
-The codebase is divided into 20-30 line chunks, each run through an embedding service. The results must then be stored on an accessible machine for local Cody clients.
-
-For an enterprise admin who has set up Cody with a Code Search instance, developers on their local machines can seamlessly access it. However, indexing a codebase to produce embeddings can take minutes or even hours for users with Cody solely on their local machine.
-
-Cody employs keyword-based context to ensure a good user experience until embeddings become available. When they are accessible, Cody will use them, thereby enhancing the quality of responses.
+For an enterprise admin who has set up Cody with a Code Search instance, developers on their local machines can seamlessly access it.
diff --git a/docs/cody/faq.mdx b/docs/cody/faq.mdx
@@ -22,7 +22,6 @@ This process ensures that Cody can provide helpful answers to your questions whi
 Yes, Cody is compatible with self-hosted Sourcegraph instances. However, there are a few considerations:
 
 - Cody operates by sending code snippets (up to 28 KB per request) to a third-party cloud service. By default, this service is Anthropic but can also be OpenAI
-- For certain repositories, Cody may utilize embeddings, which involves sending repository data to another third-party service like OpenAI
 - To use Cody effectively, your self-hosted Sourcegraph instance must have internet access for these interactions with external services
 
 ### Is Cody licensed for private code, and does it allow GPL-licensed code?
@@ -48,56 +47,44 @@ Cody's response quality on a programming language depends on many factors, inclu
 
 ### Can Cody answer non-programming questions?
 
-Cody Chat is optimized for coding related use cases and can be used primarily for reviewing, analysis, testing, writing, and editing of software code. Use of Cody for any other purposes is against our [acceptable use policy](https://sourcegraph.com/terms/aup) and may result in your account being restricted. 
+Cody Chat is optimized for coding related use cases and can be used primarily for reviewing, analysis, testing, writing, and editing of software code. Use of Cody for any other purposes is against our [acceptable use policy](https://sourcegraph.com/terms/aup) and may result in your account being restricted.
 
 ### What happened to the Cody App?
 
-We’ve deprecated the Cody App to streamline the experience for our Cody Free and Cody Pro users. Now, anyone with a Sourcegraph.com account can generate local embeddings for their personal projects within the VS Code extension without downloading and connecting the Cody App. Local embeddings are only supported for VS Code, but we’re working on adding the same functionality to JetBrains IDEs.
+We’ve deprecated the Cody App to streamline the experience for our Cody Free and Cody Pro users.The Cody App is no longer available for download.
 
 ## Embeddings
 
-### What are embeddings for?
-
-Embeddings help Sourcegraph retrieve relevant code to feed the Large Language Model as context. Embeddings, often associated with vector search, complement other strategies in the code retrieval process.
-
-While embeddings excel in semantic matching — determining "what is this code about" and "what does it do" — they may not capture syntax and other specific matching details as effectively. Sourcegraph's approach involves getting the best results from various sources to deliver the most accurate and comprehensive answers possible.
-
 ### Why were embeddings removed once my instance was upgraded to v5.3?
 
-Cody now leverages **Sourcegraph Search** as a primary context provider, which comes with the following benefits:
+Cody leverages **Sourcegraph Search** as a primary context provider, which comes with the following benefits:
 
 - **More secure**: No code being sent to a third-party embedding API
 - **Easier to manage**: Less tech debt for embeddings setup and need for refreshes
 - **More repos**: Sourcegraph Search scales to larger repos and a greater number. Users on Enterprise instances will now be able to select multiple repos as context sources from within the IDE
 - **Equal, or better, quality**: Sourcegraph Search provides high-quality retrieval, as tested over the last ten years. When a customer sees degradation, we will be ready to respond quickly.
 
-Embeddings are just one mechanism for retrieval. We leverage multiple retrieval mechanisms to give Cody the right context and will be constantly iterating to improve Cody's quality. The most important aspect is getting the files from the codebase, not the specific algorithm used to find those files.
+We leverage multiple retrieval mechanisms to give Cody the right context and will be constantly iterating to improve Cody's quality. The most important aspect is getting the files from the codebase, not the specific algorithm used to find those files.
 
-### Why are embeddings no longer supported on Cody Enterprise?
+### Why are embeddings no longer supported on Cody PLG and Enterprise?
 
-There are two driving factors:
+Cody does not support embeddings on Cody PLG and Cody Enterprise because we have replaced them with Sourcegraph Search. There are two driving factors:
 
 - The need for a retrieval system that can scale across repos and to repos of greater size
 - A system that is secure and requires low maintenance on the part of users
 
-Leveraging Sourcegraph Search allowed us to deliver these enhancements. Early evidence suggests that this context fetching works as well as embeddings and is sometimes better.
-
-### Why are embeddings only supported for Cody Free and Cody Pro users in VS Code and not for JetBrains?
-
-Only users on VSCode will have continued access to local embeddings and only as a backup source to our local search index. The rationale here is that we want a place to continue to test embeddings to empirically measure their value and see if there are areas where we should consider returning them. This aligns with our strategy to use PLG as a test bed for features before they come to enterprise customers.
+Leveraging Sourcegraph Search allowed us to deliver these enhancements.
 
 ## Third party dependencies
 
-### What is the default `sourcegraph` provider for completions and embeddings?
+### What is the default `sourcegraph` provider for completions?
 
-The default provider for completions and embeddings, specified as `"provider": "sourcegraph"` refers to the [Sourcegraph Cody Gateway](/cody/core-concepts/cody-gateway). The Cody Gateway facilitates access to completions and embeddings for Sourcegraph enterprise instances by leveraging third-party services such as Anthropic and OpenAI.
+The default provider for completions, specified as `"provider": "sourcegraph"` refers to the [Sourcegraph Cody Gateway](/cody/core-concepts/cody-gateway). The Cody Gateway facilitates access to completions for Sourcegraph enterprise instances by leveraging third-party services such as Anthropic and OpenAI.
 
 ### What third-party cloud services does Cody depend on?
 
 Cody relies on one primary third-party dependency, i.e., Anthropic's Claude API. Users can use this with the OpenAI API configuration.
 
-Additionally, Cody can optionally use OpenAI for generating embeddings, enhancing the quality of its context snippets, although this is not mandatory.
-
 It's worth noting that these dependencies remain consistent when utilizing the [default `sourcegraph` provider, Cody Gateway](/cody/core-concepts/cody-gateway), which uses the same third-party providers.
 
 ### What is the retention policy for Anthropic and OpenAI?
diff --git a/docs/cody/troubleshooting.mdx b/docs/cody/troubleshooting.mdx
@@ -61,9 +61,6 @@ To troubleshoot further:
 1. Look for log messages such as the following:
 
 ```bash
-█ SimpleChatPanelProvider: getEnhancedContext > embeddings (start)
-█ SimpleChatPanelProvider: getEnhancedContext > searching local embeddings
-█ SimpleChatPanelProvider: getEnhancedContext > embeddings (end)
 █ SimpleChatPanelProvider: getEnhancedContext > search
 █ symf: using downloaded symf "/Users/beyang/Library/Application Support/Code/User/globalStorage/sourcegraph.cody-ai/symf/symf-v0.0.6-aarch64-macos"
 █ SimpleChatPanelProvider: getEnhancedContext > search (end)
diff --git a/docs/cody/usage-and-pricing.mdx b/docs/cody/usage-and-pricing.mdx
@@ -8,7 +8,7 @@ Cody provides three subscription plans: **Free**, **Pro**, and **Enterprise**. E
 
 The free plan is designed for individuals to get started with Cody. It comes with a set of features to enhance your coding experience. It includes **unlimited autocompletion suggestions** per month, covering both whole and multi-line suggestions. You also get **200 chats/prompts** per month with access to creating custom prompts.
 
-The free plan provides access to local context to enhance Cody's understanding of your code and ability to provide accurate suggestions. Local context uses keyword search and embeddings to search the local repository (the one currently open in your IDE).
+The free plan provides access to local context to enhance Cody's understanding of your code and ability to provide accurate suggestions. Local context uses keyword search to search the local repository (the one currently open in your IDE).
 
 Cody Free uses the DeepSeek-V2 model for autocomplete. It uses the Claude 3.5 Sonnet model for chat and prompts, and users have several other model options they can switch to. See [Supported LLMs](https://sourcegraph.com/docs/cody/capabilities/supported-models) for more information.
 
@@ -113,7 +113,6 @@ The following table shows a high-level comparison of the three plans available o
 | **Autocompletion suggestions**    | Unlimited                                                  | Unlimited                                                                                     | Unlimited                                          |
 | **Chat/Command Executions**       | 200 per month                                              | Unlimited                                                                                     | Unlimited                                          |
 | **Custom Commands**               | Supported                                                  | Supported                                                                                     | Supported                                          |
-| **Embeddings (public code)**      | Supported                                                  | Supported                                                                                     | Supported                                          |
 | **Keyword Context (local code)**  | Supported                                                  | Supported                                                                                     | Supported                                          |
 | **Developer Limitations**         | 1 developer                                                | 1 developer                                                                                 | Scalable, consumption-based pricing                |
 | **LLM Support**                   | Anthropic + StarCoder for Chat, Commands, and Autocomplete | Choice of LLMs for Chat and Commands (VS Code and JetBrains only), StarCoder and Autocomplete | Claude 3 and Bring Your Own LLM Key (experimental) |
