docs: Add embedding section to context doc (#2375)

* docs: Add embedding section to context doc Signed-off-by: TennyZhuang <[email protected]> * update doc Signed-off-by: TennyZhuang <[email protected]> * refine doc Signed-off-by: TennyZhuang <[email protected]> * Apply suggestions from code review --------- Signed-off-by: TennyZhuang <[email protected]> Co-authored-by: Meng Zhang <[email protected]>
TabbyML · Jun 9, 2024 · 02560fd · 02560fd
1 parent 94d35f0
commit 02560fd
Show file tree

Hide file tree

Showing 2 changed files with 31 additions and 3 deletions.
diff --git a/crates/tabby-scheduler/src/doc/mod.rs b/crates/tabby-scheduler/src/doc/mod.rs
@@ -72,7 +72,7 @@ impl IndexAttributeBuilder<SourceDocument> for DocBuilder {
                 }
 
                 let chunk = json!({
-                        doc::fields::CHUNK_TEXT: chunk_text,
+                    doc::fields::CHUNK_TEXT: chunk_text,
                 });
 
                 yield (chunk_embedding_tokens, chunk)

diff --git a/website/docs/administration/context/index.mdx b/website/docs/administration/context/index.mdx
@@ -25,7 +25,7 @@ The repository context is used to connect Tabby with a source code repository fr
     <img src={GitUrl} alt="Git" width="700" />
 
   - For GitHub / GitLab, a personal access token is required to access private repositories.
-    * Check the instructions in the corresponding tab to create a token. 
+    * Check the instructions in the corresponding tab to create a token.
 
     <img src={GitHubGitLabUrl} alt="GitHub or GitLab" width="700" />
 
@@ -59,4 +59,32 @@ Once connected, the indexing job will start automatically. You can check the sta
 
 Additionally, you can also visit the **Code Browser** page to view the connected repository.
 
-<img src={CodeBrowserUrl} alt="code browser" width="800" />
+<img src={CodeBrowserUrl} alt="code browser" width="800" />'
+
+## Internal: Vector Index
+
+When adding a document, it is converted into vectors that help quickly find relevant context. During searches or chats, queries and messages are also converted into vectors to locate the most similar documents.
+
+### Use the default embedding model
+
+The default embedding model is "Nomic-Embed-Text", which is a high-performing open embedding model with a large token context window.
+
+Currently, "Nomic-Embed-Text" is the only supported local embedding model.
+
+### Using a remote embedding model provider
+
+You can add also a remote embedding model provider by adding a new section to the `~/.tabby/config.toml` file.
+
+```toml
+[model.embedding.http]
+kind = "openai/embedding"
+api_key = "sk-..."
+model_name = "gpt-4"
+```
+
+Following embedding model providers are supported:
+
+* `openai/embedding`
+* `voyageai/embedding`
+* `llama.cpp/embedding`
+* `ollama/embedding`