fix

Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.19.0-beta.4"
+current_version = "0.19.0-beta.5"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

docs/mkdocs.yml

@@ -162,6 +162,7 @@ nav:
               - Choosing right query type: guides/tuning_retrievers/1_query_types.md
               - Reranking: guides/tuning_retrievers/2_reranking.md
               - Embedding fine-tuning: guides/tuning_retrievers/3_embed_tuning.md
+          - Build MCP with LanceDB: guides/mcp.md
       - 🧬 Managing embeddings:
           - Understand Embeddings: embeddings/understanding_embeddings.md
           - Get Started: embeddings/index.md
@@ -293,6 +294,7 @@ nav:
           - Choosing right query type: guides/tuning_retrievers/1_query_types.md
           - Reranking: guides/tuning_retrievers/2_reranking.md
           - Embedding fine-tuning: guides/tuning_retrievers/3_embed_tuning.md
+      - Build MCP with LanceDB: guides/mcp.md
   - Managing Embeddings:
       - Understand Embeddings: embeddings/understanding_embeddings.md
       - Get Started: embeddings/index.md

docs/src/guides/mcp.md

@@ -0,0 +1,126 @@
+# MCP server with LanceDB
+
+The Model Context Protocol (MCP) is an open protocol that enables seamless integration between LLM applications and external data sources and tools. Whether you're building an AI-powered IDE, enhancing a chat interface, or creating custom AI workflows, MCP provides a standardized way to connect LLMs with the context they need.
+
+With LanceDB, your MCP can be embedded in your application. Let's implement 2 simple MCP tools using LanceDB
+1. Add data - add data to LanceDB
+2. Retreive data - retrieve data from LanceDB
+
+You need to install `mcp[cli]` python package.
+
+First, let's define some configs:
+```python
+# mcp_server.py
+
+LANCEDB_URI = "~/lancedb"
+TABLE_NAME =  "mcp_data"
+EMBEDDING_FUNCTION = "sentence-transformers"
+MODEL_NAME = "all-MiniLM-L6-v2" 
+```
+
+Then initialize the table that we'll use to store and retreive data:
+```python
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry
+
+model = get_registry().get(EMBEDDING_FUNCTION).create(model_name=MODEL_NAME)
+
+class Schema(LanceModel):
+    text: str = model.SourceField()
+    vector: Vector(model.ndims()) = model.VectorField()
+
+db = lancedb.connect(LANCEDB_URI)
+if TABLE_NAME not in db.table_names():
+    db.create_table(TABLE_NAME, schema=Schema)
+```
+
+!!! Note "Using LanceDB cloud"
+    If you want to use LanceDB cloud, you'll need to set the uri to your remote table
+    instance and also provide a token. Every other functionality will remain the same
+
+## Defining the tools
+Tools let LLMs take actions through your server. There are other components like `resources` that allow you to expose certain data sources to LLMs. For our use case, we need to define tools that LLMs can call in order to inget or retrieve data
+We'll use `FastMCP` interface of the MCP package. The FastMCP server is your core interface to the MCP protocol. It handles connection management, protocol compliance, and message routing.
+
+```python
+from mcp.server.fastmcp import FastMCP
+
+mcp = FastMCP("lancedb-example")
+```
+### Add data ingestion tool
+This function takes a string as input and adds it to the LanceDB table.
+
+```python
+@mcp.tool()
+async def ingest_data(content: str) -> str:
+    """
+    Add a new memory to the vector database
+    Args:
+        content: Content of the memory
+    """
+    tbl = db[TABLE_NAME]
+    tbl.add([
+        {"text": content}
+        ])
+    return f"Added memory: {content}"
+```
+### Retreive data tool
+
+```python
+@mcp.tool()
+async def retrieve_data(query: str, limit: int = 5) -> str:
+    """
+    Search db using vector search
+    Args:
+        query: The search query
+        limit: Maximum number of results to return
+    """
+    tbl = db[TABLE_NAME]
+    rs = tbl.search(query).limit(limit).to_list()
+    data = [
+        r["text"] for r in rs
+    ]
+    if not data:
+        return "No relevant data found."
+
+    return "\n\n".join(data)
+```
+This function takes a string and limit as input and searches the LanceDB table for the most relevant memories.
+
+
+## Install it on Claude desktop
+
+To install this MCP, you can simply run this command and it'll be registered on you Claude desktop
+```
+mcp install mcp_server.py
+```
+You'll see logs similar to this:
+```
+[04/07/25 20:18:08] INFO     Load pretrained SentenceTransformer: BAAI/bge-small-en-v1.5       SentenceTransformer.py:218
+Batches: 100%|█████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00,  4.06it/s]
+[04/07/25 20:18:11] INFO     Added server 'lancedb' to Claude config                                        claude.py:129
+                    INFO     Successfully installed lancedb in Claude app                                      cli.py:467
+```
+
+Now simply fire up claude desktop and you can start using it.
+
+1. If installed correctly, you'll `lancedb` in the MCP apps list
+![Screenshot 2025-04-08 at 8 07 39 AM](https://github.com/user-attachments/assets/6dede8ae-7e39-4931-ae60-b57ce620b328)
+
+2. You can now use the `ingest_data` tool to add data to the table. To do that, you can simply ask claude using natural language
+![Screenshot 2025-04-08 at 8 10 37 AM](https://github.com/user-attachments/assets/0cd4df4e-98bb-4bf1-8566-1671eb310a1d)
+
+3. Now you can start asking questions using the `retrieve_data` tool. It'll automatically search the table for relevant data. You should see something like this
+![Screenshot 2025-04-08 at 8 11 49 AM](https://github.com/user-attachments/assets/71b5b232-601c-4864-9d52-9b84f16adad9)
+
+4. Claude tries to set the params for tool calling on its own but you can also specify the details.
+![Screenshot 2025-04-08 at 8 12 30 AM](https://github.com/user-attachments/assets/5f362bd1-b2fc-4145-8f1e-968d453bf615)
+
+
+## Community examples
+
+- Find a minimal LanceDB mcp server similar to this [here](https://github.com/kyryl-opens-ml/mcp-server-lancedb/blob/main/src/mcp_lance_db/server.py)
+
+- You can find an implementation of a more complex MCP server that uses LanceDB to implement an advanced CodeQA feature [here](https://github.com/lancedb/MCPExample).
+

java/core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lancedb-parent</artifactId>
-        <version>0.19.0-beta.4</version>
+        <version>0.19.0-beta.5</version>
         <relativePath>../pom.xml</relativePath>
     </parent>
 

java/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-parent</artifactId>
-    <version>0.19.0-beta.4</version>
+    <version>0.19.0-beta.5</version>
     <packaging>pom</packaging>
 
     <name>LanceDB Parent</name>

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.19.0-beta.4",
+  "version": "0.19.0-beta.5",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.19.0-beta.4",
+      "version": "0.19.0-beta.5",
       "cpu": [
         "x64",
         "arm64"
@@ -52,11 +52,11 @@
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.19.0-beta.4",
+        "@lancedb/vectordb-darwin-arm64": "0.19.0-beta.5",
-        "@lancedb/vectordb-darwin-x64": "0.19.0-beta.4",
+        "@lancedb/vectordb-darwin-x64": "0.19.0-beta.5",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.19.0-beta.4",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.19.0-beta.5",
-        "@lancedb/vectordb-linux-x64-gnu": "0.19.0-beta.4",
+        "@lancedb/vectordb-linux-x64-gnu": "0.19.0-beta.5",
-        "@lancedb/vectordb-win32-x64-msvc": "0.19.0-beta.4"
+        "@lancedb/vectordb-win32-x64-msvc": "0.19.0-beta.5"
       },
       "peerDependencies": {
         "@apache-arrow/ts": "^14.0.2",
@@ -327,60 +327,65 @@
       }
     },
     "node_modules/@lancedb/vectordb-darwin-arm64": {
-      "version": "0.19.0-beta.4",
+      "version": "0.19.0-beta.5",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.19.0-beta.4.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.19.0-beta.5.tgz",
-      "integrity": "sha512-uS5AuT3Q4swrtM9JAhF8mM8Nt+kvewmB3DQWGiuYbhmMismSu8WlOHQAs9Yyh8N7NBdWENSTjroSExqjHPdFhQ==",
+      "integrity": "sha512-NuJVGaV4b6XgH3dlkCEquvtGM1cY5sIJE5M/LgJ3HYYvAbco/seBQM5AHTV/7CULoPEY9eQeJZOj9fWP5oQLYQ==",
       "cpu": [
         "arm64"
       ],
+      "license": "Apache-2.0",
       "optional": true,
       "os": [
         "darwin"
       ]
     },
     "node_modules/@lancedb/vectordb-darwin-x64": {
-      "version": "0.19.0-beta.4",
+      "version": "0.19.0-beta.5",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.19.0-beta.4.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.19.0-beta.5.tgz",
-      "integrity": "sha512-kjn3iTqZSx57ek9PN2AdPvJMx14tFkXc8sUFd3MLhY7FdWafx7Wvl0SLz2LubotJVFd6LMxvsPPNJEM5bEgMOw==",
+      "integrity": "sha512-hbadwvQcUgKJfluUHhN+mx+XeFRwTuh9mD0L3Tf3t3BkDTxyHpEG5WNgOpWrh6e1RU6zW54CoCyQuSEaVqGgGw==",
       "cpu": [
         "x64"
       ],
+      "license": "Apache-2.0",
       "optional": true,
       "os": [
         "darwin"
       ]
     },
     "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.19.0-beta.4",
+      "version": "0.19.0-beta.5",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.19.0-beta.4.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.19.0-beta.5.tgz",
-      "integrity": "sha512-iZlR7ffKC+XA1mGuuwXJojgFcUvXkgMt6pKR6lP3hsxXh8UOTWDljN7jkI8jKHcJez3rrqoqt1VjH3xD69fwtA==",
+      "integrity": "sha512-fu/EOYLr3mx76/SP4dEgbq0vSYHfuTf68lVl5/tL6eIb1Purz42l22+jNKLJ/S3Plase2SkXdxyY90K2Y/CvSg==",
       "cpu": [
         "arm64"
       ],
+      "license": "Apache-2.0",
       "optional": true,
       "os": [
         "linux"
       ]
     },
     "node_modules/@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.19.0-beta.4",
+      "version": "0.19.0-beta.5",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.19.0-beta.4.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.19.0-beta.5.tgz",
-      "integrity": "sha512-uxLeerlT5FuWzuvHlTDLdLCakyUJ+qJitReoCKT6tKhfcjIkbr+NEoLZEHifJC4dRFPtbddVgiYN6VHlnPPD/w==",
+      "integrity": "sha512-pzb8fl5M8155sc/mEFnKmuh9rCfQohHBlb+j+5qNMe84AyygQ8Me1H3b1h9fOkUPu2Y168zYfuGkjNv4Bjm9eA==",
       "cpu": [
         "x64"
       ],
+      "license": "Apache-2.0",
       "optional": true,
       "os": [
         "linux"
       ]
     },
     "node_modules/@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.19.0-beta.4",
+      "version": "0.19.0-beta.5",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.19.0-beta.4.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.19.0-beta.5.tgz",
-      "integrity": "sha512-QSugxudXooLCF7trudaAo9PfOzX7SFBIiHOoL4N6nwjC61u/JAsoiytw1Xjs/+0pOG5cT2WUMufBzBPgJyOxbw==",
+      "integrity": "sha512-5z6BSfTuZYJdDL2wwRrEQlnfluahzaUH2U7vj3i4ik4zaAwvaYcrjmdYCTLRYhFscUqzxd2pVFHbfRYe+maYzA==",
       "cpu": [
         "x64"
       ],
+      "license": "Apache-2.0",
       "optional": true,
       "os": [
         "win32"

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.19.0-beta.4",
+  "version": "0.19.0-beta.5",
   "description": " Serverless, low-latency vector database for AI applications",
   "private": false,
   "main": "dist/index.js",
@@ -89,10 +89,10 @@
     }
   },
   "optionalDependencies": {
-    "@lancedb/vectordb-darwin-x64": "0.19.0-beta.4",
+    "@lancedb/vectordb-darwin-x64": "0.19.0-beta.5",
-    "@lancedb/vectordb-darwin-arm64": "0.19.0-beta.4",
+    "@lancedb/vectordb-darwin-arm64": "0.19.0-beta.5",
-    "@lancedb/vectordb-linux-x64-gnu": "0.19.0-beta.4",
+    "@lancedb/vectordb-linux-x64-gnu": "0.19.0-beta.5",
-    "@lancedb/vectordb-linux-arm64-gnu": "0.19.0-beta.4",
+    "@lancedb/vectordb-linux-arm64-gnu": "0.19.0-beta.5",
-    "@lancedb/vectordb-win32-x64-msvc": "0.19.0-beta.4"
+    "@lancedb/vectordb-win32-x64-msvc": "0.19.0-beta.5"
   }
 }

nodejs/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.19.0-beta.4"
+version = "0.19.0-beta.5"
 license.workspace = true
 description.workspace = true
 repository.workspace = true

nodejs/npm/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.19.0-beta.4",
+	"version": "0.19.0-beta.5",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",

nodejs/npm/darwin-x64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-x64",
-	"version": "0.19.0-beta.4",
+	"version": "0.19.0-beta.5",
 	"os": ["darwin"],
 	"cpu": ["x64"],
 	"main": "lancedb.darwin-x64.node",

nodejs/npm/linux-arm64-gnu/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.19.0-beta.4",
+	"version": "0.19.0-beta.5",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-gnu.node",

nodejs/npm/linux-arm64-musl/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-musl",
-	"version": "0.19.0-beta.4",
+	"version": "0.19.0-beta.5",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-musl.node",

nodejs/npm/linux-x64-gnu/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-gnu",
-	"version": "0.19.0-beta.4",
+	"version": "0.19.0-beta.5",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-gnu.node",

nodejs/npm/linux-x64-musl/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-musl",
-	"version": "0.19.0-beta.4",
+	"version": "0.19.0-beta.5",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-musl.node",

nodejs/npm/win32-arm64-msvc/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lancedb/lancedb-win32-arm64-msvc",
-  "version": "0.19.0-beta.4",
+  "version": "0.19.0-beta.5",
   "os": [
     "win32"
   ],

nodejs/npm/win32-x64-msvc/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.19.0-beta.4",
+	"version": "0.19.0-beta.5",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",

nodejs/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "@lancedb/lancedb",
-  "version": "0.19.0-beta.4",
+  "version": "0.19.0-beta.5",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "@lancedb/lancedb",
-      "version": "0.19.0-beta.4",
+      "version": "0.19.0-beta.5",
       "cpu": [
         "x64",
         "arm64"

nodejs/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.19.0-beta.4",
+  "version": "0.19.0-beta.5",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-node"
-version = "0.19.0-beta.4"
+version = "0.19.0-beta.5"
 description = "Serverless, low-latency vector database for AI applications"
 license.workspace = true
 edition.workspace = true

rust/lancedb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb"
-version = "0.19.0-beta.4"
+version = "0.19.0-beta.5"
 edition.workspace = true
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license.workspace = true