update to lance 5.0.0-beta.2

Add support for routing merge_insert requests to a separate WAL/ingest
host. The WAL host is auto-derived as {db}.{region}.wal.lancedb.com or
can be explicitly overridden via ConnectBuilder::wal_host_override().
Callers opt in per-operation with MergeInsertBuilder::use_wal(true).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.27.2-beta.1"
+current_version = "0.27.2"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

Cargo.lock

@@ -3072,9 +3072,8 @@ checksum = "42703706b716c37f96a77aea830392ad231f44c9e9a67872fa5548707e11b11c"
 
 [[package]]
 name = "fsst"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "2195cc7f87e84bd695586137de99605e7e9579b26ec5e01b82960ddb4d0922f2"
 dependencies = [
  "arrow-array",
  "rand 0.9.2",
@@ -4124,9 +4123,8 @@ dependencies = [
 
 [[package]]
 name = "lance"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "efe6c3ddd79cdfd2b7e1c23cafae52806906bc40fbd97de9e8cf2f8c7a75fc04"
 dependencies = [
  "arrow",
  "arrow-arith",
@@ -4192,9 +4190,8 @@ dependencies = [
 
 [[package]]
 name = "lance-arrow"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "5d9f5d95bdda2a2b790f1fb8028b5b6dcf661abeb3133a8bca0f3d24b054af87"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4214,9 +4211,8 @@ dependencies = [
 
 [[package]]
 name = "lance-bitpacking"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "f827d6ab9f8f337a9509d5ad66a12f3314db8713868260521c344ef6135eb4e4"
 dependencies = [
  "arrayref",
  "paste",
@@ -4225,9 +4221,8 @@ dependencies = [
 
 [[package]]
 name = "lance-core"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "0f1e25df6a79bf72ee6bcde0851f19b1cd36c5848c1b7db83340882d3c9fdecb"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4264,9 +4259,8 @@ dependencies = [
 
 [[package]]
 name = "lance-datafusion"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "93146de8ae720cb90edef81c2f2d0a1b065fc2f23ecff2419546f389b0fa70a4"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -4296,9 +4290,8 @@ dependencies = [
 
 [[package]]
 name = "lance-datagen"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "ccec8ce4d8e0a87a99c431dab2364398029f2ffb649c1a693c60c79e05ed30dd"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -4316,9 +4309,8 @@ dependencies = [
 
 [[package]]
 name = "lance-encoding"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "5c1aec0bbbac6bce829bc10f1ba066258126100596c375fb71908ecf11c2c2a5"
 dependencies = [
  "arrow-arith",
  "arrow-array",
@@ -4355,9 +4347,8 @@ dependencies = [
 
 [[package]]
 name = "lance-file"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "14a8c548804f5b17486dc2d3282356ed1957095a852780283bc401fdd69e9075"
 dependencies = [
  "arrow-arith",
  "arrow-array",
@@ -4389,9 +4380,8 @@ dependencies = [
 
 [[package]]
 name = "lance-index"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "2da212f0090ea59f79ac3686660f596520c167fe1cb5f408900cf71d215f0e03"
 dependencies = [
  "arrow",
  "arrow-arith",
@@ -4455,9 +4445,8 @@ dependencies = [
 
 [[package]]
 name = "lance-io"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "41d958eb4b56f03bbe0f5f85eb2b4e9657882812297b6f711f201ffc995f259f"
 dependencies = [
  "arrow",
  "arrow-arith",
@@ -4498,9 +4487,8 @@ dependencies = [
 
 [[package]]
 name = "lance-linalg"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "0285b70da35def7ed95e150fae1d5308089554e1290470403ed3c50cb235bc5e"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4516,9 +4504,8 @@ dependencies = [
 
 [[package]]
 name = "lance-namespace"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "5f78e2a828b654e062a495462c6e3eb4fcf0e7e907d761b8f217fc09ccd3ceac"
 dependencies = [
  "arrow",
  "async-trait",
@@ -4531,9 +4518,8 @@ dependencies = [
 
 [[package]]
 name = "lance-namespace-impls"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "a2392314f3da38f00d166295e44244208a65ccfc256e274fa8631849fc3f4d94"
 dependencies = [
  "arrow",
  "arrow-ipc",
@@ -4547,6 +4533,7 @@ dependencies = [
  "lance-core",
  "lance-index",
  "lance-io",
+ "lance-linalg",
  "lance-namespace",
  "lance-table",
  "log",
@@ -4577,9 +4564,8 @@ dependencies = [
 
 [[package]]
 name = "lance-table"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "3df9c4adca3eb2074b3850432a9fb34248a3d90c3d6427d158b13ff9355664ee"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -4618,9 +4604,8 @@ dependencies = [
 
 [[package]]
 name = "lance-testing"
-version = "4.0.0"
+version = "5.0.0-beta.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
+source = "git+https://github.com/lance-format/lance.git?tag=v5.0.0-beta.2#34e311c7632f62d8e4ff3a6e8bd124f84f0b70dc"
-checksum = "7ed7119bdd6983718387b4ac44af873a165262ca94f181b104cd6f97912eb3bf"
 dependencies = [
  "arrow-array",
  "arrow-schema",
@@ -4631,7 +4616,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb"
-version = "0.27.2-beta.1"
+version = "0.27.2"
 dependencies = [
  "ahash",
  "anyhow",
@@ -4713,7 +4698,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-nodejs"
-version = "0.27.2-beta.1"
+version = "0.27.2"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4735,7 +4720,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-python"
-version = "0.30.2-beta.1"
+version = "0.30.2"
 dependencies = [
  "arrow",
  "async-trait",

Cargo.toml

@@ -15,20 +15,20 @@ categories = ["database-implementations"]
 rust-version = "1.91.0"
 
 [workspace.dependencies]
-lance = { version = "=4.0.0", default-features = false }
+lance = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2", default-features = false }
-lance-core = { version = "=4.0.0" }
+lance-core = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-datagen = { version = "=4.0.0" }
+lance-datagen = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-file = { version = "=4.0.0" }
+lance-file = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-io = { version = "=4.0.0", default-features = false }
+lance-io = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2", default-features = false }
-lance-index = { version = "=4.0.0" }
+lance-index = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-linalg = { version = "=4.0.0" }
+lance-linalg = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-namespace = { version = "=4.0.0" }
+lance-namespace = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-namespace-impls = { version = "=4.0.0", default-features = false }
+lance-namespace-impls = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2", default-features = false }
-lance-table = { version = "=4.0.0" }
+lance-table = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-testing = { version = "=4.0.0" }
+lance-testing = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-datafusion = { version = "=4.0.0" }
+lance-datafusion = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-encoding = { version = "=4.0.0" }
+lance-encoding = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
-lance-arrow = { version = "=4.0.0" }
+lance-arrow = { git = "https://github.com/lance-format/lance.git", tag = "v5.0.0-beta.2" }
 ahash = "0.8"
 # Note that this one does not include pyarrow
 arrow = { version = "57.2", optional = false }

docs/src/java/java.md

@@ -14,7 +14,7 @@ Add the following dependency to your `pom.xml`:
 <dependency>
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-core</artifactId>
-    <version>0.27.2-beta.1</version>
+    <version>0.27.2</version>
 </dependency>
 ```
 

java/lancedb-core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
       <groupId>com.lancedb</groupId>
       <artifactId>lancedb-parent</artifactId>
-      <version>0.27.2-beta.1</version>
+      <version>0.27.2-final.0</version>
       <relativePath>../pom.xml</relativePath>
     </parent>
 

java/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-parent</artifactId>
-    <version>0.27.2-beta.1</version>
+    <version>0.27.2-final.0</version>
     <packaging>pom</packaging>
     <name>${project.artifactId}</name>
     <description>LanceDB Java SDK Parent POM</description>

nodejs/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.27.2-beta.1"
+version = "0.27.2"
 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": "0.27.2-beta.1",
+	"version": "0.27.2",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.27.2-beta.1",
+	"version": "0.27.2",
 	"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": "0.27.2-beta.1",
+	"version": "0.27.2",
 	"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": "0.27.2-beta.1",
+	"version": "0.27.2",
 	"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": "0.27.2-beta.1",
+	"version": "0.27.2",
 	"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": "0.27.2-beta.1",
+  "version": "0.27.2",
   "os": [
     "win32"
   ],

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.27.2-beta.1",
+	"version": "0.27.2",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",

nodejs/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "@lancedb/lancedb",
-  "version": "0.27.2-beta.1",
+  "version": "0.27.2",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "@lancedb/lancedb",
-      "version": "0.27.2-beta.1",
+      "version": "0.27.2",
       "cpu": [
         "x64",
         "arm64"

nodejs/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.27.2-beta.1",
+  "version": "0.27.2",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",

rust/lancedb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb"
-version = "0.27.2-beta.1"
+version = "0.27.2"
 edition.workspace = true
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license.workspace = true

rust/lancedb/src/connection.rs

@@ -30,7 +30,10 @@ use crate::error::{Error, Result};
 #[cfg(feature = "remote")]
 use crate::remote::{
     client::ClientConfig,
-    db::{OPT_REMOTE_API_KEY, OPT_REMOTE_HOST_OVERRIDE, OPT_REMOTE_REGION},
+    db::{
+        OPT_REMOTE_API_KEY, OPT_REMOTE_HOST_OVERRIDE, OPT_REMOTE_REGION,
+        OPT_REMOTE_WAL_HOST_OVERRIDE,
+    },
 };
 use lance::io::ObjectStoreParams;
 pub use lance_encoding::version::LanceFileVersion;
@@ -664,6 +667,24 @@ impl ConnectBuilder {
         self
     }
 
+    /// Set the WAL host override for routing merge_insert requests
+    /// to a separate WAL/ingest service.
+    ///
+    /// This option is only used when connecting to LanceDB Cloud (db:// URIs)
+    /// and will be ignored for other URIs.
+    ///
+    /// # Arguments
+    ///
+    /// * `wal_host_override` - The WAL host override to use for the connection
+    #[cfg(feature = "remote")]
+    pub fn wal_host_override(mut self, wal_host_override: &str) -> Self {
+        self.request.options.insert(
+            OPT_REMOTE_WAL_HOST_OVERRIDE.to_string(),
+            wal_host_override.to_string(),
+        );
+        self
+    }
+
     /// Set the database specific options
     ///
     /// See [crate::database::listing::ListingDatabaseOptions] for the options available for
@@ -817,6 +838,7 @@ impl ConnectBuilder {
             &api_key,
             &region,
             options.host_override,
+            options.wal_host_override,
             self.request.client_config,
             storage_options.into(),
         )?);

rust/lancedb/src/remote/client.rs

@@ -190,6 +190,7 @@ pub struct RetryConfig {
 pub struct RestfulLanceDbClient<S: HttpSend = Sender> {
     client: reqwest::Client,
     host: String,
+    wal_host: String,
     pub(crate) retry_config: ResolvedRetryConfig,
     pub(crate) sender: S,
     pub(crate) id_delimiter: String,
@@ -200,6 +201,7 @@ impl<S: HttpSend> std::fmt::Debug for RestfulLanceDbClient<S> {
     fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
         f.debug_struct("RestfulLanceDbClient")
             .field("host", &self.host)
+            .field("wal_host", &self.wal_host)
             .field("retry_config", &self.retry_config)
             .field("sender", &self.sender)
             .field("id_delimiter", &self.id_delimiter)
@@ -285,6 +287,7 @@ impl RestfulLanceDbClient<Sender> {
         parsed_url: &ParsedDbUrl,
         region: &str,
         host_override: Option<String>,
+        wal_host_override: Option<String>,
         default_headers: HeaderMap,
         client_config: ClientConfig,
     ) -> Result<Self> {
@@ -372,11 +375,16 @@ impl RestfulLanceDbClient<Sender> {
             Some(host_override) => host_override,
             None => format!("https://{}.{}.api.lancedb.com", parsed_url.db_name, region),
         };
-        debug!("Created client for host: {}", host);
+        let wal_host = match wal_host_override {
+            Some(wal_host_override) => wal_host_override,
+            None => format!("https://{}.{}.wal.lancedb.com", parsed_url.db_name, region),
+        };
+        debug!("Created client for host: {}, wal_host: {}", host, wal_host);
         let retry_config = client_config.retry_config.clone().try_into()?;
         Ok(Self {
             client,
             host,
+            wal_host,
             retry_config,
             sender: Sender,
             id_delimiter: client_config
@@ -479,6 +487,12 @@ impl<S: HttpSend> RestfulLanceDbClient<S> {
         self.add_id_delimiter_query_param(builder)
     }
 
+    pub fn post_wal(&self, uri: &str) -> RequestBuilder {
+        let full_uri = format!("{}{}", self.wal_host, uri);
+        let builder = self.client.post(full_uri);
+        self.add_id_delimiter_query_param(builder)
+    }
+
     fn add_id_delimiter_query_param(&self, req: RequestBuilder) -> RequestBuilder {
         if self.id_delimiter != "$" {
             req.query(&[("delimiter", self.id_delimiter.clone())])
@@ -791,6 +805,7 @@ pub mod test_utils {
         RestfulLanceDbClient {
             client: reqwest::Client::new(),
             host: "http://localhost".to_string(),
+            wal_host: "http://localhost-wal".to_string(),
             retry_config: RetryConfig::default().try_into().unwrap(),
             sender: MockSender {
                 f: Arc::new(wrapper),
@@ -815,6 +830,7 @@ pub mod test_utils {
         RestfulLanceDbClient {
             client: reqwest::Client::new(),
             host: "http://localhost".to_string(),
+            wal_host: "http://localhost-wal".to_string(),
             retry_config: config.retry_config.try_into().unwrap(),
             sender: MockSender {
                 f: Arc::new(wrapper),
@@ -982,6 +998,7 @@ mod tests {
         let client = RestfulLanceDbClient {
             client: reqwest::Client::new(),
             host: "https://example.com".to_string(),
+            wal_host: "https://example.com".to_string(),
             retry_config: RetryConfig::default().try_into().unwrap(),
             sender: Sender,
             id_delimiter: "+".to_string(),
@@ -1017,6 +1034,7 @@ mod tests {
         let client = RestfulLanceDbClient {
             client: reqwest::Client::new(),
             host: "https://example.com".to_string(),
+            wal_host: "https://example.com".to_string(),
             retry_config: RetryConfig::default().try_into().unwrap(),
             sender: Sender,
             id_delimiter: "+".to_string(),
@@ -1054,6 +1072,7 @@ mod tests {
         let client = RestfulLanceDbClient {
             client: reqwest::Client::new(),
             host: "https://example.com".to_string(),
+            wal_host: "https://example.com".to_string(),
             retry_config: RetryConfig::default().try_into().unwrap(),
             sender: Sender,
             id_delimiter: "+".to_string(),

rust/lancedb/src/remote/db.rs

@@ -82,6 +82,7 @@ pub const OPT_REMOTE_PREFIX: &str = "remote_database_";
 pub const OPT_REMOTE_API_KEY: &str = "remote_database_api_key";
 pub const OPT_REMOTE_REGION: &str = "remote_database_region";
 pub const OPT_REMOTE_HOST_OVERRIDE: &str = "remote_database_host_override";
+pub const OPT_REMOTE_WAL_HOST_OVERRIDE: &str = "remote_database_wal_host_override";
 // TODO: add support for configuring client config via key/value options
 
 #[derive(Clone, Debug, Default)]
@@ -95,6 +96,11 @@ pub struct RemoteDatabaseOptions {
     /// This is required when connecting to LanceDB Enterprise and should be
     /// provided if using an on-premises LanceDB Enterprise instance.
     pub host_override: Option<String>,
+    /// The WAL host override
+    ///
+    /// When set, merge_insert operations using WAL routing will be sent to
+    /// this host instead of the auto-derived WAL host.
+    pub wal_host_override: Option<String>,
     /// Storage options configure the storage layer (e.g. S3, GCS, Azure, etc.)
     ///
     /// See available options at <https://lancedb.com/docs/storage/>
@@ -113,6 +119,7 @@ impl RemoteDatabaseOptions {
         let api_key = map.get(OPT_REMOTE_API_KEY).cloned();
         let region = map.get(OPT_REMOTE_REGION).cloned();
         let host_override = map.get(OPT_REMOTE_HOST_OVERRIDE).cloned();
+        let wal_host_override = map.get(OPT_REMOTE_WAL_HOST_OVERRIDE).cloned();
         let storage_options = map
             .iter()
             .filter(|(key, _)| !key.starts_with(OPT_REMOTE_PREFIX))
@@ -122,6 +129,7 @@ impl RemoteDatabaseOptions {
             api_key,
             region,
             host_override,
+            wal_host_override,
             storage_options,
         })
     }
@@ -141,6 +149,12 @@ impl DatabaseOptions for RemoteDatabaseOptions {
         if let Some(host_override) = &self.host_override {
             map.insert(OPT_REMOTE_HOST_OVERRIDE.to_string(), host_override.clone());
         }
+        if let Some(wal_host_override) = &self.wal_host_override {
+            map.insert(
+                OPT_REMOTE_WAL_HOST_OVERRIDE.to_string(),
+                wal_host_override.clone(),
+            );
+        }
     }
 }
 
@@ -185,6 +199,19 @@ impl RemoteDatabaseOptionsBuilder {
         self.options.host_override = Some(host_override);
         self
     }
+
+    /// Set the WAL host override
+    ///
+    /// When set, merge_insert operations using WAL routing will be sent to
+    /// this host instead of the auto-derived WAL host.
+    ///
+    /// # Arguments
+    ///
+    /// * `wal_host_override` - The WAL host override
+    pub fn wal_host_override(mut self, wal_host_override: String) -> Self {
+        self.options.wal_host_override = Some(wal_host_override);
+        self
+    }
 }
 
 #[derive(Debug)]
@@ -204,6 +231,7 @@ impl RemoteDatabase {
         api_key: &str,
         region: &str,
         host_override: Option<String>,
+        wal_host_override: Option<String>,
         client_config: ClientConfig,
         options: RemoteOptions,
     ) -> Result<Self> {
@@ -231,6 +259,7 @@ impl RemoteDatabase {
             &parsed,
             region,
             host_override,
+            wal_host_override,
             header_map,
             client_config.clone(),
         )?;

rust/lancedb/src/remote/table.rs

@@ -1610,13 +1610,17 @@ impl<S: HttpSend> BaseTable for RemoteTable<S> {
         self.check_mutable().await?;
 
         let timeout = params.timeout;
+        let use_wal = params.use_wal;
 
         let query = MergeInsertRequest::try_from(params)?;
-        let mut request = self
+        let path = format!("/v1/table/{}/merge_insert/", self.identifier);
-            .client
+        let mut request = if use_wal {
-            .post(&format!("/v1/table/{}/merge_insert/", self.identifier))
+            self.client.post_wal(&path)
-            .query(&query)
+        } else {
-            .header(CONTENT_TYPE, ARROW_STREAM_CONTENT_TYPE);
+            self.client.post(&path)
+        }
+        .query(&query)
+        .header(CONTENT_TYPE, ARROW_STREAM_CONTENT_TYPE);
 
         if let Some(timeout) = timeout {
             // (If it doesn't fit into u64, it's not worth sending anyways.)
@@ -2705,6 +2709,43 @@ mod tests {
         }
     }
 
+    #[tokio::test]
+    async fn test_merge_insert_use_wal() {
+        let batch = RecordBatch::try_new(
+            Arc::new(Schema::new(vec![Field::new("a", DataType::Int32, false)])),
+            vec![Arc::new(Int32Array::from(vec![1, 2, 3]))],
+        )
+        .unwrap();
+        let data: Box<dyn RecordBatchReader + Send> = Box::new(RecordBatchIterator::new(
+            [Ok(batch.clone())],
+            batch.schema(),
+        ));
+
+        let table = Table::new_with_handler("my_table", move |request| {
+            if request.url().path() == "/v1/table/my_table/merge_insert/" {
+                // Verify the request was sent to the WAL host
+                assert_eq!(
+                    request.url().host_str().unwrap(),
+                    "localhost-wal",
+                    "merge_insert with use_wal should route to WAL host"
+                );
+
+                http::Response::builder()
+                    .status(200)
+                    .body(r#"{"version": 1, "num_deleted_rows": 0, "num_inserted_rows": 3, "num_updated_rows": 0}"#)
+                    .unwrap()
+            } else {
+                panic!("Unexpected request path: {}", request.url().path());
+            }
+        });
+
+        let mut builder = table.merge_insert(&["some_col"]);
+        builder.use_wal(true);
+        let result = builder.execute(data).await.unwrap();
+
+        assert_eq!(result.num_inserted_rows, 3);
+    }
+
     #[tokio::test]
     async fn test_merge_insert_retries_on_409() {
         let batch = RecordBatch::try_new(

rust/lancedb/src/table.rs

@@ -23,7 +23,7 @@ use lance::index::vector::VectorIndexParams;
 use lance::index::vector::utils::infer_vector_dim;
 use lance::io::{ObjectStoreParams, WrappingObjectStore};
 use lance_datafusion::utils::StreamingWriteSource;
-use lance_index::DatasetIndexExt;
+use lance::index::DatasetIndexExt;
 use lance_index::IndexType;
 use lance_index::scalar::{BuiltinIndexType, ScalarIndexParams};
 use lance_index::vector::bq::RQBuildParams;

rust/lancedb/src/table/merge.rs

@@ -55,6 +55,7 @@ pub struct MergeInsertBuilder {
     pub(crate) when_not_matched_by_source_delete_filt: Option<String>,
     pub(crate) timeout: Option<Duration>,
     pub(crate) use_index: bool,
+    pub(crate) use_wal: bool,
 }
 
 impl MergeInsertBuilder {
@@ -69,6 +70,7 @@ impl MergeInsertBuilder {
             when_not_matched_by_source_delete_filt: None,
             timeout: None,
             use_index: true,
+            use_wal: false,
         }
     }
 
@@ -148,6 +150,18 @@ impl MergeInsertBuilder {
         self
     }
 
+    /// Controls whether to route the merge insert operation through the WAL host.
+    ///
+    /// When set to `true`, the operation will be sent to the WAL host instead of
+    /// the main API host. The WAL host is auto-derived from the database connection
+    /// or can be explicitly set via [`crate::connection::ConnectBuilder::wal_host_override`].
+    ///
+    /// Defaults to `false`.
+    pub fn use_wal(&mut self, use_wal: bool) -> &mut Self {
+        self.use_wal = use_wal;
+        self
+    }
+
     /// Executes the merge insert operation
     ///
     /// Returns version and statistics about the merge operation including the number of rows

rust/lancedb/src/table/optimize.rs

@@ -10,7 +10,7 @@ use std::sync::Arc;
 
 use lance::dataset::cleanup::RemovalStats;
 use lance::dataset::optimize::{CompactionMetrics, IndexRemapperOptions, compact_files};
-use lance_index::DatasetIndexExt;
+use lance::index::DatasetIndexExt;
 use lance_index::optimize::OptimizeOptions;
 use log::info;