diff --git a/rust/Cargo.lock b/rust/Cargo.lock
index 4193655805..9c7cc758f2 100644
--- a/rust/Cargo.lock
+++ b/rust/Cargo.lock
@@ -5947,7 +5947,7 @@ dependencies = [
 
 [[package]]
 name = "vllm-proto"
-version = "0.5.0"
+version = "0.6.0"
 dependencies = [
  "prost",
  "prost-types",
diff --git a/rust/proto/Cargo.toml b/rust/proto/Cargo.toml
index c2e32d43a7..11babde435 100644
--- a/rust/proto/Cargo.toml
+++ b/rust/proto/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "vllm-proto"
-version = "0.5.0"
+version = "0.6.0"
 edition.workspace = true
 license.workspace = true
 description = "Protobuf messages and gRPC bindings for the vLLM Rust frontend"
diff --git a/rust/proto/README.md b/rust/proto/README.md
index 3a60c05dfd..d77c301a0a 100644
--- a/rust/proto/README.md
+++ b/rust/proto/README.md
@@ -16,7 +16,7 @@ the schemas on pull requests, including on forks.
 
 ```toml
 [dependencies]
-vllm-proto = "0.5"
+vllm-proto = "0.6"
 ```
 
 For example, import `vllm_proto::inference_client::InferenceClient` or
diff --git a/rust/proto/control.proto b/rust/proto/control.proto
index e42f958582..52188ae0df 100644
--- a/rust/proto/control.proto
+++ b/rust/proto/control.proto
@@ -54,6 +54,7 @@ message ServerInfo {
   uint64 max_batched_tokens = 9;
   uint32 max_loras = 10;
   RlCapabilities rl_capabilities = 11;
+  string proto_version = 12;
   // Full-attention block size in tokens after initialization; absent if unavailable.
   optional uint64 effective_attention_block_size = 14;
 }
diff --git a/rust/proto/src/lib.rs b/rust/proto/src/lib.rs
index d4d44ea099..d1a90ee7a7 100644
--- a/rust/proto/src/lib.rs
+++ b/rust/proto/src/lib.rs
@@ -4,3 +4,6 @@
 //! Protobuf messages and gRPC clients and servers for vLLM's Inference and Control APIs.
 
 tonic::include_proto!("vllm");
+
+/// Schema version of this crate, reported by servers as `ServerInfo.proto_version`.
+pub const VERSION: &str = env!("CARGO_PKG_VERSION");
diff --git a/rust/src/server/src/grpc/control.rs b/rust/src/server/src/grpc/control.rs
index d1b5ac4fe0..72e52e5a8f 100644
--- a/rust/src/server/src/grpc/control.rs
+++ b/rust/src/server/src/grpc/control.rs
@@ -216,6 +216,7 @@ impl pb::control_server::Control for ControlServiceImpl {
             max_batched_tokens: ready.max_num_batched_tokens,
             max_loras: ready.max_loras,
             rl_capabilities: Some(self.rl_capabilities()),
+            proto_version: pb::VERSION.to_string(),
             effective_attention_block_size: self.client().effective_attention_block_size(),
         }))
     }
diff --git a/rust/src/server/src/grpc/tests.rs b/rust/src/server/src/grpc/tests.rs
index f46c658b01..fa041bdf33 100644
--- a/rust/src/server/src/grpc/tests.rs
+++ b/rust/src/server/src/grpc/tests.rs
@@ -2013,6 +2013,7 @@ async fn control_reports_server_and_model_info() {
     assert_eq!(server.engine_version, "test-vllm-version");
     assert_eq!(server.effective_attention_block_size, Some(64));
     assert_eq!(server.api_version, "vllm");
+    assert_eq!(server.proto_version, pb::VERSION);
     assert_eq!(server.instance_id, "test-instance");
     assert_eq!(server.max_model_len, DEFAULT_MOCK_MAX_MODEL_LEN as u32);
     assert_eq!(server.kv_block_size, DEFAULT_MOCK_BLOCK_SIZE as u32);