diff --git a/.release-please-manifest.json b/.release-please-manifest.json
index fdce8724..3cb6257c 100644
--- a/.release-please-manifest.json
+++ b/.release-please-manifest.json
@@ -1,3 +1,3 @@
 {
-  ".": "0.62.0"
+  ".": "0.63.0"
 }
\ No newline at end of file
diff --git a/.stats.yml b/.stats.yml
index ff6b8935..6359ad06 100644
--- a/.stats.yml
+++ b/.stats.yml
@@ -1,4 +1,4 @@
 configured_endpoints: 232
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai/openai-371f497afe4d6070f6e252e5febbe8f453c7058a8dff0c26a01b4d88442a4ac2.yml
-openapi_spec_hash: d39f46e8fda45f77096448105efd175a
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai/openai-50d816559ef0935e64d07789ff936a2b762e26ab0714a2fa6bc06d06d4484294.yml
+openapi_spec_hash: c5d8f37edbf66c1fef627d787b4c54fd
 config_hash: b64135fff1fe9cf4069b9ecf59ae8b07
diff --git a/CHANGELOG.md b/CHANGELOG.md
index e6b669e8..812bd395 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,13 @@
 # Changelog
 
+## 0.63.0 (2026-05-13)
+
+Full Changelog: [v0.62.0...v0.63.0](https://github.com/openai/openai-ruby/compare/v0.62.0...v0.63.0)
+
+### Features
+
+* **api:** add service_tier parameter to responses compact method ([b0ed0cb](https://github.com/openai/openai-ruby/commit/b0ed0cbce8afbf6a2d099dbfed4c4d734ac10d0d))
+
 ## 0.62.0 (2026-05-07)
 
 Full Changelog: [v0.61.0...v0.62.0](https://github.com/openai/openai-ruby/compare/v0.61.0...v0.62.0)
diff --git a/Gemfile.lock b/Gemfile.lock
index d0ebeb70..33c3e46e 100644
--- a/Gemfile.lock
+++ b/Gemfile.lock
@@ -11,7 +11,7 @@ GIT
 PATH
   remote: .
   specs:
-    openai (0.62.0)
+    openai (0.63.0)
       base64
       cgi
       connection_pool
diff --git a/README.md b/README.md
index 916c1753..27260e3c 100644
--- a/README.md
+++ b/README.md
@@ -15,7 +15,7 @@ To use this gem, install via Bundler by adding the following to your application
 <!-- x-release-please-start-version -->
 
 ```ruby
-gem "openai", "~> 0.62.0"
+gem "openai", "~> 0.63.0"
 ```
 
 <!-- x-release-please-end -->
@@ -111,6 +111,7 @@ Note that you can also pass a raw `IO` descriptor, but this disables retries, as
 
 For secure, automated environments like cloud-managed Kubernetes, Azure, and GCP, you can use workload identity authentication with short-lived tokens from cloud identity providers instead of long-lived API keys.
 
+`client_id` remains available as an optional parameter for token exchange setups that require an explicit OAuth client ID.
 
 ### Kubernetes Service Account
 
@@ -121,7 +122,6 @@ require "openai"
 provider = OpenAI::Auth::SubjectTokenProviders::K8sServiceAccountTokenProvider.new
 
 workload_identity = OpenAI::Auth::WorkloadIdentity.new(
-  client_id: ENV["OAUTH_CLIENT_ID"], # This is the default and can be omitted
   identity_provider_id: ENV["IDENTITY_PROVIDER_ID"], # This is the default and can be omitted
   service_account_id: ENV["SERVICE_ACCOUNT_ID"], # This is the default and can be omitted
   provider: provider
@@ -143,7 +143,6 @@ response = client.chat.completions.create(
 provider = OpenAI::Auth::SubjectTokenProviders::AzureManagedIdentityTokenProvider.new
 
 workload_identity = OpenAI::Auth::WorkloadIdentity.new(
-  client_id: ENV["OAUTH_CLIENT_ID"], # This is the default and can be omitted
   identity_provider_id: ENV["IDENTITY_PROVIDER_ID"], # This is the default and can be omitted
   service_account_id: ENV["SERVICE_ACCOUNT_ID"], # This is the default and can be omitted
   provider: provider
@@ -160,7 +159,6 @@ client = OpenAI::Client.new(
 provider = OpenAI::Auth::SubjectTokenProviders::GCPIDTokenProvider.new
 
 workload_identity = OpenAI::Auth::WorkloadIdentity.new(
-  client_id: ENV["OAUTH_CLIENT_ID"], # This is the default and can be omitted
   identity_provider_id: ENV["IDENTITY_PROVIDER_ID"], # This is the default and can be omitted
   service_account_id: ENV["SERVICE_ACCOUNT_ID"], # This is the default and can be omitted
   provider: provider
@@ -191,7 +189,6 @@ end
 provider = CustomProvider.new
 
 workload_identity = OpenAI::Auth::WorkloadIdentity.new(
-  client_id: ENV["OAUTH_CLIENT_ID"], # This is the default and can be omitted
   identity_provider_id: ENV["IDENTITY_PROVIDER_ID"], # This is the default and can be omitted
   service_account_id: ENV["SERVICE_ACCOUNT_ID"], # This is the default and can be omitted
   provider: provider
diff --git a/lib/openai/auth/workload_identity.rb b/lib/openai/auth/workload_identity.rb
index c05a1e9b..87a645f9 100644
--- a/lib/openai/auth/workload_identity.rb
+++ b/lib/openai/auth/workload_identity.rb
@@ -6,13 +6,13 @@ class WorkloadIdentity
       attr_reader :client_id, :identity_provider_id, :service_account_id, :provider, :refresh_buffer_seconds
 
       def initialize(
-        client_id:,
         identity_provider_id:,
         service_account_id:,
         provider:,
+        client_id: nil,
         refresh_buffer_seconds: 1200
       )
-        @client_id = client_id.to_s
+        @client_id = client_id&.to_s
         @identity_provider_id = identity_provider_id.to_s
         @service_account_id = service_account_id.to_s
         @provider = provider
diff --git a/lib/openai/auth/workload_identity_auth.rb b/lib/openai/auth/workload_identity_auth.rb
index 22a08d10..c5c257f0 100644
--- a/lib/openai/auth/workload_identity_auth.rb
+++ b/lib/openai/auth/workload_identity_auth.rb
@@ -92,14 +92,15 @@ def fetch_token_from_exchange
 
         request = Net::HTTP::Post.new(@token_exchange_url)
         request["Content-Type"] = "application/json"
-        request.body = JSON.generate(
+        body = {
           grant_type: TOKEN_EXCHANGE_GRANT_TYPE,
-          client_id: @config.client_id,
           subject_token: subject_token,
           subject_token_type: subject_token_type,
           identity_provider_id: @config.identity_provider_id,
           service_account_id: @config.service_account_id
-        )
+        }
+        body[:client_id] = @config.client_id unless @config.client_id.nil?
+        request.body = JSON.generate(body)
 
         response = Net::HTTP.start(
           @token_exchange_url.hostname,
diff --git a/lib/openai/models/responses/response_compact_params.rb b/lib/openai/models/responses/response_compact_params.rb
index a30231d6..62f97f11 100644
--- a/lib/openai/models/responses/response_compact_params.rb
+++ b/lib/openai/models/responses/response_compact_params.rb
@@ -56,7 +56,13 @@ class ResponseCompactParams < OpenAI::Internal::Type::BaseModel
                  enum: -> { OpenAI::Responses::ResponseCompactParams::PromptCacheRetention },
                  nil?: true
 
-        # @!method initialize(model:, input: nil, instructions: nil, previous_response_id: nil, prompt_cache_key: nil, prompt_cache_retention: nil, request_options: {})
+        # @!attribute service_tier
+        #   The service tier to use for this request.
+        #
+        #   @return [Symbol, OpenAI::Models::Responses::ResponseCompactParams::ServiceTier, nil]
+        optional :service_tier, enum: -> { OpenAI::Responses::ResponseCompactParams::ServiceTier }, nil?: true
+
+        # @!method initialize(model:, input: nil, instructions: nil, previous_response_id: nil, prompt_cache_key: nil, prompt_cache_retention: nil, service_tier: nil, request_options: {})
         #   Some parameter documentations has been truncated, see
         #   {OpenAI::Models::Responses::ResponseCompactParams} for more details.
         #
@@ -72,6 +78,8 @@ class ResponseCompactParams < OpenAI::Internal::Type::BaseModel
         #
         #   @param prompt_cache_retention [Symbol, OpenAI::Models::Responses::ResponseCompactParams::PromptCacheRetention, nil] How long to retain a prompt cache entry created by this request.
         #
+        #   @param service_tier [Symbol, OpenAI::Models::Responses::ResponseCompactParams::ServiceTier, nil] The service tier to use for this request.
+        #
         #   @param request_options [OpenAI::RequestOptions, Hash{Symbol=>Object}]
 
         # Model ID used to generate the response, like `gpt-5` or `o3`. OpenAI offers a
@@ -401,6 +409,19 @@ module PromptCacheRetention
           # @!method self.values
           #   @return [Array<Symbol>]
         end
+
+        # The service tier to use for this request.
+        module ServiceTier
+          extend OpenAI::Internal::Type::Enum
+
+          AUTO = :auto
+          DEFAULT = :default
+          FLEX = :flex
+          PRIORITY = :priority
+
+          # @!method self.values
+          #   @return [Array<Symbol>]
+        end
       end
     end
   end
diff --git a/lib/openai/resources/responses.rb b/lib/openai/resources/responses.rb
index cd4e73e8..2e4afd34 100644
--- a/lib/openai/resources/responses.rb
+++ b/lib/openai/resources/responses.rb
@@ -483,7 +483,7 @@ def cancel(response_id, params = {})
       # For ZDR-compatible compaction details, see
       # [Compaction (advanced)](https://platform.openai.com/docs/guides/conversation-state#compaction-advanced).
       #
-      # @overload compact(model:, input: nil, instructions: nil, previous_response_id: nil, prompt_cache_key: nil, prompt_cache_retention: nil, request_options: {})
+      # @overload compact(model:, input: nil, instructions: nil, previous_response_id: nil, prompt_cache_key: nil, prompt_cache_retention: nil, service_tier: nil, request_options: {})
       #
       # @param model [Symbol, String, OpenAI::Models::Responses::ResponseCompactParams::Model, nil] Model ID used to generate the response, like `gpt-5` or `o3`. OpenAI offers a wi
       #
@@ -497,6 +497,8 @@ def cancel(response_id, params = {})
       #
       # @param prompt_cache_retention [Symbol, OpenAI::Models::Responses::ResponseCompactParams::PromptCacheRetention, nil] How long to retain a prompt cache entry created by this request.
       #
+      # @param service_tier [Symbol, OpenAI::Models::Responses::ResponseCompactParams::ServiceTier, nil] The service tier to use for this request.
+      #
       # @param request_options [OpenAI::RequestOptions, Hash{Symbol=>Object}, nil]
       #
       # @return [OpenAI::Models::Responses::CompactedResponse]
diff --git a/lib/openai/version.rb b/lib/openai/version.rb
index 74b41238..2d5c6103 100644
--- a/lib/openai/version.rb
+++ b/lib/openai/version.rb
@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module OpenAI
-  VERSION = "0.62.0"
+  VERSION = "0.63.0"
 end
diff --git a/rbi/openai/auth.rbi b/rbi/openai/auth.rbi
index f3d217c2..e8f7fbdb 100644
--- a/rbi/openai/auth.rbi
+++ b/rbi/openai/auth.rbi
@@ -18,7 +18,7 @@ module OpenAI
     end
 
     class WorkloadIdentity
-      sig { returns(String) }
+      sig { returns(T.nilable(String)) }
       attr_reader :client_id
 
       sig { returns(String) }
@@ -35,18 +35,18 @@ module OpenAI
 
       sig do
         params(
-          client_id: T.any(String, Symbol),
           identity_provider_id: T.any(String, Symbol),
           service_account_id: T.any(String, Symbol),
           provider: SubjectTokenProvider,
+          client_id: T.nilable(T.any(String, Symbol)),
           refresh_buffer_seconds: Integer
         ).void
       end
       def initialize(
-        client_id:,
         identity_provider_id:,
         service_account_id:,
         provider:,
+        client_id: nil,
         refresh_buffer_seconds: 1200
       )
       end
diff --git a/rbi/openai/models/responses/response_compact_params.rbi b/rbi/openai/models/responses/response_compact_params.rbi
index 46564650..395bcb96 100644
--- a/rbi/openai/models/responses/response_compact_params.rbi
+++ b/rbi/openai/models/responses/response_compact_params.rbi
@@ -68,6 +68,16 @@ module OpenAI
         end
         attr_accessor :prompt_cache_retention
 
+        # The service tier to use for this request.
+        sig do
+          returns(
+            T.nilable(
+              OpenAI::Responses::ResponseCompactParams::ServiceTier::OrSymbol
+            )
+          )
+        end
+        attr_accessor :service_tier
+
         sig do
           params(
             model:
@@ -88,6 +98,10 @@ module OpenAI
               T.nilable(
                 OpenAI::Responses::ResponseCompactParams::PromptCacheRetention::OrSymbol
               ),
+            service_tier:
+              T.nilable(
+                OpenAI::Responses::ResponseCompactParams::ServiceTier::OrSymbol
+              ),
             request_options: OpenAI::RequestOptions::OrHash
           ).returns(T.attached_class)
         end
@@ -114,6 +128,8 @@ module OpenAI
           prompt_cache_key: nil,
           # How long to retain a prompt cache entry created by this request.
           prompt_cache_retention: nil,
+          # The service tier to use for this request.
+          service_tier: nil,
           request_options: {}
         )
         end
@@ -139,6 +155,10 @@ module OpenAI
                 T.nilable(
                   OpenAI::Responses::ResponseCompactParams::PromptCacheRetention::OrSymbol
                 ),
+              service_tier:
+                T.nilable(
+                  OpenAI::Responses::ResponseCompactParams::ServiceTier::OrSymbol
+                ),
               request_options: OpenAI::RequestOptions
             }
           )
@@ -705,6 +725,51 @@ module OpenAI
           def self.values
           end
         end
+
+        # The service tier to use for this request.
+        module ServiceTier
+          extend OpenAI::Internal::Type::Enum
+
+          TaggedSymbol =
+            T.type_alias do
+              T.all(
+                Symbol,
+                OpenAI::Responses::ResponseCompactParams::ServiceTier
+              )
+            end
+          OrSymbol = T.type_alias { T.any(Symbol, String) }
+
+          AUTO =
+            T.let(
+              :auto,
+              OpenAI::Responses::ResponseCompactParams::ServiceTier::TaggedSymbol
+            )
+          DEFAULT =
+            T.let(
+              :default,
+              OpenAI::Responses::ResponseCompactParams::ServiceTier::TaggedSymbol
+            )
+          FLEX =
+            T.let(
+              :flex,
+              OpenAI::Responses::ResponseCompactParams::ServiceTier::TaggedSymbol
+            )
+          PRIORITY =
+            T.let(
+              :priority,
+              OpenAI::Responses::ResponseCompactParams::ServiceTier::TaggedSymbol
+            )
+
+          sig do
+            override.returns(
+              T::Array[
+                OpenAI::Responses::ResponseCompactParams::ServiceTier::TaggedSymbol
+              ]
+            )
+          end
+          def self.values
+          end
+        end
       end
     end
   end
diff --git a/rbi/openai/resources/responses.rbi b/rbi/openai/resources/responses.rbi
index 6542cb5a..167d04b2 100644
--- a/rbi/openai/resources/responses.rbi
+++ b/rbi/openai/resources/responses.rbi
@@ -994,6 +994,10 @@ module OpenAI
             T.nilable(
               OpenAI::Responses::ResponseCompactParams::PromptCacheRetention::OrSymbol
             ),
+          service_tier:
+            T.nilable(
+              OpenAI::Responses::ResponseCompactParams::ServiceTier::OrSymbol
+            ),
           request_options: OpenAI::RequestOptions::OrHash
         ).returns(OpenAI::Responses::CompactedResponse)
       end
@@ -1020,6 +1024,8 @@ module OpenAI
         prompt_cache_key: nil,
         # How long to retain a prompt cache entry created by this request.
         prompt_cache_retention: nil,
+        # The service tier to use for this request.
+        service_tier: nil,
         request_options: {}
       )
       end
diff --git a/sig/openai/models/responses/response_compact_params.rbs b/sig/openai/models/responses/response_compact_params.rbs
index 14c59c83..226c79e2 100644
--- a/sig/openai/models/responses/response_compact_params.rbs
+++ b/sig/openai/models/responses/response_compact_params.rbs
@@ -8,7 +8,8 @@ module OpenAI
           instructions: String?,
           previous_response_id: String?,
           prompt_cache_key: String?,
-          prompt_cache_retention: OpenAI::Models::Responses::ResponseCompactParams::prompt_cache_retention?
+          prompt_cache_retention: OpenAI::Models::Responses::ResponseCompactParams::prompt_cache_retention?,
+          service_tier: OpenAI::Models::Responses::ResponseCompactParams::service_tier?
         }
         & OpenAI::Internal::Type::request_parameters
 
@@ -28,6 +29,8 @@ module OpenAI
 
         attr_accessor prompt_cache_retention: OpenAI::Models::Responses::ResponseCompactParams::prompt_cache_retention?
 
+        attr_accessor service_tier: OpenAI::Models::Responses::ResponseCompactParams::service_tier?
+
         def initialize: (
           model: OpenAI::Models::Responses::ResponseCompactParams::model?,
           ?input: OpenAI::Models::Responses::ResponseCompactParams::input?,
@@ -35,6 +38,7 @@ module OpenAI
           ?previous_response_id: String?,
           ?prompt_cache_key: String?,
           ?prompt_cache_retention: OpenAI::Models::Responses::ResponseCompactParams::prompt_cache_retention?,
+          ?service_tier: OpenAI::Models::Responses::ResponseCompactParams::service_tier?,
           ?request_options: OpenAI::request_opts
         ) -> void
 
@@ -45,6 +49,7 @@ module OpenAI
           previous_response_id: String?,
           prompt_cache_key: String?,
           prompt_cache_retention: OpenAI::Models::Responses::ResponseCompactParams::prompt_cache_retention?,
+          service_tier: OpenAI::Models::Responses::ResponseCompactParams::service_tier?,
           request_options: OpenAI::RequestOptions
         }
 
@@ -263,6 +268,19 @@ module OpenAI
 
           def self?.values: -> ::Array[OpenAI::Models::Responses::ResponseCompactParams::prompt_cache_retention]
         end
+
+        type service_tier = :auto | :default | :flex | :priority
+
+        module ServiceTier
+          extend OpenAI::Internal::Type::Enum
+
+          AUTO: :auto
+          DEFAULT: :default
+          FLEX: :flex
+          PRIORITY: :priority
+
+          def self?.values: -> ::Array[OpenAI::Models::Responses::ResponseCompactParams::service_tier]
+        end
       end
     end
   end
diff --git a/sig/openai/resources/responses.rbs b/sig/openai/resources/responses.rbs
index 88fa787f..de05b4b1 100644
--- a/sig/openai/resources/responses.rbs
+++ b/sig/openai/resources/responses.rbs
@@ -130,6 +130,7 @@ module OpenAI
         ?previous_response_id: String?,
         ?prompt_cache_key: String?,
         ?prompt_cache_retention: OpenAI::Models::Responses::ResponseCompactParams::prompt_cache_retention?,
+        ?service_tier: OpenAI::Models::Responses::ResponseCompactParams::service_tier?,
         ?request_options: OpenAI::request_opts
       ) -> OpenAI::Responses::CompactedResponse
 
diff --git a/test/openai/auth/workload_identity_test.rb b/test/openai/auth/workload_identity_test.rb
index 0731284a..78d55fb4 100644
--- a/test/openai/auth/workload_identity_test.rb
+++ b/test/openai/auth/workload_identity_test.rb
@@ -119,6 +119,39 @@ def test_workload_identity_auth_token_exchange
       token_path: @token_path
     )
 
+    stub_request(:post, "https://auth.openai.com/oauth/token")
+      .with do |request|
+        JSON.parse(request.body) == {
+          "grant_type" => "urn:ietf:params:oauth:grant-type:token-exchange",
+          "subject_token" => "k8s-jwt-token",
+          "subject_token_type" => "urn:ietf:params:oauth:token-type:jwt",
+          "identity_provider_id" => "idp-123",
+          "service_account_id" => "sa-456"
+        }
+      end
+      .to_return(
+        status: 200,
+        body: JSON.generate({"access_token" => "oauth-access-token", "expires_in" => 3600})
+      )
+
+    config = OpenAI::Auth::WorkloadIdentity.new(
+      identity_provider_id: "idp-123",
+      service_account_id: "sa-456",
+      provider: provider
+    )
+
+    auth = OpenAI::Auth::WorkloadIdentityAuth.new(config, "org-123")
+    token = auth.get_token
+
+    assert_equal("oauth-access-token", token)
+  end
+
+  def test_workload_identity_auth_token_exchange_with_optional_client_id
+    File.write(@token_path, "k8s-jwt-token")
+    provider = OpenAI::Auth::SubjectTokenProviders::K8sServiceAccountTokenProvider.new(
+      token_path: @token_path
+    )
+
     stub_request(:post, "https://auth.openai.com/oauth/token")
       .with(
         body: hash_including(
@@ -136,10 +169,10 @@ def test_workload_identity_auth_token_exchange
       )
 
     config = OpenAI::Auth::WorkloadIdentity.new(
-      client_id: "test-client",
       identity_provider_id: "idp-123",
       service_account_id: "sa-456",
-      provider: provider
+      provider: provider,
+      client_id: "test-client"
     )
 
     auth = OpenAI::Auth::WorkloadIdentityAuth.new(config, "org-123")
@@ -166,7 +199,6 @@ def test_workload_identity_auth_oauth_error
       )
 
     config = OpenAI::Auth::WorkloadIdentity.new(
-      client_id: "bad-client",
       identity_provider_id: "idp-123",
       service_account_id: "sa-456",
       provider: provider
@@ -190,7 +222,6 @@ def test_workload_identity_client_initialization
     )
 
     config = OpenAI::Auth::WorkloadIdentity.new(
-      client_id: "test-client",
       identity_provider_id: "idp-123",
       service_account_id: "sa-456",
       provider: provider
@@ -210,7 +241,6 @@ def test_workload_identity_client_initialization
   def test_workload_identity_mutually_exclusive_with_api_key
     provider = OpenAI::Auth::SubjectTokenProviders::K8sServiceAccountTokenProvider.new
     config = OpenAI::Auth::WorkloadIdentity.new(
-      client_id: "test-client",
       identity_provider_id: "idp-123",
       service_account_id: "sa-456",
       provider: provider
@@ -233,7 +263,6 @@ def test_workload_identity_preserves_admin_auth_requests
       token_path: @token_path
     )
     config = OpenAI::Auth::WorkloadIdentity.new(
-      client_id: "test-client",
       identity_provider_id: "idp-123",
       service_account_id: "sa-456",
       provider: provider
@@ -316,7 +345,6 @@ def test_401_retry_with_token_invalidation
       )
 
     config = OpenAI::Auth::WorkloadIdentity.new(
-      client_id: "test-client",
       identity_provider_id: "idp-123",
       service_account_id: "sa-456",
       provider: provider
@@ -379,7 +407,6 @@ def test_workload_identity_token_caching
       )
 
     config = OpenAI::Auth::WorkloadIdentity.new(
-      client_id: "test-client",
       identity_provider_id: "idp-123",
       service_account_id: "sa-456",
       provider: provider