openai 0.10.0 → 0.11.0

data/lib/openai/resources/responses.rb

@@ -23,7 +23,7 @@ module OpenAI
       # [file search](https://platform.openai.com/docs/guides/tools-file-search) to use
       # your own data as input for the model's response.
       #
-      # @overload create(background: nil, include: nil, input: nil, instructions: nil, max_output_tokens: nil, metadata: nil, model: nil, parallel_tool_calls: nil, previous_response_id: nil, prompt: nil, reasoning: nil, service_tier: nil, store: nil, temperature: nil, text: nil, tool_choice: nil, tools: nil, top_p: nil, truncation: nil, user: nil, request_options: {})
+      # @overload create(background: nil, include: nil, input: nil, instructions: nil, max_output_tokens: nil, max_tool_calls: nil, metadata: nil, model: nil, parallel_tool_calls: nil, previous_response_id: nil, prompt: nil, reasoning: nil, service_tier: nil, store: nil, temperature: nil, text: nil, tool_choice: nil, tools: nil, top_logprobs: nil, top_p: nil, truncation: nil, user: nil, request_options: {})
       #
       # @param background [Boolean, nil] Whether to run the model response in the background.
       #
@@ -35,6 +35,8 @@ module OpenAI
       #
       # @param max_output_tokens [Integer, nil] An upper bound for the number of tokens that can be generated for a response, in
       #
+      # @param max_tool_calls [Integer, nil] The maximum number of total calls to built-in tools that can be processed in a r
+      #
       # @param metadata [Hash{Symbol=>String}, nil] Set of 16 key-value pairs that can be attached to an object. This can be
       #
       # @param model [String, Symbol, OpenAI::Models::ChatModel, OpenAI::Models::ResponsesModel::ResponsesOnlyModel] Model ID used to generate the response, like `gpt-4o` or `o3`. OpenAI
@@ -47,7 +49,7 @@ module OpenAI
       #
       # @param reasoning [OpenAI::Models::Reasoning, nil] **o-series models only**
       #
-      # @param service_tier [Symbol, OpenAI::Models::Responses::ResponseCreateParams::ServiceTier, nil] Specifies the latency tier to use for processing the request. This parameter is
+      # @param service_tier [Symbol, OpenAI::Models::Responses::ResponseCreateParams::ServiceTier, nil] Specifies the processing type used for serving the request.
       #
       # @param store [Boolean, nil] Whether to store the generated model response for later retrieval via
       #
@@ -55,10 +57,12 @@ module OpenAI
       #
       # @param text [OpenAI::Models::Responses::ResponseTextConfig] Configuration options for a text response from the model. Can be plain
       #
-      # @param tool_choice [Symbol, OpenAI::Models::Responses::ToolChoiceOptions, OpenAI::Models::Responses::ToolChoiceTypes, OpenAI::Models::Responses::ToolChoiceFunction] How the model should select which tool (or tools) to use when generating
+      # @param tool_choice [Symbol, OpenAI::Models::Responses::ToolChoiceOptions, OpenAI::Models::Responses::ToolChoiceTypes, OpenAI::Models::Responses::ToolChoiceFunction, OpenAI::Models::Responses::ToolChoiceMcp] How the model should select which tool (or tools) to use when generating
       #
       # @param tools [Array<OpenAI::Models::Responses::FunctionTool, OpenAI::Models::Responses::FileSearchTool, OpenAI::Models::Responses::ComputerTool, OpenAI::Models::Responses::Tool::Mcp, OpenAI::Models::Responses::Tool::CodeInterpreter, OpenAI::Models::Responses::Tool::ImageGeneration, OpenAI::Models::Responses::Tool::LocalShell, OpenAI::Models::Responses::WebSearchTool>] An array of tools the model may call while generating a response. You
       #
+      # @param top_logprobs [Integer, nil] An integer between 0 and 20 specifying the number of most likely tokens to
+      #
       # @param top_p [Float, nil] An alternative to sampling with temperature, called nucleus sampling,
       #
       # @param truncation [Symbol, OpenAI::Models::Responses::ResponseCreateParams::Truncation, nil] The truncation strategy to use for the model response.
@@ -183,7 +187,7 @@ module OpenAI
       # [file search](https://platform.openai.com/docs/guides/tools-file-search) to use
       # your own data as input for the model's response.
       #
-      # @overload stream_raw(background: nil, include: nil, input: nil, instructions: nil, max_output_tokens: nil, metadata: nil, model: nil, parallel_tool_calls: nil, previous_response_id: nil, prompt: nil, reasoning: nil, service_tier: nil, store: nil, temperature: nil, text: nil, tool_choice: nil, tools: nil, top_p: nil, truncation: nil, user: nil, request_options: {})
+      # @overload stream_raw(background: nil, include: nil, input: nil, instructions: nil, max_output_tokens: nil, max_tool_calls: nil, metadata: nil, model: nil, parallel_tool_calls: nil, previous_response_id: nil, prompt: nil, reasoning: nil, service_tier: nil, store: nil, temperature: nil, text: nil, tool_choice: nil, tools: nil, top_logprobs: nil, top_p: nil, truncation: nil, user: nil, request_options: {})
       #
       # @param background [Boolean, nil] Whether to run the model response in the background.
       #
@@ -195,6 +199,8 @@ module OpenAI
       #
       # @param max_output_tokens [Integer, nil] An upper bound for the number of tokens that can be generated for a response, in
       #
+      # @param max_tool_calls [Integer, nil] The maximum number of total calls to built-in tools that can be processed in a r
+      #
       # @param metadata [Hash{Symbol=>String}, nil] Set of 16 key-value pairs that can be attached to an object. This can be
       #
       # @param model [String, Symbol, OpenAI::Models::ChatModel, OpenAI::Models::ResponsesModel::ResponsesOnlyModel] Model ID used to generate the response, like `gpt-4o` or `o3`. OpenAI
@@ -207,7 +213,7 @@ module OpenAI
       #
       # @param reasoning [OpenAI::Models::Reasoning, nil] **o-series models only**
       #
-      # @param service_tier [Symbol, OpenAI::Models::Responses::ResponseCreateParams::ServiceTier, nil] Specifies the latency tier to use for processing the request. This parameter is
+      # @param service_tier [Symbol, OpenAI::Models::Responses::ResponseCreateParams::ServiceTier, nil] Specifies the processing type used for serving the request.
       #
       # @param store [Boolean, nil] Whether to store the generated model response for later retrieval via
       #
@@ -215,10 +221,12 @@ module OpenAI
       #
       # @param text [OpenAI::Models::Responses::ResponseTextConfig] Configuration options for a text response from the model. Can be plain
       #
-      # @param tool_choice [Symbol, OpenAI::Models::Responses::ToolChoiceOptions, OpenAI::Models::Responses::ToolChoiceTypes, OpenAI::Models::Responses::ToolChoiceFunction] How the model should select which tool (or tools) to use when generating
+      # @param tool_choice [Symbol, OpenAI::Models::Responses::ToolChoiceOptions, OpenAI::Models::Responses::ToolChoiceTypes, OpenAI::Models::Responses::ToolChoiceFunction, OpenAI::Models::Responses::ToolChoiceMcp] How the model should select which tool (or tools) to use when generating
       #
       # @param tools [Array<OpenAI::Models::Responses::FunctionTool, OpenAI::Models::Responses::FileSearchTool, OpenAI::Models::Responses::ComputerTool, OpenAI::Models::Responses::Tool::Mcp, OpenAI::Models::Responses::Tool::CodeInterpreter, OpenAI::Models::Responses::Tool::ImageGeneration, OpenAI::Models::Responses::Tool::LocalShell, OpenAI::Models::Responses::WebSearchTool>] An array of tools the model may call while generating a response. You
       #
+      # @param top_logprobs [Integer, nil] An integer between 0 and 20 specifying the number of most likely tokens to
+      #
       # @param top_p [Float, nil] An alternative to sampling with temperature, called nucleus sampling,
       #
       # @param truncation [Symbol, OpenAI::Models::Responses::ResponseCreateParams::Truncation, nil] The truncation strategy to use for the model response.

data/lib/openai/resources/webhooks.rb

@@ -0,0 +1,124 @@
+# frozen_string_literal: true
+
+require "openssl"
+require "base64"
+
+module OpenAI
+  module Resources
+    class Webhooks
+      # Validates that the given payload was sent by OpenAI and parses the payload.
+      #
+      # @param payload [String] The raw webhook payload as a string
+      # @param headers [Hash] The webhook headers
+      # @param webhook_secret [String, nil] The webhook secret (optional, will use client webhook secret or ENV["OPENAI_WEBHOOK_SECRET"] if not provided)
+      #
+      # @return [OpenAI::Models::Webhooks::BatchCancelledWebhookEvent, OpenAI::Models::Webhooks::BatchCompletedWebhookEvent, OpenAI::Models::Webhooks::BatchExpiredWebhookEvent, OpenAI::Models::Webhooks::BatchFailedWebhookEvent, OpenAI::Models::Webhooks::EvalRunCanceledWebhookEvent, OpenAI::Models::Webhooks::EvalRunFailedWebhookEvent, OpenAI::Models::Webhooks::EvalRunSucceededWebhookEvent, OpenAI::Models::Webhooks::FineTuningJobCancelledWebhookEvent, OpenAI::Models::Webhooks::FineTuningJobFailedWebhookEvent, OpenAI::Models::Webhooks::FineTuningJobSucceededWebhookEvent, OpenAI::Models::Webhooks::ResponseCancelledWebhookEvent, OpenAI::Models::Webhooks::ResponseCompletedWebhookEvent, OpenAI::Models::Webhooks::ResponseCreatedWebhookEvent, OpenAI::Models::Webhooks::ResponseFailedWebhookEvent, OpenAI::Models::Webhooks::ResponseIncompleteWebhookEvent]
+      #
+      # @raise [ArgumentError] if signature verification fails
+      def unwrap(
+        payload,
+        headers = {},
+        webhook_secret = @client.webhook_secret || ENV["OPENAI_WEBHOOK_SECRET"]
+      )
+        verify_signature(payload, headers, webhook_secret)
+
+        parsed = JSON.parse(payload, symbolize_names: true)
+        OpenAI::Internal::Type::Converter.coerce(OpenAI::Models::Webhooks::UnwrapWebhookEvent, parsed)
+      end
+
+      # Validates whether or not the webhook payload was sent by OpenAI.
+      #
+      # @param payload [String] The webhook payload as a string
+      # @param headers [Hash] The webhook headers
+      # @param webhook_secret [String, nil] The webhook secret (optional, will use client webhook secret or ENV["OPENAI_WEBHOOK_SECRET"] if not provided)
+      # @param tolerance [Integer] Maximum age of the webhook in seconds (default: 300 = 5 minutes)
+      #
+      # @raise [ArgumentError] if the signature is invalid
+      def verify_signature(
+        payload,
+        headers,
+        webhook_secret = @client.webhook_secret || ENV["OPENAI_WEBHOOK_SECRET"],
+        tolerance = 300
+      )
+        if webhook_secret.nil?
+          raise ArgumentError,
+                "The webhook secret must either be set using the env var, OPENAI_WEBHOOK_SECRET, " \
+                "or passed to this function"
+        end
+
+        # Extract required headers
+        signature_header = headers["webhook-signature"] || headers[:webhook_signature]
+        timestamp_header = headers["webhook-timestamp"] || headers[:webhook_timestamp]
+        webhook_id = headers["webhook-id"] || headers[:webhook_id]
+
+        if signature_header.nil?
+          raise ArgumentError, "Missing required webhook-signature header"
+        end
+
+        if timestamp_header.nil?
+          raise ArgumentError, "Missing required webhook-timestamp header"
+        end
+
+        if webhook_id.nil?
+          raise ArgumentError, "Missing required webhook-id header"
+        end
+
+        # Validate timestamp to prevent replay attacks
+        begin
+          timestamp_seconds = timestamp_header.to_i
+        rescue ArgumentError
+          raise ArgumentError, "Invalid webhook timestamp format"
+        end
+
+        now = Time.now.to_i
+
+        if now - timestamp_seconds > tolerance
+          raise OpenAI::Errors::InvalidWebhookSignatureError, "Webhook timestamp is too old"
+        end
+
+        if timestamp_seconds > now + tolerance
+          raise OpenAI::Errors::InvalidWebhookSignatureError, "Webhook timestamp is too new"
+        end
+
+        # Extract signatures from v1,<base64> format
+        # The signature header can have multiple values, separated by spaces.
+        # Each value is in the format v1,<base64>. We should accept if any match.
+        signatures = signature_header.split.map do |part|
+          if part.start_with?("v1,")
+            part[3..]
+          else
+            part
+          end
+        end
+
+        # Decode the secret if it starts with whsec_
+        decoded_secret = if webhook_secret.start_with?("whsec_")
+          Base64.decode64(webhook_secret[6..])
+        else
+          webhook_secret
+        end
+
+        # Create the signed payload: {webhook_id}.{timestamp}.{payload}
+        signed_payload = "#{webhook_id}.#{timestamp_header}.#{payload}"
+
+        # Compute HMAC-SHA256 signature
+        expected_signature = Base64.encode64(
+          OpenSSL::HMAC.digest("sha256", decoded_secret, signed_payload)
+        ).strip
+
+        # Accept if any signature matches using timing-safe comparison
+        return if signatures.any? { |signature| OpenSSL.secure_compare(expected_signature, signature) }
+
+        raise OpenAI::Errors::InvalidWebhookSignatureError,
+              "The given webhook signature does not match the expected signature"
+      end
+
+      # @api private
+      #
+      # @param client [OpenAI::Client]
+      def initialize(client:)
+        @client = client
+      end
+    end
+  end
+end

data/lib/openai/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module OpenAI
-  VERSION = "0.10.0"
+  VERSION = "0.11.0"
 end

data/lib/openai.rb

@@ -441,6 +441,7 @@ require_relative "openai/models/responses/response_web_search_call_in_progress_e
 require_relative "openai/models/responses/response_web_search_call_searching_event"
 require_relative "openai/models/responses/tool"
 require_relative "openai/models/responses/tool_choice_function"
+require_relative "openai/models/responses/tool_choice_mcp"
 require_relative "openai/models/responses/tool_choice_options"
 require_relative "openai/models/responses/tool_choice_types"
 require_relative "openai/models/responses/web_search_tool"
@@ -477,6 +478,22 @@ require_relative "openai/models/vector_stores/vector_store_file_deleted"
 require_relative "openai/models/vector_store_search_params"
 require_relative "openai/models/vector_store_search_response"
 require_relative "openai/models/vector_store_update_params"
+require_relative "openai/models/webhooks/batch_cancelled_webhook_event"
+require_relative "openai/models/webhooks/batch_completed_webhook_event"
+require_relative "openai/models/webhooks/batch_expired_webhook_event"
+require_relative "openai/models/webhooks/batch_failed_webhook_event"
+require_relative "openai/models/webhooks/eval_run_canceled_webhook_event"
+require_relative "openai/models/webhooks/eval_run_failed_webhook_event"
+require_relative "openai/models/webhooks/eval_run_succeeded_webhook_event"
+require_relative "openai/models/webhooks/fine_tuning_job_cancelled_webhook_event"
+require_relative "openai/models/webhooks/fine_tuning_job_failed_webhook_event"
+require_relative "openai/models/webhooks/fine_tuning_job_succeeded_webhook_event"
+require_relative "openai/models/webhooks/response_cancelled_webhook_event"
+require_relative "openai/models/webhooks/response_completed_webhook_event"
+require_relative "openai/models/webhooks/response_failed_webhook_event"
+require_relative "openai/models/webhooks/response_incomplete_webhook_event"
+require_relative "openai/models/webhooks/unwrap_webhook_event"
+require_relative "openai/models/webhooks/webhook_unwrap_params"
 require_relative "openai/models"
 require_relative "openai/resources/audio"
 require_relative "openai/resources/audio/speech"
@@ -521,3 +538,4 @@ require_relative "openai/resources/uploads/parts"
 require_relative "openai/resources/vector_stores"
 require_relative "openai/resources/vector_stores/file_batches"
 require_relative "openai/resources/vector_stores/files"
+require_relative "openai/resources/webhooks"

data/rbi/openai/client.rbi

@@ -52,6 +52,9 @@ module OpenAI
     sig { returns(OpenAI::Resources::VectorStores) }
     attr_reader :vector_stores
 
+    sig { returns(OpenAI::Resources::Webhooks) }
+    attr_reader :webhooks
+
     sig { returns(OpenAI::Resources::Beta) }
     attr_reader :beta
 

data/rbi/openai/models/all_models.rbi

@@ -35,6 +35,26 @@ module OpenAI
             :"o3-pro-2025-06-10",
             OpenAI::AllModels::ResponsesOnlyModel::TaggedSymbol
           )
+        O3_DEEP_RESEARCH =
+          T.let(
+            :"o3-deep-research",
+            OpenAI::AllModels::ResponsesOnlyModel::TaggedSymbol
+          )
+        O3_DEEP_RESEARCH_2025_06_26 =
+          T.let(
+            :"o3-deep-research-2025-06-26",
+            OpenAI::AllModels::ResponsesOnlyModel::TaggedSymbol
+          )
+        O4_MINI_DEEP_RESEARCH =
+          T.let(
+            :"o4-mini-deep-research",
+            OpenAI::AllModels::ResponsesOnlyModel::TaggedSymbol
+          )
+        O4_MINI_DEEP_RESEARCH_2025_06_26 =
+          T.let(
+            :"o4-mini-deep-research-2025-06-26",
+            OpenAI::AllModels::ResponsesOnlyModel::TaggedSymbol
+          )
         COMPUTER_USE_PREVIEW =
           T.let(
             :"computer-use-preview",

data/rbi/openai/models/chat/chat_completion.rbi

@@ -32,23 +32,23 @@ module OpenAI
         sig { returns(Symbol) }
         attr_accessor :object
 
-        # Specifies the latency tier to use for processing the request. This parameter is
-        # relevant for customers subscribed to the scale tier service:
+        # Specifies the processing type used for serving the request.
         #
-        # - If set to 'auto', and the Project is Scale tier enabled, the system will
-        #   utilize scale tier credits until they are exhausted.
-        # - If set to 'auto', and the Project is not Scale tier enabled, the request will
-        #   be processed using the default service tier with a lower uptime SLA and no
-        #   latency guarantee.
-        # - If set to 'default', the request will be processed using the default service
-        #   tier with a lower uptime SLA and no latency guarantee.
-        # - If set to 'flex', the request will be processed with the Flex Processing
-        #   service tier.
-        #   [Learn more](https://platform.openai.com/docs/guides/flex-processing).
+        # - If set to 'auto', then the request will be processed with the service tier
+        #   configured in the Project settings. Unless otherwise configured, the Project
+        #   will use 'default'.
+        # - If set to 'default', then the requset will be processed with the standard
+        #   pricing and performance for the selected model.
+        # - If set to '[flex](https://platform.openai.com/docs/guides/flex-processing)' or
+        #   'priority', then the request will be processed with the corresponding service
+        #   tier. [Contact sales](https://openai.com/contact-sales) to learn more about
+        #   Priority processing.
         # - When not set, the default behavior is 'auto'.
         #
-        # When this parameter is set, the response body will include the `service_tier`
-        # utilized.
+        # When the `service_tier` parameter is set, the response body will include the
+        # `service_tier` value based on the processing mode actually used to serve the
+        # request. This response value may be different from the value set in the
+        # parameter.
         sig do
           returns(
             T.nilable(OpenAI::Chat::ChatCompletion::ServiceTier::TaggedSymbol)
@@ -98,23 +98,23 @@ module OpenAI
           created:,
           # The model used for the chat completion.
           model:,
-          # Specifies the latency tier to use for processing the request. This parameter is
-          # relevant for customers subscribed to the scale tier service:
+          # Specifies the processing type used for serving the request.
           #
-          # - If set to 'auto', and the Project is Scale tier enabled, the system will
-          #   utilize scale tier credits until they are exhausted.
-          # - If set to 'auto', and the Project is not Scale tier enabled, the request will
-          #   be processed using the default service tier with a lower uptime SLA and no
-          #   latency guarantee.
-          # - If set to 'default', the request will be processed using the default service
-          #   tier with a lower uptime SLA and no latency guarantee.
-          # - If set to 'flex', the request will be processed with the Flex Processing
-          #   service tier.
-          #   [Learn more](https://platform.openai.com/docs/guides/flex-processing).
+          # - If set to 'auto', then the request will be processed with the service tier
+          #   configured in the Project settings. Unless otherwise configured, the Project
+          #   will use 'default'.
+          # - If set to 'default', then the requset will be processed with the standard
+          #   pricing and performance for the selected model.
+          # - If set to '[flex](https://platform.openai.com/docs/guides/flex-processing)' or
+          #   'priority', then the request will be processed with the corresponding service
+          #   tier. [Contact sales](https://openai.com/contact-sales) to learn more about
+          #   Priority processing.
           # - When not set, the default behavior is 'auto'.
           #
-          # When this parameter is set, the response body will include the `service_tier`
-          # utilized.
+          # When the `service_tier` parameter is set, the response body will include the
+          # `service_tier` value based on the processing mode actually used to serve the
+          # request. This response value may be different from the value set in the
+          # parameter.
           service_tier: nil,
           # This fingerprint represents the backend configuration that the model runs with.
           #
@@ -363,23 +363,23 @@ module OpenAI
           end
         end
 
-        # Specifies the latency tier to use for processing the request. This parameter is
-        # relevant for customers subscribed to the scale tier service:
+        # Specifies the processing type used for serving the request.
         #
-        # - If set to 'auto', and the Project is Scale tier enabled, the system will
-        #   utilize scale tier credits until they are exhausted.
-        # - If set to 'auto', and the Project is not Scale tier enabled, the request will
-        #   be processed using the default service tier with a lower uptime SLA and no
-        #   latency guarantee.
-        # - If set to 'default', the request will be processed using the default service
-        #   tier with a lower uptime SLA and no latency guarantee.
-        # - If set to 'flex', the request will be processed with the Flex Processing
-        #   service tier.
-        #   [Learn more](https://platform.openai.com/docs/guides/flex-processing).
+        # - If set to 'auto', then the request will be processed with the service tier
+        #   configured in the Project settings. Unless otherwise configured, the Project
+        #   will use 'default'.
+        # - If set to 'default', then the requset will be processed with the standard
+        #   pricing and performance for the selected model.
+        # - If set to '[flex](https://platform.openai.com/docs/guides/flex-processing)' or
+        #   'priority', then the request will be processed with the corresponding service
+        #   tier. [Contact sales](https://openai.com/contact-sales) to learn more about
+        #   Priority processing.
         # - When not set, the default behavior is 'auto'.
         #
-        # When this parameter is set, the response body will include the `service_tier`
-        # utilized.
+        # When the `service_tier` parameter is set, the response body will include the
+        # `service_tier` value based on the processing mode actually used to serve the
+        # request. This response value may be different from the value set in the
+        # parameter.
         module ServiceTier
           extend OpenAI::Internal::Type::Enum
 
@@ -409,6 +409,11 @@ module OpenAI
               :scale,
               OpenAI::Chat::ChatCompletion::ServiceTier::TaggedSymbol
             )
+          PRIORITY =
+            T.let(
+              :priority,
+              OpenAI::Chat::ChatCompletion::ServiceTier::TaggedSymbol
+            )
 
           sig do
             override.returns(

data/rbi/openai/models/chat/chat_completion_chunk.rbi

@@ -34,23 +34,23 @@ module OpenAI
         sig { returns(Symbol) }
         attr_accessor :object
 
-        # Specifies the latency tier to use for processing the request. This parameter is
-        # relevant for customers subscribed to the scale tier service:
+        # Specifies the processing type used for serving the request.
         #
-        # - If set to 'auto', and the Project is Scale tier enabled, the system will
-        #   utilize scale tier credits until they are exhausted.
-        # - If set to 'auto', and the Project is not Scale tier enabled, the request will
-        #   be processed using the default service tier with a lower uptime SLA and no
-        #   latency guarantee.
-        # - If set to 'default', the request will be processed using the default service
-        #   tier with a lower uptime SLA and no latency guarantee.
-        # - If set to 'flex', the request will be processed with the Flex Processing
-        #   service tier.
-        #   [Learn more](https://platform.openai.com/docs/guides/flex-processing).
+        # - If set to 'auto', then the request will be processed with the service tier
+        #   configured in the Project settings. Unless otherwise configured, the Project
+        #   will use 'default'.
+        # - If set to 'default', then the requset will be processed with the standard
+        #   pricing and performance for the selected model.
+        # - If set to '[flex](https://platform.openai.com/docs/guides/flex-processing)' or
+        #   'priority', then the request will be processed with the corresponding service
+        #   tier. [Contact sales](https://openai.com/contact-sales) to learn more about
+        #   Priority processing.
         # - When not set, the default behavior is 'auto'.
         #
-        # When this parameter is set, the response body will include the `service_tier`
-        # utilized.
+        # When the `service_tier` parameter is set, the response body will include the
+        # `service_tier` value based on the processing mode actually used to serve the
+        # request. This response value may be different from the value set in the
+        # parameter.
         sig do
           returns(
             T.nilable(
@@ -113,23 +113,23 @@ module OpenAI
           created:,
           # The model to generate the completion.
           model:,
-          # Specifies the latency tier to use for processing the request. This parameter is
-          # relevant for customers subscribed to the scale tier service:
+          # Specifies the processing type used for serving the request.
           #
-          # - If set to 'auto', and the Project is Scale tier enabled, the system will
-          #   utilize scale tier credits until they are exhausted.
-          # - If set to 'auto', and the Project is not Scale tier enabled, the request will
-          #   be processed using the default service tier with a lower uptime SLA and no
-          #   latency guarantee.
-          # - If set to 'default', the request will be processed using the default service
-          #   tier with a lower uptime SLA and no latency guarantee.
-          # - If set to 'flex', the request will be processed with the Flex Processing
-          #   service tier.
-          #   [Learn more](https://platform.openai.com/docs/guides/flex-processing).
+          # - If set to 'auto', then the request will be processed with the service tier
+          #   configured in the Project settings. Unless otherwise configured, the Project
+          #   will use 'default'.
+          # - If set to 'default', then the requset will be processed with the standard
+          #   pricing and performance for the selected model.
+          # - If set to '[flex](https://platform.openai.com/docs/guides/flex-processing)' or
+          #   'priority', then the request will be processed with the corresponding service
+          #   tier. [Contact sales](https://openai.com/contact-sales) to learn more about
+          #   Priority processing.
           # - When not set, the default behavior is 'auto'.
           #
-          # When this parameter is set, the response body will include the `service_tier`
-          # utilized.
+          # When the `service_tier` parameter is set, the response body will include the
+          # `service_tier` value based on the processing mode actually used to serve the
+          # request. This response value may be different from the value set in the
+          # parameter.
           service_tier: nil,
           # This fingerprint represents the backend configuration that the model runs with.
           # Can be used in conjunction with the `seed` request parameter to understand when
@@ -783,23 +783,23 @@ module OpenAI
           end
         end
 
-        # Specifies the latency tier to use for processing the request. This parameter is
-        # relevant for customers subscribed to the scale tier service:
+        # Specifies the processing type used for serving the request.
         #
-        # - If set to 'auto', and the Project is Scale tier enabled, the system will
-        #   utilize scale tier credits until they are exhausted.
-        # - If set to 'auto', and the Project is not Scale tier enabled, the request will
-        #   be processed using the default service tier with a lower uptime SLA and no
-        #   latency guarantee.
-        # - If set to 'default', the request will be processed using the default service
-        #   tier with a lower uptime SLA and no latency guarantee.
-        # - If set to 'flex', the request will be processed with the Flex Processing
-        #   service tier.
-        #   [Learn more](https://platform.openai.com/docs/guides/flex-processing).
+        # - If set to 'auto', then the request will be processed with the service tier
+        #   configured in the Project settings. Unless otherwise configured, the Project
+        #   will use 'default'.
+        # - If set to 'default', then the requset will be processed with the standard
+        #   pricing and performance for the selected model.
+        # - If set to '[flex](https://platform.openai.com/docs/guides/flex-processing)' or
+        #   'priority', then the request will be processed with the corresponding service
+        #   tier. [Contact sales](https://openai.com/contact-sales) to learn more about
+        #   Priority processing.
         # - When not set, the default behavior is 'auto'.
         #
-        # When this parameter is set, the response body will include the `service_tier`
-        # utilized.
+        # When the `service_tier` parameter is set, the response body will include the
+        # `service_tier` value based on the processing mode actually used to serve the
+        # request. This response value may be different from the value set in the
+        # parameter.
         module ServiceTier
           extend OpenAI::Internal::Type::Enum
 
@@ -829,6 +829,11 @@ module OpenAI
               :scale,
               OpenAI::Chat::ChatCompletionChunk::ServiceTier::TaggedSymbol
             )
+          PRIORITY =
+            T.let(
+              :priority,
+              OpenAI::Chat::ChatCompletionChunk::ServiceTier::TaggedSymbol
+            )
 
           sig do
             override.returns(