openai 0.14.0 → 0.16.0

data/lib/openai/resources/images.rb

@@ -39,7 +39,7 @@ module OpenAI
         )
       end
 
-      # See {OpenAI::Resources::Images#stream_raw} for streaming counterpart.
+      # See {OpenAI::Resources::Images#edit_stream_raw} for streaming counterpart.
       #
       # Some parameter documentations has been truncated, see
       # {OpenAI::Models::ImageEditParams} for more details.
@@ -85,7 +85,7 @@ module OpenAI
       def edit(params)
         parsed, options = OpenAI::ImageEditParams.dump_request(params)
         if parsed[:stream]
-          message = "Please use `#stream_raw` for the streaming use case."
+          message = "Please use `#edit_stream_raw` for the streaming use case."
           raise ArgumentError.new(message)
         end
         @client.request(
@@ -106,7 +106,7 @@ module OpenAI
       # Creates an edited or extended image given one or more source images and a
       # prompt. This endpoint only supports `gpt-image-1` and `dall-e-2`.
       #
-      # @overload stream_raw(image:, prompt:, background: nil, input_fidelity: nil, mask: nil, model: nil, n: nil, output_compression: nil, output_format: nil, partial_images: nil, quality: nil, response_format: nil, size: nil, user: nil, request_options: {})
+      # @overload edit_stream_raw(image:, prompt:, background: nil, input_fidelity: nil, mask: nil, model: nil, n: nil, output_compression: nil, output_format: nil, partial_images: nil, quality: nil, response_format: nil, size: nil, user: nil, request_options: {})
       #
       # @param image [Pathname, StringIO, IO, String, OpenAI::FilePart, Array<Pathname, StringIO, IO, String, OpenAI::FilePart>] The image(s) to edit. Must be a supported image file or an array of images.
       #
@@ -159,7 +159,7 @@ module OpenAI
         )
       end
 
-      # See {OpenAI::Resources::Images#stream_raw} for streaming counterpart.
+      # See {OpenAI::Resources::Images#generate_stream_raw} for streaming counterpart.
       #
       # Some parameter documentations has been truncated, see
       # {OpenAI::Models::ImageGenerateParams} for more details.
@@ -203,7 +203,7 @@ module OpenAI
       def generate(params)
         parsed, options = OpenAI::ImageGenerateParams.dump_request(params)
         if parsed[:stream]
-          message = "Please use `#stream_raw` for the streaming use case."
+          message = "Please use `#generate_stream_raw` for the streaming use case."
           raise ArgumentError.new(message)
         end
         @client.request(
@@ -223,7 +223,7 @@ module OpenAI
       # Creates an image given a prompt.
       # [Learn more](https://platform.openai.com/docs/guides/images).
       #
-      # @overload stream_raw(prompt:, background: nil, model: nil, moderation: nil, n: nil, output_compression: nil, output_format: nil, partial_images: nil, quality: nil, response_format: nil, size: nil, style: nil, user: nil, request_options: {})
+      # @overload generate_stream_raw(prompt:, background: nil, model: nil, moderation: nil, n: nil, output_compression: nil, output_format: nil, partial_images: nil, quality: nil, response_format: nil, size: nil, style: nil, user: nil, request_options: {})
       #
       # @param prompt [String] A text description of the desired image(s). The maximum length is 32000 characte
       #

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, 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: {})
+      # @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, prompt_cache_key: nil, reasoning: nil, safety_identifier: 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.
       #
@@ -47,8 +47,12 @@ module OpenAI
       #
       # @param prompt [OpenAI::Models::Responses::ResponsePrompt, nil] Reference to a prompt template and its variables.
       #
+      # @param prompt_cache_key [String] Used by OpenAI to cache responses for similar requests to optimize your cache hi
+      #
       # @param reasoning [OpenAI::Models::Reasoning, nil] **o-series models only**
       #
+      # @param safety_identifier [String] A stable identifier used to help detect users of your application that may be vi
+      #
       # @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
@@ -67,7 +71,7 @@ module OpenAI
       #
       # @param truncation [Symbol, OpenAI::Models::Responses::ResponseCreateParams::Truncation, nil] The truncation strategy to use for the model response.
       #
-      # @param user [String] A stable identifier for your end-users.
+      # @param user [String] This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
       #
       # @param request_options [OpenAI::RequestOptions, Hash{Symbol=>Object}, nil]
       #
@@ -170,20 +174,22 @@ module OpenAI
         end
         model, tool_models = get_structured_output_models(parsed)
 
+        unwrap = ->(raw) do
+          if raw[:type] == "response.completed" && raw[:response]
+            parse_structured_outputs!(raw[:response], model, tool_models)
+          end
+          raw
+        end
+
         if previous_response_id
-          retrieve_params = {}
-          retrieve_params[:include] = params[:include] if params[:include]
-          retrieve_params[:request_options] = params[:request_options] if params[:request_options]
+          retrieve_params = params.slice(:include, :request_options)
 
-          raw_stream = retrieve_streaming(previous_response_id, retrieve_params)
+          raw_stream = retrieve_streaming_internal(
+            previous_response_id,
+            params: retrieve_params,
+            unwrap: unwrap
+          )
         else
-          unwrap = ->(raw) do
-            if raw[:type] == "response.completed" && raw[:response]
-              parse_structured_outputs!(raw[:response], model, tool_models)
-            end
-            raw
-          end
-
           parsed[:stream] = true
 
           raw_stream = @client.request(
@@ -222,7 +228,7 @@ 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, 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: {})
+      # @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, prompt_cache_key: nil, reasoning: nil, safety_identifier: 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.
       #
@@ -246,8 +252,12 @@ module OpenAI
       #
       # @param prompt [OpenAI::Models::Responses::ResponsePrompt, nil] Reference to a prompt template and its variables.
       #
+      # @param prompt_cache_key [String] Used by OpenAI to cache responses for similar requests to optimize your cache hi
+      #
       # @param reasoning [OpenAI::Models::Reasoning, nil] **o-series models only**
       #
+      # @param safety_identifier [String] A stable identifier used to help detect users of your application that may be vi
+      #
       # @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
@@ -266,11 +276,11 @@ module OpenAI
       #
       # @param truncation [Symbol, OpenAI::Models::Responses::ResponseCreateParams::Truncation, nil] The truncation strategy to use for the model response.
       #
-      # @param user [String] A stable identifier for your end-users.
+      # @param user [String] This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
       #
       # @param request_options [OpenAI::RequestOptions, Hash{Symbol=>Object}, nil]
       #
-      # @return [OpenAI::Internal::Stream<OpenAI::Models::Responses::ResponseAudioDeltaEvent, OpenAI::Models::Responses::ResponseAudioDoneEvent, OpenAI::Models::Responses::ResponseAudioTranscriptDeltaEvent, OpenAI::Models::Responses::ResponseAudioTranscriptDoneEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCodeDeltaEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCodeDoneEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCompletedEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallInProgressEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallInterpretingEvent, OpenAI::Models::Responses::ResponseCompletedEvent, OpenAI::Models::Responses::ResponseContentPartAddedEvent, OpenAI::Models::Responses::ResponseContentPartDoneEvent, OpenAI::Models::Responses::ResponseCreatedEvent, OpenAI::Models::Responses::ResponseErrorEvent, OpenAI::Models::Responses::ResponseFileSearchCallCompletedEvent, OpenAI::Models::Responses::ResponseFileSearchCallInProgressEvent, OpenAI::Models::Responses::ResponseFileSearchCallSearchingEvent, OpenAI::Models::Responses::ResponseFunctionCallArgumentsDeltaEvent, OpenAI::Models::Responses::ResponseFunctionCallArgumentsDoneEvent, OpenAI::Models::Responses::ResponseInProgressEvent, OpenAI::Models::Responses::ResponseFailedEvent, OpenAI::Models::Responses::ResponseIncompleteEvent, OpenAI::Models::Responses::ResponseOutputItemAddedEvent, OpenAI::Models::Responses::ResponseOutputItemDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryPartAddedEvent, OpenAI::Models::Responses::ResponseReasoningSummaryPartDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryTextDeltaEvent, OpenAI::Models::Responses::ResponseReasoningSummaryTextDoneEvent, OpenAI::Models::Responses::ResponseRefusalDeltaEvent, OpenAI::Models::Responses::ResponseRefusalDoneEvent, OpenAI::Models::Responses::ResponseTextDeltaEvent, OpenAI::Models::Responses::ResponseTextDoneEvent, OpenAI::Models::Responses::ResponseWebSearchCallCompletedEvent, OpenAI::Models::Responses::ResponseWebSearchCallInProgressEvent, OpenAI::Models::Responses::ResponseWebSearchCallSearchingEvent, OpenAI::Models::Responses::ResponseImageGenCallCompletedEvent, OpenAI::Models::Responses::ResponseImageGenCallGeneratingEvent, OpenAI::Models::Responses::ResponseImageGenCallInProgressEvent, OpenAI::Models::Responses::ResponseImageGenCallPartialImageEvent, OpenAI::Models::Responses::ResponseMcpCallArgumentsDeltaEvent, OpenAI::Models::Responses::ResponseMcpCallArgumentsDoneEvent, OpenAI::Models::Responses::ResponseMcpCallCompletedEvent, OpenAI::Models::Responses::ResponseMcpCallFailedEvent, OpenAI::Models::Responses::ResponseMcpCallInProgressEvent, OpenAI::Models::Responses::ResponseMcpListToolsCompletedEvent, OpenAI::Models::Responses::ResponseMcpListToolsFailedEvent, OpenAI::Models::Responses::ResponseMcpListToolsInProgressEvent, OpenAI::Models::Responses::ResponseOutputTextAnnotationAddedEvent, OpenAI::Models::Responses::ResponseQueuedEvent, OpenAI::Models::Responses::ResponseReasoningDeltaEvent, OpenAI::Models::Responses::ResponseReasoningDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryDeltaEvent, OpenAI::Models::Responses::ResponseReasoningSummaryDoneEvent>]
+      # @return [OpenAI::Internal::Stream<OpenAI::Models::Responses::ResponseAudioDeltaEvent, OpenAI::Models::Responses::ResponseAudioDoneEvent, OpenAI::Models::Responses::ResponseAudioTranscriptDeltaEvent, OpenAI::Models::Responses::ResponseAudioTranscriptDoneEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCodeDeltaEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCodeDoneEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCompletedEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallInProgressEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallInterpretingEvent, OpenAI::Models::Responses::ResponseCompletedEvent, OpenAI::Models::Responses::ResponseContentPartAddedEvent, OpenAI::Models::Responses::ResponseContentPartDoneEvent, OpenAI::Models::Responses::ResponseCreatedEvent, OpenAI::Models::Responses::ResponseErrorEvent, OpenAI::Models::Responses::ResponseFileSearchCallCompletedEvent, OpenAI::Models::Responses::ResponseFileSearchCallInProgressEvent, OpenAI::Models::Responses::ResponseFileSearchCallSearchingEvent, OpenAI::Models::Responses::ResponseFunctionCallArgumentsDeltaEvent, OpenAI::Models::Responses::ResponseFunctionCallArgumentsDoneEvent, OpenAI::Models::Responses::ResponseInProgressEvent, OpenAI::Models::Responses::ResponseFailedEvent, OpenAI::Models::Responses::ResponseIncompleteEvent, OpenAI::Models::Responses::ResponseOutputItemAddedEvent, OpenAI::Models::Responses::ResponseOutputItemDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryPartAddedEvent, OpenAI::Models::Responses::ResponseReasoningSummaryPartDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryTextDeltaEvent, OpenAI::Models::Responses::ResponseReasoningSummaryTextDoneEvent, OpenAI::Models::Responses::ResponseRefusalDeltaEvent, OpenAI::Models::Responses::ResponseRefusalDoneEvent, OpenAI::Models::Responses::ResponseTextDeltaEvent, OpenAI::Models::Responses::ResponseTextDoneEvent, OpenAI::Models::Responses::ResponseWebSearchCallCompletedEvent, OpenAI::Models::Responses::ResponseWebSearchCallInProgressEvent, OpenAI::Models::Responses::ResponseWebSearchCallSearchingEvent, OpenAI::Models::Responses::ResponseImageGenCallCompletedEvent, OpenAI::Models::Responses::ResponseImageGenCallGeneratingEvent, OpenAI::Models::Responses::ResponseImageGenCallInProgressEvent, OpenAI::Models::Responses::ResponseImageGenCallPartialImageEvent, OpenAI::Models::Responses::ResponseMcpCallArgumentsDeltaEvent, OpenAI::Models::Responses::ResponseMcpCallArgumentsDoneEvent, OpenAI::Models::Responses::ResponseMcpCallCompletedEvent, OpenAI::Models::Responses::ResponseMcpCallFailedEvent, OpenAI::Models::Responses::ResponseMcpCallInProgressEvent, OpenAI::Models::Responses::ResponseMcpListToolsCompletedEvent, OpenAI::Models::Responses::ResponseMcpListToolsFailedEvent, OpenAI::Models::Responses::ResponseMcpListToolsInProgressEvent, OpenAI::Models::Responses::ResponseOutputTextAnnotationAddedEvent, OpenAI::Models::Responses::ResponseQueuedEvent, OpenAI::Models::Responses::ResponseReasoningSummaryDeltaEvent, OpenAI::Models::Responses::ResponseReasoningSummaryDoneEvent>]
       #
       # @see OpenAI::Models::Responses::ResponseCreateParams
       def stream_raw(params = {})
@@ -344,7 +354,7 @@ module OpenAI
       #
       # @param request_options [OpenAI::RequestOptions, Hash{Symbol=>Object}, nil]
       #
-      # @return [OpenAI::Internal::Stream<OpenAI::Models::Responses::ResponseAudioDeltaEvent, OpenAI::Models::Responses::ResponseAudioDoneEvent, OpenAI::Models::Responses::ResponseAudioTranscriptDeltaEvent, OpenAI::Models::Responses::ResponseAudioTranscriptDoneEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCodeDeltaEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCodeDoneEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCompletedEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallInProgressEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallInterpretingEvent, OpenAI::Models::Responses::ResponseCompletedEvent, OpenAI::Models::Responses::ResponseContentPartAddedEvent, OpenAI::Models::Responses::ResponseContentPartDoneEvent, OpenAI::Models::Responses::ResponseCreatedEvent, OpenAI::Models::Responses::ResponseErrorEvent, OpenAI::Models::Responses::ResponseFileSearchCallCompletedEvent, OpenAI::Models::Responses::ResponseFileSearchCallInProgressEvent, OpenAI::Models::Responses::ResponseFileSearchCallSearchingEvent, OpenAI::Models::Responses::ResponseFunctionCallArgumentsDeltaEvent, OpenAI::Models::Responses::ResponseFunctionCallArgumentsDoneEvent, OpenAI::Models::Responses::ResponseInProgressEvent, OpenAI::Models::Responses::ResponseFailedEvent, OpenAI::Models::Responses::ResponseIncompleteEvent, OpenAI::Models::Responses::ResponseOutputItemAddedEvent, OpenAI::Models::Responses::ResponseOutputItemDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryPartAddedEvent, OpenAI::Models::Responses::ResponseReasoningSummaryPartDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryTextDeltaEvent, OpenAI::Models::Responses::ResponseReasoningSummaryTextDoneEvent, OpenAI::Models::Responses::ResponseRefusalDeltaEvent, OpenAI::Models::Responses::ResponseRefusalDoneEvent, OpenAI::Models::Responses::ResponseTextDeltaEvent, OpenAI::Models::Responses::ResponseTextDoneEvent, OpenAI::Models::Responses::ResponseWebSearchCallCompletedEvent, OpenAI::Models::Responses::ResponseWebSearchCallInProgressEvent, OpenAI::Models::Responses::ResponseWebSearchCallSearchingEvent, OpenAI::Models::Responses::ResponseImageGenCallCompletedEvent, OpenAI::Models::Responses::ResponseImageGenCallGeneratingEvent, OpenAI::Models::Responses::ResponseImageGenCallInProgressEvent, OpenAI::Models::Responses::ResponseImageGenCallPartialImageEvent, OpenAI::Models::Responses::ResponseMcpCallArgumentsDeltaEvent, OpenAI::Models::Responses::ResponseMcpCallArgumentsDoneEvent, OpenAI::Models::Responses::ResponseMcpCallCompletedEvent, OpenAI::Models::Responses::ResponseMcpCallFailedEvent, OpenAI::Models::Responses::ResponseMcpCallInProgressEvent, OpenAI::Models::Responses::ResponseMcpListToolsCompletedEvent, OpenAI::Models::Responses::ResponseMcpListToolsFailedEvent, OpenAI::Models::Responses::ResponseMcpListToolsInProgressEvent, OpenAI::Models::Responses::ResponseOutputTextAnnotationAddedEvent, OpenAI::Models::Responses::ResponseQueuedEvent, OpenAI::Models::Responses::ResponseReasoningDeltaEvent, OpenAI::Models::Responses::ResponseReasoningDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryDeltaEvent, OpenAI::Models::Responses::ResponseReasoningSummaryDoneEvent>]
+      # @return [OpenAI::Internal::Stream<OpenAI::Models::Responses::ResponseAudioDeltaEvent, OpenAI::Models::Responses::ResponseAudioDoneEvent, OpenAI::Models::Responses::ResponseAudioTranscriptDeltaEvent, OpenAI::Models::Responses::ResponseAudioTranscriptDoneEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCodeDeltaEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCodeDoneEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallCompletedEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallInProgressEvent, OpenAI::Models::Responses::ResponseCodeInterpreterCallInterpretingEvent, OpenAI::Models::Responses::ResponseCompletedEvent, OpenAI::Models::Responses::ResponseContentPartAddedEvent, OpenAI::Models::Responses::ResponseContentPartDoneEvent, OpenAI::Models::Responses::ResponseCreatedEvent, OpenAI::Models::Responses::ResponseErrorEvent, OpenAI::Models::Responses::ResponseFileSearchCallCompletedEvent, OpenAI::Models::Responses::ResponseFileSearchCallInProgressEvent, OpenAI::Models::Responses::ResponseFileSearchCallSearchingEvent, OpenAI::Models::Responses::ResponseFunctionCallArgumentsDeltaEvent, OpenAI::Models::Responses::ResponseFunctionCallArgumentsDoneEvent, OpenAI::Models::Responses::ResponseInProgressEvent, OpenAI::Models::Responses::ResponseFailedEvent, OpenAI::Models::Responses::ResponseIncompleteEvent, OpenAI::Models::Responses::ResponseOutputItemAddedEvent, OpenAI::Models::Responses::ResponseOutputItemDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryPartAddedEvent, OpenAI::Models::Responses::ResponseReasoningSummaryPartDoneEvent, OpenAI::Models::Responses::ResponseReasoningSummaryTextDeltaEvent, OpenAI::Models::Responses::ResponseReasoningSummaryTextDoneEvent, OpenAI::Models::Responses::ResponseRefusalDeltaEvent, OpenAI::Models::Responses::ResponseRefusalDoneEvent, OpenAI::Models::Responses::ResponseTextDeltaEvent, OpenAI::Models::Responses::ResponseTextDoneEvent, OpenAI::Models::Responses::ResponseWebSearchCallCompletedEvent, OpenAI::Models::Responses::ResponseWebSearchCallInProgressEvent, OpenAI::Models::Responses::ResponseWebSearchCallSearchingEvent, OpenAI::Models::Responses::ResponseImageGenCallCompletedEvent, OpenAI::Models::Responses::ResponseImageGenCallGeneratingEvent, OpenAI::Models::Responses::ResponseImageGenCallInProgressEvent, OpenAI::Models::Responses::ResponseImageGenCallPartialImageEvent, OpenAI::Models::Responses::ResponseMcpCallArgumentsDeltaEvent, OpenAI::Models::Responses::ResponseMcpCallArgumentsDoneEvent, OpenAI::Models::Responses::ResponseMcpCallCompletedEvent, OpenAI::Models::Responses::ResponseMcpCallFailedEvent, OpenAI::Models::Responses::ResponseMcpCallInProgressEvent, OpenAI::Models::Responses::ResponseMcpListToolsCompletedEvent, OpenAI::Models::Responses::ResponseMcpListToolsFailedEvent, OpenAI::Models::Responses::ResponseMcpListToolsInProgressEvent, OpenAI::Models::Responses::ResponseOutputTextAnnotationAddedEvent, OpenAI::Models::Responses::ResponseQueuedEvent, OpenAI::Models::Responses::ResponseReasoningSummaryDeltaEvent, OpenAI::Models::Responses::ResponseReasoningSummaryDoneEvent>]
       #
       # @see OpenAI::Models::Responses::ResponseRetrieveParams
       def retrieve_streaming(response_id, params = {})
@@ -365,6 +375,21 @@ module OpenAI
         )
       end
 
+      private def retrieve_streaming_internal(response_id, params:, unwrap:)
+        parsed, options = OpenAI::Responses::ResponseRetrieveParams.dump_request(params)
+        parsed.store(:stream, true)
+        @client.request(
+          method: :get,
+          path: ["responses/%1$s", response_id],
+          query: parsed,
+          headers: {"accept" => "text/event-stream"},
+          stream: OpenAI::Internal::Stream,
+          model: OpenAI::Responses::ResponseStreamEvent,
+          options: options,
+          unwrap: unwrap
+        )
+      end
+
       # Deletes a model response with the given ID.
       #
       # @overload delete(response_id, request_options: {})

data/lib/openai/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module OpenAI
-  VERSION = "0.14.0"
+  VERSION = "0.16.0"
 end

data/lib/openai.rb

@@ -425,8 +425,6 @@ require_relative "openai/models/responses/response_output_text"
 require_relative "openai/models/responses/response_output_text_annotation_added_event"
 require_relative "openai/models/responses/response_prompt"
 require_relative "openai/models/responses/response_queued_event"
-require_relative "openai/models/responses/response_reasoning_delta_event"
-require_relative "openai/models/responses/response_reasoning_done_event"
 require_relative "openai/models/responses/response_reasoning_item"
 require_relative "openai/models/responses/response_reasoning_summary_delta_event"
 require_relative "openai/models/responses/response_reasoning_summary_done_event"

data/rbi/openai/models/audio/speech_create_params.rbi

@@ -216,15 +216,6 @@ module OpenAI
             )
           ECHO =
             T.let(:echo, OpenAI::Audio::SpeechCreateParams::Voice::TaggedSymbol)
-          FABLE =
-            T.let(
-              :fable,
-              OpenAI::Audio::SpeechCreateParams::Voice::TaggedSymbol
-            )
-          ONYX =
-            T.let(:onyx, OpenAI::Audio::SpeechCreateParams::Voice::TaggedSymbol)
-          NOVA =
-            T.let(:nova, OpenAI::Audio::SpeechCreateParams::Voice::TaggedSymbol)
           SAGE =
             T.let(:sage, OpenAI::Audio::SpeechCreateParams::Voice::TaggedSymbol)
           SHIMMER =

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

@@ -37,7 +37,7 @@ module OpenAI
         # - 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
+        # - If set to 'default', then the request 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
@@ -103,7 +103,7 @@ module OpenAI
           # - 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
+          # - If set to 'default', then the request 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
@@ -368,7 +368,7 @@ module OpenAI
         # - 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
+        # - If set to 'default', then the request 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

data/rbi/openai/models/chat/chat_completion_audio_param.rbi

@@ -176,21 +176,6 @@ module OpenAI
               :echo,
               OpenAI::Chat::ChatCompletionAudioParam::Voice::TaggedSymbol
             )
-          FABLE =
-            T.let(
-              :fable,
-              OpenAI::Chat::ChatCompletionAudioParam::Voice::TaggedSymbol
-            )
-          ONYX =
-            T.let(
-              :onyx,
-              OpenAI::Chat::ChatCompletionAudioParam::Voice::TaggedSymbol
-            )
-          NOVA =
-            T.let(
-              :nova,
-              OpenAI::Chat::ChatCompletionAudioParam::Voice::TaggedSymbol
-            )
           SAGE =
             T.let(
               :sage,

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

@@ -39,7 +39,7 @@ module OpenAI
         # - 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
+        # - If set to 'default', then the request 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
@@ -118,7 +118,7 @@ module OpenAI
           # - 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
+          # - If set to 'default', then the request 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
@@ -788,7 +788,7 @@ module OpenAI
         # - 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
+        # - If set to 'default', then the request 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

data/rbi/openai/models/chat/chat_completion_store_message.rbi

@@ -18,17 +18,82 @@ module OpenAI
         sig { returns(String) }
         attr_accessor :id
 
+        # If a content parts array was provided, this is an array of `text` and
+        # `image_url` parts. Otherwise, null.
+        sig do
+          returns(
+            T.nilable(
+              T::Array[
+                OpenAI::Chat::ChatCompletionStoreMessage::ContentPart::Variants
+              ]
+            )
+          )
+        end
+        attr_accessor :content_parts
+
         # A chat completion message generated by the model.
-        sig { params(id: String).returns(T.attached_class) }
+        sig do
+          params(
+            id: String,
+            content_parts:
+              T.nilable(
+                T::Array[
+                  T.any(
+                    OpenAI::Chat::ChatCompletionContentPartText::OrHash,
+                    OpenAI::Chat::ChatCompletionContentPartImage::OrHash
+                  )
+                ]
+              )
+          ).returns(T.attached_class)
+        end
         def self.new(
           # The identifier of the chat message.
-          id:
+          id:,
+          # If a content parts array was provided, this is an array of `text` and
+          # `image_url` parts. Otherwise, null.
+          content_parts: nil
         )
         end
 
-        sig { override.returns({ id: String }) }
+        sig do
+          override.returns(
+            {
+              id: String,
+              content_parts:
+                T.nilable(
+                  T::Array[
+                    OpenAI::Chat::ChatCompletionStoreMessage::ContentPart::Variants
+                  ]
+                )
+            }
+          )
+        end
         def to_hash
         end
+
+        # Learn about
+        # [text inputs](https://platform.openai.com/docs/guides/text-generation).
+        module ContentPart
+          extend OpenAI::Internal::Type::Union
+
+          Variants =
+            T.type_alias do
+              T.any(
+                OpenAI::Chat::ChatCompletionContentPartText,
+                OpenAI::Chat::ChatCompletionContentPartImage
+              )
+            end
+
+          sig do
+            override.returns(
+              T::Array[
+                OpenAI::Chat::ChatCompletionStoreMessage::ContentPart::Variants
+              ]
+            )
+          end
+          def self.variants
+          end
+        end
       end
     end
   end

data/rbi/openai/models/chat/completion_create_params.rbi

@@ -216,6 +216,15 @@ module OpenAI
         sig { returns(T.nilable(Float)) }
         attr_accessor :presence_penalty
 
+        # Used by OpenAI to cache responses for similar requests to optimize your cache
+        # hit rates. Replaces the `user` field.
+        # [Learn more](https://platform.openai.com/docs/guides/prompt-caching).
+        sig { returns(T.nilable(String)) }
+        attr_reader :prompt_cache_key
+
+        sig { params(prompt_cache_key: String).void }
+        attr_writer :prompt_cache_key
+
         # **o-series models only**
         #
         # Constrains effort on reasoning for
@@ -262,6 +271,17 @@ module OpenAI
         end
         attr_writer :response_format
 
+        # A stable identifier used to help detect users of your application that may be
+        # violating OpenAI's usage policies. The IDs should be a string that uniquely
+        # identifies each user. We recommend hashing their username or email address, in
+        # order to avoid sending us any identifying information.
+        # [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#safety-identifiers).
+        sig { returns(T.nilable(String)) }
+        attr_reader :safety_identifier
+
+        sig { params(safety_identifier: String).void }
+        attr_writer :safety_identifier
+
         # This feature is in Beta. If specified, our system will make a best effort to
         # sample deterministically, such that repeated requests with the same `seed` and
         # parameters should return the same result. Determinism is not guaranteed, and you
@@ -275,7 +295,7 @@ module OpenAI
         # - 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
+        # - If set to 'default', then the request 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
@@ -399,9 +419,11 @@ module OpenAI
         sig { returns(T.nilable(Float)) }
         attr_accessor :top_p
 
-        # A stable identifier for your end-users. Used to boost cache hit rates by better
-        # bucketing similar requests and to help OpenAI detect and prevent abuse.
-        # [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
+        # This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
+        # `prompt_cache_key` instead to maintain caching optimizations. A stable
+        # identifier for your end-users. Used to boost cache hit rates by better bucketing
+        # similar requests and to help OpenAI detect and prevent abuse.
+        # [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#safety-identifiers).
         sig { returns(T.nilable(String)) }
         attr_reader :user
 
@@ -465,6 +487,7 @@ module OpenAI
             prediction:
               T.nilable(OpenAI::Chat::ChatCompletionPredictionContent::OrHash),
             presence_penalty: T.nilable(Float),
+            prompt_cache_key: String,
             reasoning_effort: T.nilable(OpenAI::ReasoningEffort::OrSymbol),
             response_format:
               T.any(
@@ -473,6 +496,7 @@ module OpenAI
                 OpenAI::StructuredOutput::JsonSchemaConverter,
                 OpenAI::ResponseFormatJSONObject::OrHash
               ),
+            safety_identifier: String,
             seed: T.nilable(Integer),
             service_tier:
               T.nilable(
@@ -603,6 +627,10 @@ module OpenAI
           # whether they appear in the text so far, increasing the model's likelihood to
           # talk about new topics.
           presence_penalty: nil,
+          # Used by OpenAI to cache responses for similar requests to optimize your cache
+          # hit rates. Replaces the `user` field.
+          # [Learn more](https://platform.openai.com/docs/guides/prompt-caching).
+          prompt_cache_key: nil,
           # **o-series models only**
           #
           # Constrains effort on reasoning for
@@ -621,6 +649,12 @@ module OpenAI
           # ensures the message the model generates is valid JSON. Using `json_schema` is
           # preferred for models that support it.
           response_format: nil,
+          # A stable identifier used to help detect users of your application that may be
+          # violating OpenAI's usage policies. The IDs should be a string that uniquely
+          # identifies each user. We recommend hashing their username or email address, in
+          # order to avoid sending us any identifying information.
+          # [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#safety-identifiers).
+          safety_identifier: nil,
           # This feature is in Beta. If specified, our system will make a best effort to
           # sample deterministically, such that repeated requests with the same `seed` and
           # parameters should return the same result. Determinism is not guaranteed, and you
@@ -632,7 +666,7 @@ module OpenAI
           # - 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
+          # - If set to 'default', then the request 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
@@ -687,9 +721,11 @@ module OpenAI
           #
           # We generally recommend altering this or `temperature` but not both.
           top_p: nil,
-          # A stable identifier for your end-users. Used to boost cache hit rates by better
-          # bucketing similar requests and to help OpenAI detect and prevent abuse.
-          # [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
+          # This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
+          # `prompt_cache_key` instead to maintain caching optimizations. A stable
+          # identifier for your end-users. Used to boost cache hit rates by better bucketing
+          # similar requests and to help OpenAI detect and prevent abuse.
+          # [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#safety-identifiers).
           user: nil,
           # This tool searches the web for relevant results to use in a response. Learn more
           # about the
@@ -739,6 +775,7 @@ module OpenAI
               prediction:
                 T.nilable(OpenAI::Chat::ChatCompletionPredictionContent),
               presence_penalty: T.nilable(Float),
+              prompt_cache_key: String,
               reasoning_effort: T.nilable(OpenAI::ReasoningEffort::OrSymbol),
               response_format:
                 T.any(
@@ -746,6 +783,7 @@ module OpenAI
                   OpenAI::ResponseFormatJSONSchema,
                   OpenAI::ResponseFormatJSONObject
                 ),
+              safety_identifier: String,
               seed: T.nilable(Integer),
               service_tier:
                 T.nilable(
@@ -1017,7 +1055,7 @@ module OpenAI
         # - 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
+        # - If set to 'default', then the request 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

data/rbi/openai/models/function_definition.rbi

@@ -38,7 +38,7 @@ module OpenAI
       # set to true, the model will follow the exact schema defined in the `parameters`
       # field. Only a subset of JSON Schema is supported when `strict` is `true`. Learn
       # more about Structured Outputs in the
-      # [function calling guide](docs/guides/function-calling).
+      # [function calling guide](https://platform.openai.com/docs/guides/function-calling).
       sig { returns(T.nilable(T::Boolean)) }
       attr_accessor :strict
 
@@ -69,7 +69,7 @@ module OpenAI
         # set to true, the model will follow the exact schema defined in the `parameters`
         # field. Only a subset of JSON Schema is supported when `strict` is `true`. Learn
         # more about Structured Outputs in the
-        # [function calling guide](docs/guides/function-calling).
+        # [function calling guide](https://platform.openai.com/docs/guides/function-calling).
         strict: nil
       )
       end

data/rbi/openai/models/image_edit_params.rbi

@@ -81,6 +81,9 @@ module OpenAI
       # The number of partial images to generate. This parameter is used for streaming
       # responses that return partial images. Value must be between 0 and 3. When set to
       # 0, the response will be a single image sent in one streaming event.
+      #
+      # Note that the final image may be sent before the full number of partial images
+      # are generated if the full image is generated more quickly.
       sig { returns(T.nilable(Integer)) }
       attr_accessor :partial_images
 
@@ -182,6 +185,9 @@ module OpenAI
         # The number of partial images to generate. This parameter is used for streaming
         # responses that return partial images. Value must be between 0 and 3. When set to
         # 0, the response will be a single image sent in one streaming event.
+        #
+        # Note that the final image may be sent before the full number of partial images
+        # are generated if the full image is generated more quickly.
         partial_images: nil,
         # The quality of the image that will be generated. `high`, `medium` and `low` are
         # only supported for `gpt-image-1`. `dall-e-2` only supports `standard` quality.

data/rbi/openai/models/image_generate_params.rbi

@@ -63,6 +63,9 @@ module OpenAI
       # The number of partial images to generate. This parameter is used for streaming
       # responses that return partial images. Value must be between 0 and 3. When set to
       # 0, the response will be a single image sent in one streaming event.
+      #
+      # Note that the final image may be sent before the full number of partial images
+      # are generated if the full image is generated more quickly.
       sig { returns(T.nilable(Integer)) }
       attr_accessor :partial_images
 
@@ -165,6 +168,9 @@ module OpenAI
         # The number of partial images to generate. This parameter is used for streaming
         # responses that return partial images. Value must be between 0 and 3. When set to
         # 0, the response will be a single image sent in one streaming event.
+        #
+        # Note that the final image may be sent before the full number of partial images
+        # are generated if the full image is generated more quickly.
         partial_images: nil,
         # The quality of the image that will be generated.
         #

data/rbi/openai/models/images_response.rbi

@@ -224,7 +224,7 @@ module OpenAI
         end
         attr_writer :input_tokens_details
 
-        # The number of image tokens in the output image.
+        # The number of output tokens generated by the model.
         sig { returns(Integer) }
         attr_accessor :output_tokens
 
@@ -247,7 +247,7 @@ module OpenAI
           input_tokens:,
           # The input tokens detailed information for the image generation.
           input_tokens_details:,
-          # The number of image tokens in the output image.
+          # The number of output tokens generated by the model.
           output_tokens:,
           # The total number of tokens (images and text) used for the image generation.
           total_tokens: