feat(api): add o3 and o4-mini model IDs

aliases.go

@@ -37,6 +37,18 @@ const ChatModelGPT4_1Mini2025_04_14 = shared.ChatModelGPT4_1Mini2025_04_14
 // Equals "gpt-4.1-nano-2025-04-14"
 const ChatModelGPT4_1Nano2025_04_14 = shared.ChatModelGPT4_1Nano2025_04_14
 
+// Equals "o4-mini"
+const ChatModelO4Mini = shared.ChatModelO4Mini
+
+// Equals "o4-mini-2025-04-16"
+const ChatModelO4Mini2025_04_16 = shared.ChatModelO4Mini2025_04_16
+
+// Equals "o3"
+const ChatModelO3 = shared.ChatModelO3
+
+// Equals "o3-2025-04-16"
+const ChatModelO3_2025_04_16 = shared.ChatModelO3_2025_04_16
+
 // Equals "o3-mini"
 const ChatModelO3Mini = shared.ChatModelO3Mini
 
@@ -279,21 +291,40 @@ type Metadata = shared.Metadata
 // This is an alias to an internal type.
 type Reasoning = shared.Reasoning
 
-// **computer_use_preview only**
+// **Deprecated:** use `summary` instead.
 //
 // A summary of the reasoning performed by the model. This can be useful for
-// debugging and understanding the model's reasoning process. One of `concise` or
-// `detailed`.
+// debugging and understanding the model's reasoning process. One of `auto`,
+// `concise`, or `detailed`.
 //
 // This is an alias to an internal type.
 type ReasoningGenerateSummary = shared.ReasoningGenerateSummary
 
+// Equals "auto"
+const ReasoningGenerateSummaryAuto = shared.ReasoningGenerateSummaryAuto
+
 // Equals "concise"
 const ReasoningGenerateSummaryConcise = shared.ReasoningGenerateSummaryConcise
 
 // Equals "detailed"
 const ReasoningGenerateSummaryDetailed = shared.ReasoningGenerateSummaryDetailed
 
+// A summary of the reasoning performed by the model. This can be useful for
+// debugging and understanding the model's reasoning process. One of `auto`,
+// `concise`, or `detailed`.
+//
+// This is an alias to an internal type.
+type ReasoningSummary = shared.ReasoningSummary
+
+// Equals "auto"
+const ReasoningSummaryAuto = shared.ReasoningSummaryAuto
+
+// Equals "concise"
+const ReasoningSummaryConcise = shared.ReasoningSummaryConcise
+
+// Equals "detailed"
+const ReasoningSummaryDetailed = shared.ReasoningSummaryDetailed
+
 // **o-series models only**
 //
 // Configuration options for

chatcompletion.go

@@ -174,9 +174,25 @@ type ChatCompletion struct {
 	Model string `json:"model,required"`
 	// The object type, which is always `chat.completion`.
 	Object constant.ChatCompletion `json:"object,required"`
-	// The service tier used for processing the request.
+	// Specifies the latency tier to use for processing the request. This parameter is
+	// relevant for customers subscribed to the scale tier service:
 	//
-	// Any of "scale", "default".
+	//   - 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 guarentee.
+	//   - If set to 'default', the request will be processed using the default service
+	//     tier with a lower uptime SLA and no latency guarentee.
+	//   - 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).
+	//   - When not set, the default behavior is 'auto'.
+	//
+	// When this parameter is set, the response body will include the `service_tier`
+	// utilized.
+	//
+	// Any of "auto", "default", "flex".
 	ServiceTier ChatCompletionServiceTier `json:"service_tier,nullable"`
 	// This fingerprint represents the backend configuration that the model runs with.
 	//
@@ -260,12 +276,29 @@ func (r *ChatCompletionChoiceLogprobs) UnmarshalJSON(data []byte) error {
 	return apijson.UnmarshalRoot(data, r)
 }
 
-// The service tier used for processing the request.
+// Specifies the latency tier to use for processing the request. This parameter is
+// relevant for customers subscribed to the scale tier service:
+//
+//   - 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 guarentee.
+//   - If set to 'default', the request will be processed using the default service
+//     tier with a lower uptime SLA and no latency guarentee.
+//   - 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).
+//   - When not set, the default behavior is 'auto'.
+//
+// When this parameter is set, the response body will include the `service_tier`
+// utilized.
 type ChatCompletionServiceTier string
 
 const (
-	ChatCompletionServiceTierScale   ChatCompletionServiceTier = "scale"
+	ChatCompletionServiceTierAuto    ChatCompletionServiceTier = "auto"
 	ChatCompletionServiceTierDefault ChatCompletionServiceTier = "default"
+	ChatCompletionServiceTierFlex    ChatCompletionServiceTier = "flex"
 )
 
 // Messages sent by the model in response to user messages.
@@ -472,10 +505,10 @@ type ChatCompletionAudioParam struct {
 	// Specifies the output audio format. Must be one of `wav`, `mp3`, `flac`, `opus`,
 	// or `pcm16`.
 	//
-	// Any of "wav", "mp3", "flac", "opus", "pcm16".
+	// Any of "wav", "aac", "mp3", "flac", "opus", "pcm16".
 	Format ChatCompletionAudioParamFormat `json:"format,omitzero,required"`
 	// The voice the model uses to respond. Supported voices are `alloy`, `ash`,
-	// `ballad`, `coral`, `echo`, `sage`, and `shimmer`.
+	// `ballad`, `coral`, `echo`, `fable`, `nova`, `onyx`, `sage`, and `shimmer`.
 	Voice ChatCompletionAudioParamVoice `json:"voice,omitzero,required"`
 	paramObj
 }
@@ -494,6 +527,7 @@ type ChatCompletionAudioParamFormat string
 
 const (
 	ChatCompletionAudioParamFormatWAV   ChatCompletionAudioParamFormat = "wav"
+	ChatCompletionAudioParamFormatAAC   ChatCompletionAudioParamFormat = "aac"
 	ChatCompletionAudioParamFormatMP3   ChatCompletionAudioParamFormat = "mp3"
 	ChatCompletionAudioParamFormatFLAC  ChatCompletionAudioParamFormat = "flac"
 	ChatCompletionAudioParamFormatOpus  ChatCompletionAudioParamFormat = "opus"
@@ -501,7 +535,7 @@ const (
 )
 
 // The voice the model uses to respond. Supported voices are `alloy`, `ash`,
-// `ballad`, `coral`, `echo`, `sage`, and `shimmer`.
+// `ballad`, `coral`, `echo`, `fable`, `nova`, `onyx`, `sage`, and `shimmer`.
 type ChatCompletionAudioParamVoice string
 
 const (
@@ -535,9 +569,25 @@ type ChatCompletionChunk struct {
 	Model string `json:"model,required"`
 	// The object type, which is always `chat.completion.chunk`.
 	Object constant.ChatCompletionChunk `json:"object,required"`
-	// The service tier used for processing the request.
+	// Specifies the latency tier to use for processing the request. This parameter is
+	// relevant for customers subscribed to the scale tier service:
 	//
-	// Any of "scale", "default".
+	//   - 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 guarentee.
+	//   - If set to 'default', the request will be processed using the default service
+	//     tier with a lower uptime SLA and no latency guarentee.
+	//   - 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).
+	//   - When not set, the default behavior is 'auto'.
+	//
+	// When this parameter is set, the response body will include the `service_tier`
+	// utilized.
+	//
+	// Any of "auto", "default", "flex".
 	ServiceTier ChatCompletionChunkServiceTier `json:"service_tier,nullable"`
 	// This fingerprint represents the backend configuration that the model runs with.
 	// Can be used in conjunction with the `seed` request parameter to understand when
@@ -736,12 +786,29 @@ func (r *ChatCompletionChunkChoiceLogprobs) UnmarshalJSON(data []byte) error {
 	return apijson.UnmarshalRoot(data, r)
 }
 
-// The service tier used for processing the request.
+// Specifies the latency tier to use for processing the request. This parameter is
+// relevant for customers subscribed to the scale tier service:
+//
+//   - 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 guarentee.
+//   - If set to 'default', the request will be processed using the default service
+//     tier with a lower uptime SLA and no latency guarentee.
+//   - 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).
+//   - When not set, the default behavior is 'auto'.
+//
+// When this parameter is set, the response body will include the `service_tier`
+// utilized.
 type ChatCompletionChunkServiceTier string
 
 const (
-	ChatCompletionChunkServiceTierScale   ChatCompletionChunkServiceTier = "scale"
+	ChatCompletionChunkServiceTierAuto    ChatCompletionChunkServiceTier = "auto"
 	ChatCompletionChunkServiceTierDefault ChatCompletionChunkServiceTier = "default"
+	ChatCompletionChunkServiceTierFlex    ChatCompletionChunkServiceTier = "flex"
 )
 
 func TextContentPart(text string) ChatCompletionContentPartUnionParam {
@@ -2026,7 +2093,7 @@ type ChatCompletionNewParams struct {
 	// [images](https://platform.openai.com/docs/guides/vision), and
 	// [audio](https://platform.openai.com/docs/guides/audio).
 	Messages []ChatCompletionMessageParamUnion `json:"messages,omitzero,required"`
-	// Model ID used to generate the response, like `gpt-4o` or `o1`. OpenAI offers a
+	// Model ID used to generate the response, like `gpt-4o` or `o3`. OpenAI offers a
 	// wide range of models with different capabilities, performance characteristics,
 	// and price points. Refer to the
 	// [model guide](https://platform.openai.com/docs/models) to browse and compare
@@ -2050,7 +2117,7 @@ type ChatCompletionNewParams struct {
 	//
 	// This value is now deprecated in favor of `max_completion_tokens`, and is not
 	// compatible with
-	// [o1 series models](https://platform.openai.com/docs/guides/reasoning).
+	// [o-series models](https://platform.openai.com/docs/guides/reasoning).
 	MaxTokens param.Opt[int64] `json:"max_tokens,omitzero"`
 	// How many chat completion choices to generate for each input message. Note that
 	// you will be charged based on the number of generated tokens across all of the
@@ -2145,13 +2212,18 @@ type ChatCompletionNewParams struct {
 	//     latency guarentee.
 	//   - If set to 'default', the request will be processed using the default service
 	//     tier with a lower uptime SLA and no latency guarentee.
+	//   - 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).
 	//   - When not set, the default behavior is 'auto'.
 	//
 	// When this parameter is set, the response body will include the `service_tier`
 	// utilized.
 	//
-	// Any of "auto", "default".
+	// Any of "auto", "default", "flex".
 	ServiceTier ChatCompletionNewParamsServiceTier `json:"service_tier,omitzero"`
+	// Not supported with latest reasoning models `o3` and `o4-mini`.
+	//
 	// Up to 4 sequences where the API will stop generating further tokens. The
 	// returned text will not contain the stop sequence.
 	Stop ChatCompletionNewParamsStopUnion `json:"stop,omitzero"`
@@ -2342,6 +2414,9 @@ func (u ChatCompletionNewParamsResponseFormatUnion) GetType() *string {
 //     latency guarentee.
 //   - If set to 'default', the request will be processed using the default service
 //     tier with a lower uptime SLA and no latency guarentee.
+//   - 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).
 //   - When not set, the default behavior is 'auto'.
 //
 // When this parameter is set, the response body will include the `service_tier`
@@ -2351,6 +2426,7 @@ type ChatCompletionNewParamsServiceTier string
 const (
 	ChatCompletionNewParamsServiceTierAuto    ChatCompletionNewParamsServiceTier = "auto"
 	ChatCompletionNewParamsServiceTierDefault ChatCompletionNewParamsServiceTier = "default"
+	ChatCompletionNewParamsServiceTierFlex    ChatCompletionNewParamsServiceTier = "flex"
 )
 
 // Only one field can be non-zero.

completion.go

@@ -338,6 +338,8 @@ type CompletionNewParams struct {
 	// As an example, you can pass `{"50256": -100}` to prevent the <|endoftext|> token
 	// from being generated.
 	LogitBias map[string]int64 `json:"logit_bias,omitzero"`
+	// Not supported with latest reasoning models `o3` and `o4-mini`.
+	//
 	// Up to 4 sequences where the API will stop generating further tokens. The
 	// returned text will not contain the stop sequence.
 	Stop CompletionNewParamsStopUnion `json:"stop,omitzero"`

responses/aliases.go

@@ -37,6 +37,18 @@ const ChatModelGPT4_1Mini2025_04_14 = shared.ChatModelGPT4_1Mini2025_04_14
 // Equals "gpt-4.1-nano-2025-04-14"
 const ChatModelGPT4_1Nano2025_04_14 = shared.ChatModelGPT4_1Nano2025_04_14
 
+// Equals "o4-mini"
+const ChatModelO4Mini = shared.ChatModelO4Mini
+
+// Equals "o4-mini-2025-04-16"
+const ChatModelO4Mini2025_04_16 = shared.ChatModelO4Mini2025_04_16
+
+// Equals "o3"
+const ChatModelO3 = shared.ChatModelO3
+
+// Equals "o3-2025-04-16"
+const ChatModelO3_2025_04_16 = shared.ChatModelO3_2025_04_16
+
 // Equals "o3-mini"
 const ChatModelO3Mini = shared.ChatModelO3Mini
 
@@ -279,21 +291,40 @@ type Metadata = shared.Metadata
 // This is an alias to an internal type.
 type Reasoning = shared.Reasoning
 
-// **computer_use_preview only**
+// **Deprecated:** use `summary` instead.
 //
 // A summary of the reasoning performed by the model. This can be useful for
-// debugging and understanding the model's reasoning process. One of `concise` or
-// `detailed`.
+// debugging and understanding the model's reasoning process. One of `auto`,
+// `concise`, or `detailed`.
 //
 // This is an alias to an internal type.
 type ReasoningGenerateSummary = shared.ReasoningGenerateSummary
 
+// Equals "auto"
+const ReasoningGenerateSummaryAuto = shared.ReasoningGenerateSummaryAuto
+
 // Equals "concise"
 const ReasoningGenerateSummaryConcise = shared.ReasoningGenerateSummaryConcise
 
 // Equals "detailed"
 const ReasoningGenerateSummaryDetailed = shared.ReasoningGenerateSummaryDetailed
 
+// A summary of the reasoning performed by the model. This can be useful for
+// debugging and understanding the model's reasoning process. One of `auto`,
+// `concise`, or `detailed`.
+//
+// This is an alias to an internal type.
+type ReasoningSummary = shared.ReasoningSummary
+
+// Equals "auto"
+const ReasoningSummaryAuto = shared.ReasoningSummaryAuto
+
+// Equals "concise"
+const ReasoningSummaryConcise = shared.ReasoningSummaryConcise
+
+// Equals "detailed"
+const ReasoningSummaryDetailed = shared.ReasoningSummaryDetailed
+
 // **o-series models only**
 //
 // Configuration options for

responses/response.go

@@ -578,7 +578,7 @@ type Response struct {
 	// Keys are strings with a maximum length of 64 characters. Values are strings with
 	// a maximum length of 512 characters.
 	Metadata shared.Metadata `json:"metadata,required"`
-	// Model ID used to generate the response, like `gpt-4o` or `o1`. OpenAI offers a
+	// Model ID used to generate the response, like `gpt-4o` or `o3`. OpenAI offers a
 	// wide range of models with different capabilities, performance characteristics,
 	// and price points. Refer to the
 	// [model guide](https://platform.openai.com/docs/models) to browse and compare
@@ -639,6 +639,26 @@ type Response struct {
 	// Configuration options for
 	// [reasoning models](https://platform.openai.com/docs/guides/reasoning).
 	Reasoning shared.Reasoning `json:"reasoning,nullable"`
+	// Specifies the latency tier to use for processing the request. This parameter is
+	// relevant for customers subscribed to the scale tier service:
+	//
+	//   - 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 guarentee.
+	//   - If set to 'default', the request will be processed using the default service
+	//     tier with a lower uptime SLA and no latency guarentee.
+	//   - 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).
+	//   - When not set, the default behavior is 'auto'.
+	//
+	// When this parameter is set, the response body will include the `service_tier`
+	// utilized.
+	//
+	// Any of "auto", "default", "flex".
+	ServiceTier ResponseServiceTier `json:"service_tier,nullable"`
 	// The status of the response generation. One of `completed`, `failed`,
 	// `in_progress`, or `incomplete`.
 	//
@@ -686,6 +706,7 @@ type Response struct {
 		MaxOutputTokens    respjson.Field
 		PreviousResponseID respjson.Field
 		Reasoning          respjson.Field
+		ServiceTier        respjson.Field
 		Status             respjson.Field
 		Text               respjson.Field
 		Truncation         respjson.Field
@@ -778,6 +799,31 @@ func (r *ResponseToolChoiceUnion) UnmarshalJSON(data []byte) error {
 	return apijson.UnmarshalRoot(data, r)
 }
 
+// Specifies the latency tier to use for processing the request. This parameter is
+// relevant for customers subscribed to the scale tier service:
+//
+//   - 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 guarentee.
+//   - If set to 'default', the request will be processed using the default service
+//     tier with a lower uptime SLA and no latency guarentee.
+//   - 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).
+//   - When not set, the default behavior is 'auto'.
+//
+// When this parameter is set, the response body will include the `service_tier`
+// utilized.
+type ResponseServiceTier string
+
+const (
+	ResponseServiceTierAuto    ResponseServiceTier = "auto"
+	ResponseServiceTierDefault ResponseServiceTier = "default"
+	ResponseServiceTierFlex    ResponseServiceTier = "flex"
+)
+
 // The truncation strategy to use for the model response.
 //
 //   - `auto`: If the context of this response and previous ones exceeds the model's
@@ -7509,7 +7555,7 @@ type ResponseNewParams struct {
 	// - [Conversation state](https://platform.openai.com/docs/guides/conversation-state)
 	// - [Function calling](https://platform.openai.com/docs/guides/function-calling)
 	Input ResponseNewParamsInputUnion `json:"input,omitzero,required"`
-	// Model ID used to generate the response, like `gpt-4o` or `o1`. OpenAI offers a
+	// Model ID used to generate the response, like `gpt-4o` or `o3`. OpenAI offers a
 	// wide range of models with different capabilities, performance characteristics,
 	// and price points. Refer to the
 	// [model guide](https://platform.openai.com/docs/models) to browse and compare
@@ -7570,6 +7616,26 @@ type ResponseNewParams struct {
 	// Keys are strings with a maximum length of 64 characters. Values are strings with
 	// a maximum length of 512 characters.
 	Metadata shared.Metadata `json:"metadata,omitzero"`
+	// Specifies the latency tier to use for processing the request. This parameter is
+	// relevant for customers subscribed to the scale tier service:
+	//
+	//   - 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 guarentee.
+	//   - If set to 'default', the request will be processed using the default service
+	//     tier with a lower uptime SLA and no latency guarentee.
+	//   - 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).
+	//   - When not set, the default behavior is 'auto'.
+	//
+	// When this parameter is set, the response body will include the `service_tier`
+	// utilized.
+	//
+	// Any of "auto", "default", "flex".
+	ServiceTier ResponseNewParamsServiceTier `json:"service_tier,omitzero"`
 	// The truncation strategy to use for the model response.
 	//
 	//   - `auto`: If the context of this response and previous ones exceeds the model's
@@ -7646,6 +7712,31 @@ func (u *ResponseNewParamsInputUnion) asAny() any {
 	return nil
 }
 
+// Specifies the latency tier to use for processing the request. This parameter is
+// relevant for customers subscribed to the scale tier service:
+//
+//   - 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 guarentee.
+//   - If set to 'default', the request will be processed using the default service
+//     tier with a lower uptime SLA and no latency guarentee.
+//   - 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).
+//   - When not set, the default behavior is 'auto'.
+//
+// When this parameter is set, the response body will include the `service_tier`
+// utilized.
+type ResponseNewParamsServiceTier string
+
+const (
+	ResponseNewParamsServiceTierAuto    ResponseNewParamsServiceTier = "auto"
+	ResponseNewParamsServiceTierDefault ResponseNewParamsServiceTier = "default"
+	ResponseNewParamsServiceTierFlex    ResponseNewParamsServiceTier = "flex"
+)
+
 // Only one field can be non-zero.
 //
 // Use [param.IsOmitted] to confirm if a field is set.

responses/response_test.go

@@ -42,8 +42,10 @@ func TestResponseNewWithOptionalParams(t *testing.T) {
 		PreviousResponseID: openai.String("previous_response_id"),
 		Reasoning: shared.ReasoningParam{
 			Effort:          shared.ReasoningEffortLow,
-			GenerateSummary: shared.ReasoningGenerateSummaryConcise,
+			GenerateSummary: shared.ReasoningGenerateSummaryAuto,
+			Summary:         shared.ReasoningSummaryAuto,
 		},
+		ServiceTier: responses.ResponseNewParamsServiceTierAuto,
 		Store:       openai.Bool(true),
 		Temperature: openai.Float(1),
 		Text: responses.ResponseTextConfigParam{

shared/shared.go

@@ -27,6 +27,10 @@ const (
 	ChatModelGPT4_1_2025_04_14                ChatModel = "gpt-4.1-2025-04-14"
 	ChatModelGPT4_1Mini2025_04_14             ChatModel = "gpt-4.1-mini-2025-04-14"
 	ChatModelGPT4_1Nano2025_04_14             ChatModel = "gpt-4.1-nano-2025-04-14"
+	ChatModelO4Mini                           ChatModel = "o4-mini"
+	ChatModelO4Mini2025_04_16                 ChatModel = "o4-mini-2025-04-16"
+	ChatModelO3                               ChatModel = "o3"
+	ChatModelO3_2025_04_16                    ChatModel = "o3-2025-04-16"
 	ChatModelO3Mini                           ChatModel = "o3-mini"
 	ChatModelO3Mini2025_01_31                 ChatModel = "o3-mini-2025-01-31"
 	ChatModelO1                               ChatModel = "o1"
@@ -422,13 +426,15 @@ type Reasoning struct {
 	//
 	// Any of "low", "medium", "high".
 	Effort ReasoningEffort `json:"effort,nullable"`
-	// **computer_use_preview only**
+	// **Deprecated:** use `summary` instead.
 	//
 	// A summary of the reasoning performed by the model. This can be useful for
-	// debugging and understanding the model's reasoning process. One of `concise` or
-	// `detailed`.
+	// debugging and understanding the model's reasoning process. One of `auto`,
+	// `concise`, or `detailed`.
 	//
-	// Any of "concise", "detailed".
+	// Any of "auto", "concise", "detailed".
+	//
+	// Deprecated: deprecated
 	GenerateSummary ReasoningGenerateSummary `json:"generate_summary,nullable"`
 	// A summary of the reasoning performed by the model. This can be useful for
 	// debugging and understanding the model's reasoning process. One of `auto`,
@@ -440,6 +446,7 @@ type Reasoning struct {
 	JSON struct {
 		Effort          respjson.Field
 		GenerateSummary respjson.Field
+		Summary         respjson.Field
 		ExtraFields     map[string]respjson.Field
 		raw             string
 	} `json:"-"`
@@ -460,18 +467,30 @@ func (r Reasoning) ToParam() ReasoningParam {
 	return param.Override[ReasoningParam](r.RawJSON())
 }
 
-// **computer_use_preview only**
+// **Deprecated:** use `summary` instead.
 //
 // A summary of the reasoning performed by the model. This can be useful for
-// debugging and understanding the model's reasoning process. One of `concise` or
-// `detailed`.
+// debugging and understanding the model's reasoning process. One of `auto`,
+// `concise`, or `detailed`.
 type ReasoningGenerateSummary string
 
 const (
+	ReasoningGenerateSummaryAuto     ReasoningGenerateSummary = "auto"
 	ReasoningGenerateSummaryConcise  ReasoningGenerateSummary = "concise"
 	ReasoningGenerateSummaryDetailed ReasoningGenerateSummary = "detailed"
 )
 
+// A summary of the reasoning performed by the model. This can be useful for
+// debugging and understanding the model's reasoning process. One of `auto`,
+// `concise`, or `detailed`.
+type ReasoningSummary string
+
+const (
+	ReasoningSummaryAuto     ReasoningSummary = "auto"
+	ReasoningSummaryConcise  ReasoningSummary = "concise"
+	ReasoningSummaryDetailed ReasoningSummary = "detailed"
+)
+
 // **o-series models only**
 //
 // Configuration options for
@@ -486,14 +505,22 @@ type ReasoningParam struct {
 	//
 	// Any of "low", "medium", "high".
 	Effort ReasoningEffort `json:"effort,omitzero"`
-	// **computer_use_preview only**
+	// **Deprecated:** use `summary` instead.
 	//
 	// A summary of the reasoning performed by the model. This can be useful for
-	// debugging and understanding the model's reasoning process. One of `concise` or
-	// `detailed`.
+	// debugging and understanding the model's reasoning process. One of `auto`,
+	// `concise`, or `detailed`.
 	//
-	// Any of "concise", "detailed".
+	// Any of "auto", "concise", "detailed".
+	//
+	// Deprecated: deprecated
 	GenerateSummary ReasoningGenerateSummary `json:"generate_summary,omitzero"`
+	// A summary of the reasoning performed by the model. This can be useful for
+	// debugging and understanding the model's reasoning process. One of `auto`,
+	// `concise`, or `detailed`.
+	//
+	// Any of "auto", "concise", "detailed".
+	Summary ReasoningSummary `json:"summary,omitzero"`
 	paramObj
 }