openai · stainless-app · Nov 4, 2025 · Oct 21, 2025 · Oct 28, 2025 · Nov 3, 2025
@@ -1,3 +1,3 @@
 {
-  ".": "4.6.1"
+  ".": "4.7.0"
 }
@@ -1,4 +1,4 @@
 configured_endpoints: 135
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-a4bb37d110a22c2888f53e21281434686a6fffa3e672a40f2503ad9bd2759063.yml
-openapi_spec_hash: 2d59eefb494dff4eea8c3d008c7e2070
-config_hash: 50ee3382a63c021a9f821a935950e926
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-3c5d1593d7c6f2b38a7d78d7906041465ee9d6e9022f0651e1da194654488108.yml
+openapi_spec_hash: 0a4d8ad2469823ce24a3fd94f23f1c2b
+config_hash: 032995825500a503a76da119f5354905
@@ -1,5 +1,19 @@
 # Changelog
 
+## 4.7.0 (2025-11-04)
+
+Full Changelog: [v4.6.1...v4.7.0](https://github.com/openai/openai-java/compare/v4.6.1...v4.7.0)
+
+### Features
+
+* **api:** Realtime API token_limits, Hybrid searching ranking options ([bd9bcfd](https://github.com/openai/openai-java/commit/bd9bcfdd560cfc8df2a9336d162a0ee1f6604b84))
+* **api:** remove InputAudio from ResponseInputContent ([630fecf](https://github.com/openai/openai-java/commit/630fecf8f0e04ce82ac0f0df9b5d60df4edd0655))
+
+
+### Bug Fixes
+
+* **api:** docs updates ([3e970ec](https://github.com/openai/openai-java/commit/3e970ec8c9b3895b087f5722a0bfae23ca9e4e2c))
+
 ## 4.6.1 (2025-10-20)
 
 Full Changelog: [v4.6.0...v4.6.1](https://github.com/openai/openai-java/compare/v4.6.0...v4.6.1)

@@ -2,16 +2,16 @@
 
 <!-- x-release-please-start-version -->
 
-[![Maven Central](https://img.shields.io/maven-central/v/com.openai/openai-java)](https://central.sonatype.com/artifact/com.openai/openai-java/4.6.1)
-[![javadoc](https://javadoc.io/badge2/com.openai/openai-java/4.6.1/javadoc.svg)](https://javadoc.io/doc/com.openai/openai-java/4.6.1)
+[![Maven Central](https://img.shields.io/maven-central/v/com.openai/openai-java)](https://central.sonatype.com/artifact/com.openai/openai-java/4.7.0)
+[![javadoc](https://javadoc.io/badge2/com.openai/openai-java/4.7.0/javadoc.svg)](https://javadoc.io/doc/com.openai/openai-java/4.7.0)
 
 <!-- x-release-please-end -->
 
 The OpenAI Java SDK provides convenient access to the [OpenAI REST API](https://platform.openai.com/docs) from applications written in Java.
 
 <!-- x-release-please-start-version -->
 
-The REST API documentation can be found on [platform.openai.com](https://platform.openai.com/docs). Javadocs are available on [javadoc.io](https://javadoc.io/doc/com.openai/openai-java/4.6.1).
+The REST API documentation can be found on [platform.openai.com](https://platform.openai.com/docs). Javadocs are available on [javadoc.io](https://javadoc.io/doc/com.openai/openai-java/4.7.0).
 
 <!-- x-release-please-end -->
 
@@ -24,7 +24,7 @@ The REST API documentation can be found on [platform.openai.com](https://platfor
 ### Gradle
 
 ```kotlin
-implementation("com.openai:openai-java:4.6.1")
+implementation("com.openai:openai-java:4.7.0")
 ```
 
 ### Maven
@@ -33,7 +33,7 @@ implementation("com.openai:openai-java:4.6.1")
 <dependency>
   <groupId>com.openai</groupId>
   <artifactId>openai-java</artifactId>
-  <version>4.6.1</version>
+  <version>4.7.0</version>
 </dependency>
 ```
 
@@ -1342,7 +1342,7 @@ If you're using Spring Boot, then you can use the SDK's [Spring Boot starter](ht
 #### Gradle
 
 ```kotlin
-implementation("com.openai:openai-java-spring-boot-starter:4.6.1")
+implementation("com.openai:openai-java-spring-boot-starter:4.7.0")
 ```
 
 #### Maven
@@ -1351,7 +1351,7 @@ implementation("com.openai:openai-java-spring-boot-starter:4.6.1")
 <dependency>
   <groupId>com.openai</groupId>
   <artifactId>openai-java-spring-boot-starter</artifactId>
-  <version>4.6.1</version>
+  <version>4.7.0</version>
 </dependency>
 ```
 

@@ -8,7 +8,7 @@ repositories {
 
 allprojects {
     group = "com.openai"
-    version = "4.6.1" // x-release-please-version
+    version = "4.7.0" // x-release-please-version
 }
 
 subprojects {

@@ -38,16 +38,20 @@ private constructor(
     private val _json: JsonValue? = null,
 ) {
 
+    /** Unconstrained free-form text. */
     fun text(): Optional<JsonValue> = Optional.ofNullable(text)
 
+    /** A grammar defined by the user. */
     fun grammar(): Optional<Grammar> = Optional.ofNullable(grammar)
 
     fun isText(): Boolean = text != null
 
     fun isGrammar(): Boolean = grammar != null
 
+    /** Unconstrained free-form text. */
     fun asText(): JsonValue = text.getOrThrow("text")
 
+    /** A grammar defined by the user. */
     fun asGrammar(): Grammar = grammar.getOrThrow("grammar")
 
     fun _json(): Optional<JsonValue> = Optional.ofNullable(_json)
@@ -130,9 +134,11 @@ private constructor(
 
     companion object {
 
+        /** Unconstrained free-form text. */
         @JvmStatic
         fun ofText() = CustomToolInputFormat(text = JsonValue.from(mapOf("type" to "text")))
 
+        /** A grammar defined by the user. */
         @JvmStatic fun ofGrammar(grammar: Grammar) = CustomToolInputFormat(grammar = grammar)
     }
 
@@ -142,8 +148,10 @@ private constructor(
      */
     interface Visitor<out T> {
 
+        /** Unconstrained free-form text. */
         fun visitText(text: JsonValue): T
 
+        /** A grammar defined by the user. */
         fun visitGrammar(grammar: Grammar): T
 
         /**
@@ -202,6 +210,7 @@ private constructor(
         }
     }
 
+    /** A grammar defined by the user. */
     class Grammar
     @JsonCreator(mode = JsonCreator.Mode.DISABLED)
     private constructor(

@@ -72,6 +72,8 @@ private constructor(
      * 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`.
      *
+     * `concise` is only supported for `computer-use-preview` models.
+     *
      * @throws OpenAIInvalidDataException if the JSON field has an unexpected type (e.g. if the
      *   server responded with an unexpected value).
      */
@@ -187,6 +189,8 @@ private constructor(
         /**
          * 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`.
+         *
+         * `concise` is only supported for `computer-use-preview` models.
          */
         fun summary(summary: Summary?) = summary(JsonField.ofNullable(summary))
 
@@ -406,6 +410,8 @@ private constructor(
     /**
      * 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`.
+     *
+     * `concise` is only supported for `computer-use-preview` models.
      */
     class Summary @JsonCreator private constructor(private val value: JsonField<String>) : Enum {
 

@@ -25,17 +25,14 @@ import kotlin.io.path.name
 /**
  * Upload a file that can be used across various endpoints. Individual files can be up to 512 MB,
  * and the size of all files uploaded by one organization can be up to 1 TB.
- *
- * The Assistants API supports files up to 2 million tokens and of specific file types. See the
- * [Assistants Tools guide](https://platform.openai.com/docs/assistants/tools) for details.
- *
- * The Fine-tuning API only supports `.jsonl` files. The input also has certain required formats for
- * fine-tuning [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input) or
- * [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
- * models.
- *
- * The Batch API only supports `.jsonl` files up to 200 MB in size. The input also has a specific
- * required [format](https://platform.openai.com/docs/api-reference/batch/request-input).
+ * - The Assistants API supports files up to 2 million tokens and of specific file types. See the
+ *   [Assistants Tools guide](https://platform.openai.com/docs/assistants/tools) for details.
+ * - The Fine-tuning API only supports `.jsonl` files. The input also has certain required formats
+ *   for fine-tuning [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input)
+ *   or [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
+ *   models.
+ * - The Batch API only supports `.jsonl` files up to 200 MB in size. The input also has a specific
+ *   required [format](https://platform.openai.com/docs/api-reference/batch/request-input).
  *
  * Please [contact us](https://help.openai.com/) if you need to increase these storage limits.
  */

@@ -80,7 +80,9 @@ private constructor(
     fun background(): Optional<Background> = body.background()
 
     /**
-     *             Control how much effort the model will exert to match the style and features, especially facial features, of input images. This parameter is only supported for `gpt-image-1`. Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
+     * Control how much effort the model will exert to match the style and features, especially
+     * facial features, of input images. This parameter is only supported for `gpt-image-1`.
+     * Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
      *
      * @throws OpenAIInvalidDataException if the JSON field has an unexpected type (e.g. if the
      *   server responded with an unexpected value).
@@ -429,7 +431,9 @@ private constructor(
         }
 
         /**
-         *             Control how much effort the model will exert to match the style and features, especially facial features, of input images. This parameter is only supported for `gpt-image-1`. Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
+         * Control how much effort the model will exert to match the style and features, especially
+         * facial features, of input images. This parameter is only supported for `gpt-image-1`.
+         * Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
          */
         fun inputFidelity(inputFidelity: InputFidelity?) = apply {
             body.inputFidelity(inputFidelity)
@@ -903,7 +907,9 @@ private constructor(
         fun background(): Optional<Background> = background.value.getOptional("background")
 
         /**
-         *             Control how much effort the model will exert to match the style and features, especially facial features, of input images. This parameter is only supported for `gpt-image-1`. Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
+         * Control how much effort the model will exert to match the style and features, especially
+         * facial features, of input images. This parameter is only supported for `gpt-image-1`.
+         * Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
          *
          * @throws OpenAIInvalidDataException if the JSON field has an unexpected type (e.g. if the
          *   server responded with an unexpected value).
@@ -1297,7 +1303,10 @@ private constructor(
             }
 
             /**
-             *             Control how much effort the model will exert to match the style and features, especially facial features, of input images. This parameter is only supported for `gpt-image-1`. Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
+             * Control how much effort the model will exert to match the style and features,
+             * especially facial features, of input images. This parameter is only supported for
+             * `gpt-image-1`. Unsupported for `gpt-image-1-mini`. Supports `high` and `low`.
+             * Defaults to `low`.
              */
             fun inputFidelity(inputFidelity: InputFidelity?) =
                 inputFidelity(MultipartField.of(inputFidelity))
@@ -1981,7 +1990,9 @@ private constructor(
     }
 
     /**
-     *             Control how much effort the model will exert to match the style and features, especially facial features, of input images. This parameter is only supported for `gpt-image-1`. Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
+     * Control how much effort the model will exert to match the style and features, especially
+     * facial features, of input images. This parameter is only supported for `gpt-image-1`.
+     * Unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.
      */
     class InputFidelity @JsonCreator private constructor(private val value: JsonField<String>) :
         Enum {

@@ -221,8 +221,17 @@ private constructor(
     fun tracing(): Optional<RealtimeTracingConfig> = tracing.getOptional("tracing")
 
     /**
-     * Controls how the realtime conversation is truncated prior to model inference. The default is
-     * `auto`.
+     * When the number of tokens in a conversation exceeds the model's input token limit, the
+     * conversation be truncated, meaning messages (starting from the oldest) will not be included
+     * in the model's context. A 32k context model with 4,096 max output tokens can only include
+     * 28,224 tokens in the context before truncation occurs. Clients can configure truncation
+     * behavior to truncate with a lower max token limit, which is an effective way to control token
+     * usage and cost. Truncation will reduce the number of cached tokens on the next turn (busting
+     * the cache), since messages are dropped from the beginning of the context. However, clients
+     * can also configure truncation to retain messages up to a fraction of the maximum context
+     * size, which will reduce the need for future truncations and thus improve the cache rate.
+     * Truncation can be disabled entirely, which means the server will never truncate but would
+     * instead return an error if the conversation exceeds the model's input token limit.
      *
      * @throws OpenAIInvalidDataException if the JSON field has an unexpected type (e.g. if the
      *   server responded with an unexpected value).
@@ -666,8 +675,18 @@ private constructor(
             tracing(RealtimeTracingConfig.ofTracingConfiguration(tracingConfiguration))
 
         /**
-         * Controls how the realtime conversation is truncated prior to model inference. The default
-         * is `auto`.
+         * When the number of tokens in a conversation exceeds the model's input token limit, the
+         * conversation be truncated, meaning messages (starting from the oldest) will not be
+         * included in the model's context. A 32k context model with 4,096 max output tokens can
+         * only include 28,224 tokens in the context before truncation occurs. Clients can configure
+         * truncation behavior to truncate with a lower max token limit, which is an effective way
+         * to control token usage and cost. Truncation will reduce the number of cached tokens on
+         * the next turn (busting the cache), since messages are dropped from the beginning of the
+         * context. However, clients can also configure truncation to retain messages up to a
+         * fraction of the maximum context size, which will reduce the need for future truncations
+         * and thus improve the cache rate. Truncation can be disabled entirely, which means the
+         * server will never truncate but would instead return an error if the conversation exceeds
+         * the model's input token limit.
          */
         fun truncation(truncation: RealtimeTruncation) = truncation(JsonField.of(truncation))
 

@@ -42,7 +42,7 @@ private constructor(
     private val _json: JsonValue? = null,
 ) {
 
-    /** Default tracing mode for the session. */
+    /** Enables tracing and sets default values for tracing configuration options. Always `auto`. */
     fun auto(): Optional<JsonValue> = Optional.ofNullable(auto)
 
     /** Granular configuration for tracing. */
@@ -53,7 +53,7 @@ private constructor(
 
     fun isTracingConfiguration(): Boolean = tracingConfiguration != null
 
-    /** Default tracing mode for the session. */
+    /** Enables tracing and sets default values for tracing configuration options. Always `auto`. */
     fun asAuto(): JsonValue = auto.getOrThrow("auto")
 
     /** Granular configuration for tracing. */
@@ -144,7 +144,9 @@ private constructor(
 
     companion object {
 
-        /** Default tracing mode for the session. */
+        /**
+         * Enables tracing and sets default values for tracing configuration options. Always `auto`.
+         */
         @JvmStatic fun ofAuto() = RealtimeTracingConfig(auto = JsonValue.from("auto"))
 
         /** Granular configuration for tracing. */
@@ -159,7 +161,9 @@ private constructor(
      */
     interface Visitor<out T> {
 
-        /** Default tracing mode for the session. */
+        /**
+         * Enables tracing and sets default values for tracing configuration options. Always `auto`.
+         */
         fun visitAuto(auto: JsonValue): T
 
         /** Granular configuration for tracing. */

@@ -22,8 +22,17 @@ import java.util.Objects
 import java.util.Optional
 
 /**
- * Controls how the realtime conversation is truncated prior to model inference. The default is
- * `auto`.
+ * When the number of tokens in a conversation exceeds the model's input token limit, the
+ * conversation be truncated, meaning messages (starting from the oldest) will not be included in
+ * the model's context. A 32k context model with 4,096 max output tokens can only include 28,224
+ * tokens in the context before truncation occurs. Clients can configure truncation behavior to
+ * truncate with a lower max token limit, which is an effective way to control token usage and cost.
+ * Truncation will reduce the number of cached tokens on the next turn (busting the cache), since
+ * messages are dropped from the beginning of the context. However, clients can also configure
+ * truncation to retain messages up to a fraction of the maximum context size, which will reduce the
+ * need for future truncations and thus improve the cache rate. Truncation can be disabled entirely,
+ * which means the server will never truncate but would instead return an error if the conversation
+ * exceeds the model's input token limit.
  */
 @JsonDeserialize(using = RealtimeTruncation.Deserializer::class)
 @JsonSerialize(using = RealtimeTruncation.Serializer::class)