openai · kwhinnery-openai · Jun 2, 2025 · May 29, 2025 · Jun 2, 2025 · Jun 2, 2025
@@ -1,3 +1,3 @@
 {
-  ".": "0.5.0"
+  ".": "0.5.1"
 }
@@ -1,4 +1,4 @@
 configured_endpoints: 109
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-d4bcffecf0cdadf746faa6708ed1ec81fac451f9b857deabbab26f0a343b9314.yml
-openapi_spec_hash: 7c54a18b4381248bda7cc34c52142615
-config_hash: d23f847b9ebb3f427d0f198035bd3e9f
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-2bcc845d8635bf93ddcf9ee723af4d7928248412a417bee5fc10d863a1e13867.yml
+openapi_spec_hash: 865230cb3abeb01bd85de05891af23c4
+config_hash: ed1e6b3c5f93d12b80d31167f55c557c
@@ -1,5 +1,13 @@
 # Changelog
 
+## 0.5.1 (2025-06-02)
+
+Full Changelog: [v0.5.0...v0.5.1](https://github.com/openai/openai-ruby/compare/v0.5.0...v0.5.1)
+
+### Bug Fixes
+
+* **api:** Fix evals and code interpreter interfaces ([24a9100](https://github.com/openai/openai-ruby/commit/24a910015e6885fc19a2ad689fe70a148bed5787))
+
 ## 0.5.0 (2025-05-29)
 
 Full Changelog: [v0.4.1...v0.5.0](https://github.com/openai/openai-ruby/compare/v0.4.1...v0.5.0)

@@ -11,7 +11,7 @@ GIT
 PATH
   remote: .
   specs:
-    openai (0.5.0)
+    openai (0.5.1)
       connection_pool
 
 GEM

@@ -15,7 +15,7 @@ To use this gem, install via Bundler by adding the following to your application
 <!-- x-release-please-start-version -->
 
 ```ruby
-gem "openai", "~> 0.5.0"
+gem "openai", "~> 0.5.1"
 ```
 
 <!-- x-release-please-end -->

@@ -50,8 +50,8 @@ class Logprob < OpenAI::Internal::Type::BaseModel
           # @!attribute bytes
           #   The bytes that were used to generate the log probability.
           #
-          #   @return [Array<Object>, nil]
-          optional :bytes, OpenAI::Internal::Type::ArrayOf[OpenAI::Internal::Type::Unknown]
+          #   @return [Array<Integer>, nil]
+          optional :bytes, OpenAI::Internal::Type::ArrayOf[Integer]
 
           # @!attribute logprob
           #   The log probability of the token.
@@ -65,7 +65,7 @@ class Logprob < OpenAI::Internal::Type::BaseModel
           #
           #   @param token [String] The token that was used to generate the log probability.
           #
-          #   @param bytes [Array<Object>] The bytes that were used to generate the log probability.
+          #   @param bytes [Array<Integer>] The bytes that were used to generate the log probability.
           #
           #   @param logprob [Float] The log probability of the token.
         end

@@ -51,8 +51,8 @@ class Logprob < OpenAI::Internal::Type::BaseModel
           # @!attribute bytes
           #   The bytes that were used to generate the log probability.
           #
-          #   @return [Array<Object>, nil]
-          optional :bytes, OpenAI::Internal::Type::ArrayOf[OpenAI::Internal::Type::Unknown]
+          #   @return [Array<Integer>, nil]
+          optional :bytes, OpenAI::Internal::Type::ArrayOf[Integer]
 
           # @!attribute logprob
           #   The log probability of the token.
@@ -66,7 +66,7 @@ class Logprob < OpenAI::Internal::Type::BaseModel
           #
           #   @param token [String] The token that was used to generate the log probability.
           #
-          #   @param bytes [Array<Object>] The bytes that were used to generate the log probability.
+          #   @param bytes [Array<Integer>] The bytes that were used to generate the log probability.
           #
           #   @param logprob [Float] The log probability of the token.
         end

@@ -46,9 +46,9 @@ class ChatCompletion < OpenAI::Internal::Type::BaseModel
         #     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.
+        #     latency guarantee.
         #   - If set to 'default', the request will be processed using the default service
-        #     tier with a lower uptime SLA and no latency guarentee.
+        #     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).
@@ -195,9 +195,9 @@ class Logprobs < OpenAI::Internal::Type::BaseModel
         #   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.
+        #   latency guarantee.
         # - If set to 'default', the request will be processed using the default service
-        #   tier with a lower uptime SLA and no latency guarentee.
+        #   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).

@@ -45,9 +45,9 @@ class ChatCompletionChunk < OpenAI::Internal::Type::BaseModel
         #     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.
+        #     latency guarantee.
         #   - If set to 'default', the request will be processed using the default service
-        #     tier with a lower uptime SLA and no latency guarentee.
+        #     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).
@@ -378,9 +378,9 @@ class Logprobs < OpenAI::Internal::Type::BaseModel
         #   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.
+        #   latency guarantee.
         # - If set to 'default', the request will be processed using the default service
-        #   tier with a lower uptime SLA and no latency guarentee.
+        #   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).

@@ -226,9 +226,9 @@ class CompletionCreateParams < OpenAI::Internal::Type::BaseModel
         #     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.
+        #     latency guarantee.
         #   - If set to 'default', the request will be processed using the default service
-        #     tier with a lower uptime SLA and no latency guarentee.
+        #     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).
@@ -553,9 +553,9 @@ module ResponseFormat
         #   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.
+        #   latency guarantee.
         # - If set to 'default', the request will be processed using the default service
-        #   tier with a lower uptime SLA and no latency guarentee.
+        #   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).

@@ -16,26 +16,32 @@ class GraderRunParams < OpenAI::Internal::Type::BaseModel
           required :grader, union: -> { OpenAI::FineTuning::Alpha::GraderRunParams::Grader }
 
           # @!attribute model_sample
-          #   The model sample to be evaluated.
+          #   The model sample to be evaluated. This value will be used to populate the
+          #   `sample` namespace. See
+          #   [the guide](https://platform.openai.com/docs/guides/graders) for more details.
+          #   The `output_json` variable will be populated if the model sample is a valid JSON
+          #   string.
           #
           #   @return [String]
           required :model_sample, String
 
-          # @!attribute reference_answer
-          #   The reference answer for the evaluation.
+          # @!attribute item
+          #   The dataset item provided to the grader. This will be used to populate the
+          #   `item` namespace. See
+          #   [the guide](https://platform.openai.com/docs/guides/graders) for more details.
           #
-          #   @return [String, Object, Array<Object>, Float]
-          required :reference_answer,
-                   union: -> {
-                     OpenAI::FineTuning::Alpha::GraderRunParams::ReferenceAnswer
-                   }
+          #   @return [Object, nil]
+          optional :item, OpenAI::Internal::Type::Unknown
 
-          # @!method initialize(grader:, model_sample:, reference_answer:, request_options: {})
+          # @!method initialize(grader:, model_sample:, item: nil, request_options: {})
+          #   Some parameter documentations has been truncated, see
+          #   {OpenAI::Models::FineTuning::Alpha::GraderRunParams} for more details.
+          #
           #   @param grader [OpenAI::Models::Graders::StringCheckGrader, OpenAI::Models::Graders::TextSimilarityGrader, OpenAI::Models::Graders::PythonGrader, OpenAI::Models::Graders::ScoreModelGrader, OpenAI::Models::Graders::MultiGrader] The grader used for the fine-tuning job.
           #
-          #   @param model_sample [String] The model sample to be evaluated.
+          #   @param model_sample [String] The model sample to be evaluated. This value will be used to populate
           #
-          #   @param reference_answer [String, Object, Array<Object>, Float] The reference answer for the evaluation.
+          #   @param item [Object] The dataset item provided to the grader. This will be used to populate
           #
           #   @param request_options [OpenAI::RequestOptions, Hash{Symbol=>Object}]
 
@@ -63,25 +69,6 @@ module Grader
             # @!method self.variants
             #   @return [Array(OpenAI::Models::Graders::StringCheckGrader, OpenAI::Models::Graders::TextSimilarityGrader, OpenAI::Models::Graders::PythonGrader, OpenAI::Models::Graders::ScoreModelGrader, OpenAI::Models::Graders::MultiGrader)]
           end
-
-          # The reference answer for the evaluation.
-          module ReferenceAnswer
-            extend OpenAI::Internal::Type::Union
-
-            variant String
-
-            variant OpenAI::Internal::Type::Unknown
-
-            variant -> { OpenAI::Models::FineTuning::Alpha::GraderRunParams::ReferenceAnswer::UnionMember2Array }
-
-            variant Float
-
-            # @!method self.variants
-            #   @return [Array(String, Object, Array<Object>, Float)]
-
-            # @type [OpenAI::Internal::Type::Converter]
-            UnionMember2Array = OpenAI::Internal::Type::ArrayOf[OpenAI::Internal::Type::Unknown]
-          end
         end
       end
     end

@@ -226,7 +226,7 @@ class Hyperparameters < OpenAI::Internal::Type::BaseModel
           #   Number of examples in each batch. A larger batch size means that model
           #   parameters are updated less frequently, but with lower variance.
           #
-          #   @return [Object, Symbol, :auto, Integer, nil]
+          #   @return [Symbol, :auto, Integer, nil]
           optional :batch_size,
                    union: -> { OpenAI::FineTuning::FineTuningJob::Hyperparameters::BatchSize },
                    nil?: true
@@ -253,7 +253,7 @@ class Hyperparameters < OpenAI::Internal::Type::BaseModel
           #   The hyperparameters used for the fine-tuning job. This value will only be
           #   returned when running `supervised` jobs.
           #
-          #   @param batch_size [Object, Symbol, :auto, Integer, nil] Number of examples in each batch. A larger batch size means that model parameter
+          #   @param batch_size [Symbol, :auto, Integer, nil] Number of examples in each batch. A larger batch size means that model parameter
           #
           #   @param learning_rate_multiplier [Symbol, :auto, Float] Scaling factor for the learning rate. A smaller learning rate may be useful to a
           #
@@ -266,14 +266,12 @@ class Hyperparameters < OpenAI::Internal::Type::BaseModel
           module BatchSize
             extend OpenAI::Internal::Type::Union
 
-            variant OpenAI::Internal::Type::Unknown
-
             variant const: :auto
 
             variant Integer
 
             # @!method self.variants
-            #   @return [Array(Object, Symbol, :auto, Integer)]
+            #   @return [Array(Symbol, :auto, Integer)]
           end
 
           # Scaling factor for the learning rate. A smaller learning rate may be useful to

@@ -11,9 +11,11 @@ class MultiGrader < OpenAI::Internal::Type::BaseModel
         required :calculate_output, String
 
         # @!attribute graders
+        #   A StringCheckGrader object that performs a string comparison between input and
+        #   reference using a specified operation.
         #
-        #   @return [Hash{Symbol=>OpenAI::Models::Graders::StringCheckGrader, OpenAI::Models::Graders::TextSimilarityGrader, OpenAI::Models::Graders::PythonGrader, OpenAI::Models::Graders::ScoreModelGrader, OpenAI::Models::Graders::LabelModelGrader}]
-        required :graders, -> { OpenAI::Internal::Type::HashOf[union: OpenAI::Graders::MultiGrader::Grader] }
+        #   @return [OpenAI::Models::Graders::StringCheckGrader, OpenAI::Models::Graders::TextSimilarityGrader, OpenAI::Models::Graders::PythonGrader, OpenAI::Models::Graders::ScoreModelGrader, OpenAI::Models::Graders::LabelModelGrader]
+        required :graders, union: -> { OpenAI::Graders::MultiGrader::Graders }
 
         # @!attribute name
         #   The name of the grader.
@@ -28,20 +30,25 @@ class MultiGrader < OpenAI::Internal::Type::BaseModel
         required :type, const: :multi
 
         # @!method initialize(calculate_output:, graders:, name:, type: :multi)
+        #   Some parameter documentations has been truncated, see
+        #   {OpenAI::Models::Graders::MultiGrader} for more details.
+        #
         #   A MultiGrader object combines the output of multiple graders to produce a single
         #   score.
         #
         #   @param calculate_output [String] A formula to calculate the output based on grader results.
         #
-        #   @param graders [Hash{Symbol=>OpenAI::Models::Graders::StringCheckGrader, OpenAI::Models::Graders::TextSimilarityGrader, OpenAI::Models::Graders::PythonGrader, OpenAI::Models::Graders::ScoreModelGrader, OpenAI::Models::Graders::LabelModelGrader}]
+        #   @param graders [OpenAI::Models::Graders::StringCheckGrader, OpenAI::Models::Graders::TextSimilarityGrader, OpenAI::Models::Graders::PythonGrader, OpenAI::Models::Graders::ScoreModelGrader, OpenAI::Models::Graders::LabelModelGrader] A StringCheckGrader object that performs a string comparison between input and r
         #
         #   @param name [String] The name of the grader.
         #
         #   @param type [Symbol, :multi] The object type, which is always `multi`.
 
         # A StringCheckGrader object that performs a string comparison between input and
         # reference using a specified operation.
-        module Grader
+        #
+        # @see OpenAI::Models::Graders::MultiGrader#graders
+        module Graders
           extend OpenAI::Internal::Type::Union
 
           # A StringCheckGrader object that performs a string comparison between input and reference using a specified operation.

@@ -11,7 +11,7 @@ class ImageEditParams < OpenAI::Internal::Type::BaseModel
       #   The image(s) to edit. Must be a supported image file or an array of images.
       #
       #   For `gpt-image-1`, each image should be a `png`, `webp`, or `jpg` file less than
-      #   25MB. You can provide up to 16 images.
+      #   50MB. You can provide up to 16 images.
       #
       #   For `dall-e-2`, you can only provide one image, and it should be a square `png`
       #   file less than 4MB.
@@ -123,7 +123,7 @@ class ImageEditParams < OpenAI::Internal::Type::BaseModel
       # The image(s) to edit. Must be a supported image file or an array of images.
       #
       # For `gpt-image-1`, each image should be a `png`, `webp`, or `jpg` file less than
-      # 25MB. You can provide up to 16 images.
+      # 50MB. You can provide up to 16 images.
       #
       # For `dall-e-2`, you can only provide one image, and it should be a square `png`
       # file less than 4MB.

@@ -173,9 +173,9 @@ class Response < OpenAI::Internal::Type::BaseModel
         #     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.
+        #     latency guarantee.
         #   - If set to 'default', the request will be processed using the default service
-        #     tier with a lower uptime SLA and no latency guarentee.
+        #     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).
@@ -346,9 +346,9 @@ module ToolChoice
         #   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.
+        #   latency guarantee.
         # - If set to 'default', the request will be processed using the default service
-        #   tier with a lower uptime SLA and no latency guarentee.
+        #   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).