API_KEY='...'
X_PORTKEY_API_KEY='...'
X_PORTKEY_CONFIG='...'
X_PORTKEY_CUSTOM_HOST='...'
X_PORTKEY_PROVIDER='...'
X_PORTKEY_VIRTUAL_KEY='...'

npx @open-mcp/config add portkey-ai \
  ~/Library/Application\ Support/Claude/claude_desktop_config.json \
  --API_KEY=$API_KEY \
  --X_PORTKEY_API_KEY=$X_PORTKEY_API_KEY \
  --X_PORTKEY_CONFIG=$X_PORTKEY_CONFIG \
  --X_PORTKEY_CUSTOM_HOST=$X_PORTKEY_CUSTOM_HOST \
  --X_PORTKEY_PROVIDER=$X_PORTKEY_PROVIDER \
  --X_PORTKEY_VIRTUAL_KEY=$X_PORTKEY_VIRTUAL_KEY

npx @open-mcp/config add portkey-ai \
  .cursor/mcp.json \
  --API_KEY=$API_KEY \
  --X_PORTKEY_API_KEY=$X_PORTKEY_API_KEY \
  --X_PORTKEY_CONFIG=$X_PORTKEY_CONFIG \
  --X_PORTKEY_CUSTOM_HOST=$X_PORTKEY_CUSTOM_HOST \
  --X_PORTKEY_PROVIDER=$X_PORTKEY_PROVIDER \
  --X_PORTKEY_VIRTUAL_KEY=$X_PORTKEY_VIRTUAL_KEY

npx @open-mcp/config add portkey-ai \
  /path/to/client/config.json \
  --API_KEY=$API_KEY \
  --X_PORTKEY_API_KEY=$X_PORTKEY_API_KEY \
  --X_PORTKEY_CONFIG=$X_PORTKEY_CONFIG \
  --X_PORTKEY_CUSTOM_HOST=$X_PORTKEY_CUSTOM_HOST \
  --X_PORTKEY_PROVIDER=$X_PORTKEY_PROVIDER \
  --X_PORTKEY_VIRTUAL_KEY=$X_PORTKEY_VIRTUAL_KEY

{
  "mcpServers": {
    "portkey-ai": {
      "command": "npx",
      "args": ["-y", "@open-mcp/portkey-ai"],
      "env": {"API_KEY":"...","X_PORTKEY_API_KEY":"...","X_PORTKEY_CONFIG":"...","X_PORTKEY_CUSTOM_HOST":"...","X_PORTKEY_PROVIDER":"...","X_PORTKEY_VIRTUAL_KEY":"..."}
    }
  }
}

npx -y @modelcontextprotocol/inspector npx -y @open-mcp/portkey-ai

{
  toolName: z.string(),
  jsonPointers: z.array(z.string().startsWith("/").describe("The pointer to the JSON schema object which needs expanding")).describe("A list of JSON pointers"),
}

{
  "messages": z.array(z.object({ "content": z.string().describe("The contents of the system message."), "role": z.literal("system").describe("The role of the messages author, in this case \`system\`."), "name": z.string().describe("An optional name for the participant. Provides the model information to differentiate between participants of the same role.").optional() })).min(1).describe("A list of messages comprising the conversation so far. [Example Python code](https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models)."),
  "model": z.union([z.string(), z.enum(["gpt-4o","gpt-4o-2024-05-13","gpt-4-turbo","gpt-4-turbo-2024-04-09","gpt-4-0125-preview","gpt-4-turbo-preview","gpt-4-1106-preview","gpt-4-vision-preview","gpt-4","gpt-4-0314","gpt-4-0613","gpt-4-32k","gpt-4-32k-0314","gpt-4-32k-0613","gpt-3.5-turbo","gpt-3.5-turbo-16k","gpt-3.5-turbo-0301","gpt-3.5-turbo-0613","gpt-3.5-turbo-1106","gpt-3.5-turbo-0125","gpt-3.5-turbo-16k-0613"])]).describe("ID of the model to use. See the [model endpoint compatibility](https://platform.openai.com/docs/models/model-endpoint-compatibility) table for details on which models work with the Chat API."),
  "frequency_penalty": z.number().gte(-2).lte(2).nullable().describe("Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.\n\n[See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)\n").optional(),
  "logit_bias": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nModify the likelihood of specified tokens appearing in the completion.\n\nAccepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.\n").optional(),
  "logprobs": z.boolean().nullable().describe("Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the \`content\` of \`message\`.").optional(),
  "top_logprobs": z.number().int().gte(0).lte(20).nullable().describe("An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. \`logprobs\` must be set to \`true\` if this parameter is used.").optional(),
  "max_tokens": z.number().int().nullable().describe("The maximum number of [tokens](https://platform.openai.com/tokenizer?view=bpe) that can be generated in the chat completion.\n\nThe total length of input tokens and generated tokens is limited by the model's context length. [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken) for counting tokens.\n").optional(),
  "n": z.number().int().gte(1).lte(128).nullable().describe("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 choices. Keep \`n\` as \`1\` to minimize costs.").optional(),
  "presence_penalty": z.number().gte(-2).lte(2).nullable().describe("Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.\n\n[See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)\n").optional(),
  "response_format": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nAn object specifying the format that the model must output.\n\nSetting to \`{ \"type\": \"json_schema\", \"json_schema\": {...} }\`enables Structured Outputs which ensures the model will match your\nsupplied JSON schema. Works across all the providers that support this functionality. [OpenAI & Azure OpenAI](/integrations/llms/openai/structured-outputs), [Gemini & Vertex AI](/integrations/llms/vertex-ai/controlled-generations).\n\nSetting to \`{ \"type\": \"json_object\" }\` enables the older JSON mode, which ensures the message the model generates is valid JSON.\n\nUsing \`json_schema\` is preferred for models that support it.\n").optional(),
  "seed": z.number().int().gte(-9223372036854776000).lte(9223372036854776000).nullable().describe("This feature is in Beta.\nIf 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.\nDeterminism is not guaranteed, and you should refer to the \`system_fingerprint\` response parameter to monitor changes in the backend.\n").optional(),
  "stop": z.string().nullable().describe("Up to 4 sequences where the API will stop generating further tokens.\n").optional(),
  "stream": z.boolean().nullable().describe("If set, partial message deltas will be sent, like in ChatGPT. Tokens will be sent as data-only [server-sent events](https://developer.mozilla.org/en-UShttps://platform.openai.com/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format) as they become available, with the stream terminated by a \`data: [DONE]\` message. [Example Python code](https://cookbook.openai.com/examples/how_to_stream_completions).\n").optional(),
  "stream_options": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nOptions for streaming response. Only set this when you set \`stream: true\`.\n").optional(),
  "thinking": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nView the thinking/reasoning tokens as part of your response. Thinking models produce a long internal chain of thought before generating a response. Supported only for specific Claude models on Anthropic, Google Vertex AI, and AWS Bedrock.  Requires setting \`strict_openai_compliance = false\` in your API call.\n").optional(),
  "temperature": z.number().gte(0).lte(2).nullable().describe("What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.\n\nWe generally recommend altering this or \`top_p\` but not both.\n").optional(),
  "top_p": z.number().gte(0).lte(1).nullable().describe("An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n\nWe generally recommend altering this or \`temperature\` but not both.\n").optional(),
  "tools": z.array(z.object({ "type": z.literal("function").describe("The type of the tool. Currently, only \`function\` is supported."), "function": z.object({ "description": z.string().describe("A description of what the function does, used by the model to choose when and how to call the function.").optional(), "name": z.string().describe("The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."), "parameters": z.record(z.any()).describe("The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format. \n\nOmitting \`parameters\` defines a function with an empty parameter list.").optional(), "strict": z.boolean().nullable().describe("Whether to enable strict schema adherence when generating the function call. If 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).") }) })).describe("A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported.\n").optional(),
  "tool_choice": z.enum(["none","auto","required"]).describe("\`none\` means the model will not call any tool and instead generates a message. \`auto\` means the model can pick between generating a message or calling one or more tools. \`required\` means the model must call one or more tools.\n").describe("Controls which (if any) tool is called by the model.\n\`none\` means the model will not call any tool and instead generates a message.\n\`auto\` means the model can pick between generating a message or calling one or more tools.\n\`required\` means the model must call one or more tools.\nSpecifying a particular tool via \`{\"type\": \"function\", \"function\": {\"name\": \"my_function\"}}\` forces the model to call that tool.\n\n\`none\` is the default when no tools are present. \`auto\` is the default if tools are present.\n").optional(),
  "parallel_tool_calls": z.boolean().describe("Whether to enable [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling) during tool use.").optional(),
  "user": z.string().describe("A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).\n").optional(),
  "x-portkey-trace-id": z.string().describe("An ID you can pass to refer to one or more requests later on. If not provided, Portkey generates a trace ID automatically for each request. [Docs](https://portkey.ai/docs/product/observability/traces)").optional(),
  "x-portkey-span-id": z.string().describe("An ID you can pass to refer to a span under a trace.").optional(),
  "x-portkey-parent-span-id": z.string().describe("Link a child span to a parent span").optional(),
  "x-portkey-span-name": z.string().describe("Name for the Span ID").optional(),
  "x-portkey-metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nPass any arbitrary metadata along with your request").optional(),
  "x-portkey-cache-namespace": z.string().describe("Partition your Portkey cache store based on custom strings, ignoring metadata and other headers").optional(),
  "x-portkey-cache-force-refresh": z.boolean().describe("Forces a cache refresh for your request by making a new API call and storing the updated value").optional()
}

{
  "model": z.union([z.string(), z.enum(["gpt-3.5-turbo-instruct","davinci-002","babbage-002"])]).describe("ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models/overview) for descriptions of them.\n"),
  "prompt": z.string().nullable().describe("The prompt(s) to generate completions for, encoded as a string, array of strings, array of tokens, or array of token arrays.\n\nNote that <|endoftext|> is the document separator that the model sees during training, so if a prompt is not specified the model will generate as if from the beginning of a new document.\n"),
  "best_of": z.number().int().gte(0).lte(20).nullable().describe("Generates \`best_of\` completions server-side and returns the \"best\" (the one with the highest log probability per token). Results cannot be streamed.\n\nWhen used with \`n\`, \`best_of\` controls the number of candidate completions and \`n\` specifies how many to return – \`best_of\` must be greater than \`n\`.\n\n**Note:** Because this parameter generates many completions, it can quickly consume your token quota. Use carefully and ensure that you have reasonable settings for \`max_tokens\` and \`stop\`.\n").optional(),
  "echo": z.boolean().nullable().describe("Echo back the prompt in addition to the completion\n").optional(),
  "frequency_penalty": z.number().gte(-2).lte(2).nullable().describe("Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.\n\n[See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)\n").optional(),
  "logit_bias": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nModify the likelihood of specified tokens appearing in the completion.\n\nAccepts a JSON object that maps tokens (specified by their token ID in the GPT tokenizer) to an associated bias value from -100 to 100. You can use this [tokenizer tool](https://platform.openai.com/tokenizer?view=bpe) to convert text to token IDs. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.\n\nAs an example, you can pass \`{\"50256\": -100}\` to prevent the <|endoftext|> token from being generated.\n").optional(),
  "logprobs": z.number().int().gte(0).lte(5).nullable().describe("Include the log probabilities on the \`logprobs\` most likely output tokens, as well the chosen tokens. For example, if \`logprobs\` is 5, the API will return a list of the 5 most likely tokens. The API will always return the \`logprob\` of the sampled token, so there may be up to \`logprobs+1\` elements in the response.\n\nThe maximum value for \`logprobs\` is 5.\n").optional(),
  "max_tokens": z.number().int().gte(0).nullable().describe("The maximum number of [tokens](https://platform.openai.com/tokenizer?view=bpe) that can be generated in the completion.\n\nThe token count of your prompt plus \`max_tokens\` cannot exceed the model's context length. [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken) for counting tokens.\n").optional(),
  "n": z.number().int().gte(1).lte(128).nullable().describe("How many completions to generate for each prompt.\n\n**Note:** Because this parameter generates many completions, it can quickly consume your token quota. Use carefully and ensure that you have reasonable settings for \`max_tokens\` and \`stop\`.\n").optional(),
  "presence_penalty": z.number().gte(-2).lte(2).nullable().describe("Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.\n\n[See more information about frequency and presence penalties.](https://platform.openai.com/docs/guides/text-generation/parameter-details)\n").optional(),
  "seed": z.number().int().gte(-9223372036854776000).lte(9223372036854776000).nullable().describe("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.\n\nDeterminism is not guaranteed, and you should refer to the \`system_fingerprint\` response parameter to monitor changes in the backend.\n").optional(),
  "stop": z.string().nullable().nullable().describe("Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence.\n").optional(),
  "stream": z.boolean().nullable().describe("Whether to stream back partial progress. If set, tokens will be sent as data-only [server-sent events](https://developer.mozilla.org/en-UShttps://platform.openai.com/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format) as they become available, with the stream terminated by a \`data: [DONE]\` message. [Example Python code](https://cookbook.openai.com/examples/how_to_stream_completions).\n").optional(),
  "stream_options": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nOptions for streaming response. Only set this when you set \`stream: true\`.\n").optional(),
  "suffix": z.string().nullable().describe("The suffix that comes after a completion of inserted text.\n\nThis parameter is only supported for \`gpt-3.5-turbo-instruct\`.\n").optional(),
  "temperature": z.number().gte(0).lte(2).nullable().describe("What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.\n\nWe generally recommend altering this or \`top_p\` but not both.\n").optional(),
  "top_p": z.number().gte(0).lte(1).nullable().describe("An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n\nWe generally recommend altering this or \`temperature\` but not both.\n").optional(),
  "user": z.string().describe("A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).\n").optional()
}

{
  "workspace_id": z.string().describe("ID or slug of the workspace").optional(),
  "name": z.string().describe("Name of the collection"),
  "parent_collection_id": z.string().describe("ID or slug of the parent collection (optional)").optional()
}

{
  "workspace_id": z.string().describe("ID or slug of the workspace"),
  "current_page": z.number().int().gte(0).describe("Page number for pagination (0-indexed)").optional(),
  "page_size": z.number().int().gte(1).describe("Number of items per page").optional(),
  "search": z.string().describe("Search query to filter collections by name").optional()
}

{}

{}

{
  "name": z.string().describe("New name for the collection")
}

{}

{
  "organisation_id": z.string().uuid().describe("ID of the organisation").optional(),
  "workspace_id": z.string().describe("ID or slug of the workspace").optional(),
  "name": z.string().regex(new RegExp("^(?!latest$)(?!default$)(?!\\d+$)[a-zA-Z0-9_-]*[^@\\s]?[a-zA-Z0-9_-]*$")).describe("Name of the label"),
  "description": z.string().describe("Description of the label").optional(),
  "color_code": z.string().describe("Color code for the label").optional()
}

{
  "organisation_id": z.string().uuid().describe("ID of the organisation").optional(),
  "workspace_id": z.string().describe("ID or slug of the workspace").optional(),
  "search": z.string().describe("Search query to filter labels by name").optional(),
  "current_page": z.number().int().gte(0).describe("Page number for pagination").optional(),
  "page_size": z.number().int().gte(1).describe("Number of items per page").optional()
}

{
  "labelId": z.string().uuid().describe("ID of the label to retrieve"),
  "organisation_id": z.string().uuid().describe("ID of the organisation").optional(),
  "workspace_id": z.string().describe("ID or slug of the workspace").optional()
}

{
  "labelId": z.string().uuid().describe("ID of the label to update"),
  "name": z.string().regex(new RegExp("^(?!latest$)(?!default$)(?!\\d+$)[a-zA-Z0-9_-]*[^@\\s]?[a-zA-Z0-9_-]*$")).describe("Name of the label").optional(),
  "description": z.string().describe("Description of the label").optional(),
  "color_code": z.string().describe("Color code for the label").optional()
}

{
  "labelId": z.string().uuid().describe("ID of the label to delete")
}

{
  "name": z.string(),
  "collection_id": z.string().describe("UUID or slug of the collection"),
  "string": z.string().describe("Prompt template in string format"),
  "parameters": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nParameters for the prompt"),
  "functions": z.array(z.record(z.any())).describe("Functions for the prompt").optional(),
  "tools": z.array(z.record(z.any())).describe("Tools for the prompt").optional(),
  "tool_choice": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nTool Choice for the prompt").optional(),
  "model": z.string().describe("The model to use for the prompt").optional(),
  "virtual_key": z.string().describe("The virtual key to use for the prompt").optional(),
  "version_description": z.string().describe("The description of the prompt version").optional(),
  "template_metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nMetadata for the prompt").optional()
}

{
  "collection_id": z.string().optional(),
  "workspace_id": z.string().optional(),
  "current_page": z.number().int().optional(),
  "page_size": z.number().int().optional(),
  "search": z.string().optional()
}

{
  "promptId": z.string()
}

{
  "promptId": z.string(),
  "name": z.string().optional(),
  "collection_id": z.string().optional(),
  "string": z.string().optional(),
  "parameters": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:").optional(),
  "model": z.string().optional(),
  "virtual_key": z.string().optional(),
  "version_description": z.string().optional(),
  "functions": z.array(z.record(z.any())).optional(),
  "tools": z.array(z.record(z.any())).optional(),
  "tool_choice": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:").optional(),
  "is_raw_template": z.union([z.literal(0), z.literal(1)]).optional(),
  "prompt_metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:").optional()
}

{
  "promptId": z.string()
}

{
  "promptId": z.string()
}

{
  "promptId": z.string(),
  "versionId": z.string().uuid()
}

{
  "promptId": z.string(),
  "versionId": z.string().uuid(),
  "label_id": z.string().uuid().optional()
}

{
  "promptId": z.string(),
  "version": z.number().describe("Version Number to set as default")
}

{
  "workspace_id": z.string().describe("Required for Admin keys").optional(),
  "name": z.string(),
  "string": z.string().describe("Prompt partial template in string format"),
  "version_description": z.string().optional()
}

{
  "collection_id": z.string().optional()
}

{
  "promptPartialId": z.string()
}

{
  "promptPartialId": z.string(),
  "name": z.string().optional(),
  "string": z.string().optional(),
  "description": z.string().optional(),
  "status": z.string().optional()
}

{
  "promptPartialId": z.string()
}

{
  "promptPartialId": z.string()
}

{
  "promptPartialId": z.string(),
  "version": z.number().describe("Version Number to set as default")
}

{
  "promptId": z.string().describe("The unique identifier of the prompt template to use")
}

{
  "promptId": z.string().describe("The unique identifier of the prompt template to render")
}

{
  "prompt": z.string().describe("A text description of the desired image(s). The maximum length is 1000 characters for \`dall-e-2\` and 4000 characters for \`dall-e-3\`."),
  "model": z.union([z.string(), z.enum(["dall-e-2","dall-e-3"])]).nullable().describe("The model to use for image generation.").optional(),
  "n": z.number().int().gte(1).lte(10).nullable().describe("The number of images to generate. Must be between 1 and 10. For \`dall-e-3\`, only \`n=1\` is supported.").optional(),
  "quality": z.enum(["standard","hd"]).describe("The quality of the image that will be generated. \`hd\` creates images with finer details and greater consistency across the image. This param is only supported for \`dall-e-3\`.").optional(),
  "response_format": z.enum(["url","b64_json"]).nullable().describe("The format in which the generated images are returned. Must be one of \`url\` or \`b64_json\`. URLs are only valid for 60 minutes after the image has been generated.").optional(),
  "size": z.enum(["256x256","512x512","1024x1024","1792x1024","1024x1792"]).nullable().describe("The size of the generated images. Must be one of \`256x256\`, \`512x512\`, or \`1024x1024\` for \`dall-e-2\`. Must be one of \`1024x1024\`, \`1792x1024\`, or \`1024x1792\` for \`dall-e-3\` models.").optional(),
  "style": z.enum(["vivid","natural"]).nullable().describe("The style of the generated images. Must be one of \`vivid\` or \`natural\`. Vivid causes the model to lean towards generating hyper-real and dramatic images. Natural causes the model to produce more natural, less hyper-real looking images. This param is only supported for \`dall-e-3\`.").optional(),
  "user": z.string().describe("A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).\n").optional()
}

{}

{}

{
  "input": z.string().describe("The string that will be turned into an embedding.").describe("Input text to embed, encoded as a string or array of tokens. To embed multiple inputs in a single request, pass an array of strings or array of token arrays. The input must not exceed the max input tokens for the model (8192 tokens for \`text-embedding-ada-002\`), cannot be an empty string, and any array must be 2048 dimensions or less. [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken) for counting tokens.\n"),
  "model": z.union([z.string(), z.enum(["text-embedding-ada-002","text-embedding-3-small","text-embedding-3-large"])]).describe("ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models/overview) for descriptions of them.\n"),
  "encoding_format": z.enum(["float","base64"]).describe("The format to return the embeddings in. Can be either \`float\` or [\`base64\`](https://pypi.org/project/pybase64/).").optional(),
  "dimensions": z.number().int().gte(1).describe("The number of dimensions the resulting output embeddings should have. Only supported in \`text-embedding-3\` and later models.\n").optional(),
  "user": z.string().describe("A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).\n").optional()
}

{
  "model": z.union([z.string(), z.enum(["tts-1","tts-1-hd"])]).describe("One of the available [TTS models](https://platform.openai.com/docs/models/tts): \`tts-1\` or \`tts-1-hd\`\n"),
  "input": z.string().max(4096).describe("The text to generate audio for. The maximum length is 4096 characters."),
  "voice": z.enum(["alloy","echo","fable","onyx","nova","shimmer"]).describe("The voice to use when generating the audio. Supported voices are \`alloy\`, \`echo\`, \`fable\`, \`onyx\`, \`nova\`, and \`shimmer\`. Previews of the voices are available in the [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech/voice-options)."),
  "response_format": z.enum(["mp3","opus","aac","flac","wav","pcm"]).describe("The format to audio in. Supported formats are \`mp3\`, \`opus\`, \`aac\`, \`flac\`, \`wav\`, and \`pcm\`.").optional(),
  "speed": z.number().gte(0.25).lte(4).describe("The speed of the generated audio. Select a value from \`0.25\` to \`4.0\`. \`1.0\` is the default.").optional()
}

{}

{}

{
  "purpose": z.string().describe("Only return files with the given purpose.").optional()
}

{}

{
  "file_id": z.string().describe("The ID of the file to use for this request.")
}

{
  "file_id": z.string().describe("The ID of the file to use for this request.")
}

{
  "file_id": z.string().describe("The ID of the file to use for this request.")
}

{}

{
  "after": z.string().describe("Identifier for the last job from the previous pagination request.").optional(),
  "limit": z.number().int().describe("Number of fine-tuning jobs to retrieve.").optional()
}

{
  "fine_tuning_job_id": z.string().describe("The ID of the fine-tuning job.\n")
}

{
  "fine_tuning_job_id": z.string().describe("The ID of the fine-tuning job to get events for.\n"),
  "after": z.string().describe("Identifier for the last event from the previous pagination request.").optional(),
  "limit": z.number().int().describe("Number of events to retrieve.").optional()
}

{
  "fine_tuning_job_id": z.string().describe("The ID of the fine-tuning job to cancel.\n")
}

{
  "fine_tuning_job_id": z.string().describe("The ID of the fine-tuning job to get checkpoints for.\n"),
  "after": z.string().describe("Identifier for the last checkpoint ID from the previous pagination request.").optional(),
  "limit": z.number().int().describe("Number of checkpoints to retrieve.").optional()
}

{}

{
  "model": z.string().describe("The ID of the model to use for this request")
}

{
  "model": z.string().describe("The model to delete")
}

{
  "input": z.string().describe("The input text to classify"),
  "model": z.union([z.string(), z.enum(["text-moderation-latest","text-moderation-stable"])]).describe("Two content moderations models are available: \`text-moderation-stable\` and \`text-moderation-latest\`.\n\nThe default is \`text-moderation-latest\` which will be automatically upgraded over time. This ensures you are always using our most accurate model. If you use \`text-moderation-stable\`, we will provide advanced notice before updating the model. Accuracy of \`text-moderation-stable\` may be slightly lower than for \`text-moderation-latest\`.\n").optional()
}

{
  "limit": z.number().int().describe("A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.\n").optional(),
  "order": z.enum(["asc","desc"]).describe("Sort order by the \`created_at\` timestamp of the objects. \`asc\` for ascending order and \`desc\` for descending order.\n").optional(),
  "after": z.string().describe("A cursor for use in pagination. \`after\` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.\n").optional(),
  "before": z.string().describe("A cursor for use in pagination. \`before\` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.\n").optional()
}

{
  "model": z.union([z.string(), z.enum(["gpt-4o","gpt-4o-2024-05-13","gpt-4-turbo","gpt-4-turbo-2024-04-09","gpt-4-0125-preview","gpt-4-turbo-preview","gpt-4-1106-preview","gpt-4-vision-preview","gpt-4","gpt-4-0314","gpt-4-0613","gpt-4-32k","gpt-4-32k-0314","gpt-4-32k-0613","gpt-3.5-turbo","gpt-3.5-turbo-16k","gpt-3.5-turbo-0613","gpt-3.5-turbo-1106","gpt-3.5-turbo-0125","gpt-3.5-turbo-16k-0613"])]).describe("ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models/overview) for descriptions of them.\n"),
  "name": z.string().max(256).nullable().describe("The name of the assistant. The maximum length is 256 characters.\n").optional(),
  "description": z.string().max(512).nullable().describe("The description of the assistant. The maximum length is 512 characters.\n").optional(),
  "instructions": z.string().max(256000).nullable().describe("The system instructions that the assistant uses. The maximum length is 256,000 characters.\n").optional(),
  "tools": z.array(z.object({ "type": z.literal("code_interpreter").describe("The type of tool being defined: \`code_interpreter\`") })).max(128).describe("A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types \`code_interpreter\`, \`file_search\`, or \`function\`.\n").optional(),
  "tool_resources": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nA set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the \`code_interpreter\` tool requires a list of file IDs, while the \`file_search\` tool requires a list of vector store IDs.\n").optional(),
  "metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nSet of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional(),
  "temperature": z.number().gte(0).lte(2).nullable().describe("What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.\n").optional(),
  "top_p": z.number().gte(0).lte(1).nullable().describe("An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n\nWe generally recommend altering this or temperature but not both.\n").optional(),
  "response_format": z.enum(["none","auto"]).describe("\`auto\` is the default value\n").nullable().describe("Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since \`gpt-3.5-turbo-1106\`.\n\nSetting to \`{ \"type\": \"json_object\" }\` enables JSON mode, which guarantees the message the model generates is valid JSON.\n\n**Important:** when using JSON mode, you **must** also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly \"stuck\" request. Also note that the message content may be partially cut off if \`finish_reason=\"length\"\`, which indicates the generation exceeded \`max_tokens\` or the conversation exceeded the max context length.\n").optional()
}

{
  "assistant_id": z.string().describe("The ID of the assistant to retrieve.")
}

{
  "assistant_id": z.string().describe("The ID of the assistant to modify."),
  "model": z.string().describe("ID of the model to use. You can use the [List models](https://platform.openai.com/docs/api-reference/models/list) API to see all of your available models, or see our [Model overview](https://platform.openai.com/docs/models/overview) for descriptions of them.\n").optional(),
  "name": z.string().max(256).nullable().describe("The name of the assistant. The maximum length is 256 characters.\n").optional(),
  "description": z.string().max(512).nullable().describe("The description of the assistant. The maximum length is 512 characters.\n").optional(),
  "instructions": z.string().max(256000).nullable().describe("The system instructions that the assistant uses. The maximum length is 256,000 characters.\n").optional(),
  "tools": z.array(z.object({ "type": z.literal("code_interpreter").describe("The type of tool being defined: \`code_interpreter\`") })).max(128).describe("A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types \`code_interpreter\`, \`file_search\`, or \`function\`.\n").optional(),
  "tool_resources": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nA set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the \`code_interpreter\` tool requires a list of file IDs, while the \`file_search\` tool requires a list of vector store IDs.\n").optional(),
  "metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nSet of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional(),
  "temperature": z.number().gte(0).lte(2).nullable().describe("What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.\n").optional(),
  "top_p": z.number().gte(0).lte(1).nullable().describe("An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n\nWe generally recommend altering this or temperature but not both.\n").optional(),
  "response_format": z.enum(["none","auto"]).describe("\`auto\` is the default value\n").nullable().describe("Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since \`gpt-3.5-turbo-1106\`.\n\nSetting to \`{ \"type\": \"json_object\" }\` enables JSON mode, which guarantees the message the model generates is valid JSON.\n\n**Important:** when using JSON mode, you **must** also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly \"stuck\" request. Also note that the message content may be partially cut off if \`finish_reason=\"length\"\`, which indicates the generation exceeded \`max_tokens\` or the conversation exceeded the max context length.\n").optional()
}

{
  "assistant_id": z.string().describe("The ID of the assistant to delete.")
}

{}

{
  "response_id": z.string().describe("The ID of the response to retrieve."),
  "include": z.array(z.enum(["file_search_call.results","message.input_image.image_url","computer_call_output.output.image_url"]).describe("Specify additional output data to include in the model response. Currently\nsupported values are:\n- \`file_search_call.results\`: Include the search results of\n  the file search tool call.\n- \`message.input_image.image_url\`: Include image urls from the input message.\n- \`computer_call_output.output.image_url\`: Include image urls from the computer call output.\n")).describe("Specify additional output data to include in the response. Currently\nsupported values are:\n- \`file_search_call.results\`: Include the search results of\n  the file search tool call.\n- \`message.input_image.image_url\`: Include image urls from the input message.\n- \`computer_call_output.output.image_url\`: Include image urls from the computer call output.\n").optional()
}

{
  "response_id": z.string().describe("The ID of the response to delete.")
}

{
  "response_id": z.string().describe("The ID of the response to retrieve input items for."),
  "limit": z.number().int().describe("A limit on the number of objects to be returned. Limit can range between\n1 and 100, and the default is 20.\n").optional(),
  "order": z.enum(["asc","desc"]).describe("The order to return the input items in. Default is \`asc\`.\n- \`asc\`: Return the input items in ascending order.\n- \`desc\`: Return the input items in descending order.\n").optional(),
  "after": z.string().describe("An item ID to list items after, used in pagination.\n").optional(),
  "before": z.string().describe("An item ID to list items before, used in pagination.\n").optional()
}

{
  "messages": z.array(z.object({ "role": z.enum(["user","assistant"]).describe("The role of the entity that is creating the message. Allowed values include:\n- \`user\`: Indicates the message is sent by an actual user and should be used in most cases to represent user-generated messages.\n- \`assistant\`: Indicates the message is generated by the assistant. Use this value to insert messages from the assistant into the conversation.\n"), "content": z.string().describe("The text contents of the message."), "attachments": z.array(z.object({ "file_id": z.string().describe("The ID of the file to attach to the message.").optional(), "tools": z.array(z.object({ "type": z.literal("code_interpreter").describe("The type of tool being defined: \`code_interpreter\`") })).describe("The tools to add this file to.").optional() })).nullable().describe("A list of files attached to the message, and the tools they should be added to.").optional(), "metadata": z.record(z.any()).nullable().describe("Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional() }).strict()).describe("A list of [messages](https://platform.openai.com/docs/api-reference/messages) to start the thread with.").optional(),
  "tool_resources": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nA set of resources that are made available to the assistant's tools in this thread. The resources are specific to the type of tool. For example, the \`code_interpreter\` tool requires a list of file IDs, while the \`file_search\` tool requires a list of vector store IDs.\n").optional(),
  "metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nSet of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional()
}

{
  "thread_id": z.string().describe("The ID of the thread to retrieve.")
}

{
  "thread_id": z.string().describe("The ID of the thread to modify. Only the \`metadata\` can be modified."),
  "tool_resources": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nA set of resources that are made available to the assistant's tools in this thread. The resources are specific to the type of tool. For example, the \`code_interpreter\` tool requires a list of file IDs, while the \`file_search\` tool requires a list of vector store IDs.\n").optional(),
  "metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nSet of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional()
}

{
  "thread_id": z.string().describe("The ID of the thread to delete.")
}

{
  "thread_id": z.string().describe("The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) the messages belong to."),
  "limit": z.number().int().describe("A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.\n").optional(),
  "order": z.enum(["asc","desc"]).describe("Sort order by the \`created_at\` timestamp of the objects. \`asc\` for ascending order and \`desc\` for descending order.\n").optional(),
  "after": z.string().describe("A cursor for use in pagination. \`after\` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.\n").optional(),
  "before": z.string().describe("A cursor for use in pagination. \`before\` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.\n").optional(),
  "run_id": z.string().describe("Filter messages by the run ID that generated them.\n").optional()
}

{
  "thread_id": z.string().describe("The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) to create a message for."),
  "role": z.enum(["user","assistant"]).describe("The role of the entity that is creating the message. Allowed values include:\n- \`user\`: Indicates the message is sent by an actual user and should be used in most cases to represent user-generated messages.\n- \`assistant\`: Indicates the message is generated by the assistant. Use this value to insert messages from the assistant into the conversation.\n"),
  "content": z.string().describe("The text contents of the message."),
  "attachments": z.array(z.object({ "file_id": z.string().describe("The ID of the file to attach to the message.").optional(), "tools": z.array(z.object({ "type": z.literal("code_interpreter").describe("The type of tool being defined: \`code_interpreter\`") })).describe("The tools to add this file to.").optional() })).nullable().describe("A list of files attached to the message, and the tools they should be added to.").optional(),
  "metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nSet of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional()
}

{
  "thread_id": z.string().describe("The ID of the [thread](https://platform.openai.com/docs/api-reference/threads) to which this message belongs."),
  "message_id": z.string().describe("The ID of the message to retrieve.")
}

{
  "thread_id": z.string().describe("The ID of the thread to which this message belongs."),
  "message_id": z.string().describe("The ID of the message to modify."),
  "metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nSet of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional()
}

{
  "thread_id": z.string().describe("The ID of the thread to which this message belongs."),
  "message_id": z.string().describe("The ID of the message to delete.")
}

{
  "assistant_id": z.string().describe("The ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) to use to execute this run."),
  "thread": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nIf no thread is provided, an empty thread will be created.").optional(),
  "model": z.union([z.string(), z.enum(["gpt-4o","gpt-4o-2024-05-13","gpt-4-turbo","gpt-4-turbo-2024-04-09","gpt-4-0125-preview","gpt-4-turbo-preview","gpt-4-1106-preview","gpt-4-vision-preview","gpt-4","gpt-4-0314","gpt-4-0613","gpt-4-32k","gpt-4-32k-0314","gpt-4-32k-0613","gpt-3.5-turbo","gpt-3.5-turbo-16k","gpt-3.5-turbo-0613","gpt-3.5-turbo-1106","gpt-3.5-turbo-0125","gpt-3.5-turbo-16k-0613"])]).nullable().describe("The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to be used to execute this run. If a value is provided here, it will override the model associated with the assistant. If not, the model associated with the assistant will be used.").optional(),
  "instructions": z.string().nullable().describe("Override the default system message of the assistant. This is useful for modifying the behavior on a per-run basis.").optional(),
  "tools": z.array(z.object({ "type": z.literal("code_interpreter").describe("The type of tool being defined: \`code_interpreter\`") })).max(20).nullable().describe("Override the tools the assistant can use for this run. This is useful for modifying the behavior on a per-run basis.").optional(),
  "tool_resources": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nA set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the \`code_interpreter\` tool requires a list of file IDs, while the \`file_search\` tool requires a list of vector store IDs.\n").optional(),
  "metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nSet of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional(),
  "temperature": z.number().gte(0).lte(2).nullable().describe("What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.\n").optional(),
  "top_p": z.number().gte(0).lte(1).nullable().describe("An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n\nWe generally recommend altering this or temperature but not both.\n").optional(),
  "stream": z.boolean().nullable().describe("If \`true\`, returns a stream of events that happen during the Run as server-sent events, terminating when the Run enters a terminal state with a \`data: [DONE]\` message.\n").optional(),
  "max_prompt_tokens": z.number().int().gte(256).nullable().describe("The maximum number of prompt tokens that may be used over the course of the run. The run will make a best effort to use only the number of prompt tokens specified, across multiple turns of the run. If the run exceeds the number of prompt tokens specified, the run will end with status \`incomplete\`. See \`incomplete_details\` for more info.\n").optional(),
  "max_completion_tokens": z.number().int().gte(256).nullable().describe("The maximum number of completion tokens that may be used over the course of the run. The run will make a best effort to use only the number of completion tokens specified, across multiple turns of the run. If the run exceeds the number of completion tokens specified, the run will end with status \`incomplete\`. See \`incomplete_details\` for more info.\n").optional(),
  "truncation_strategy": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nControls for how a thread will be truncated prior to the run. Use this to control the intial context window of the run.").optional(),
  "tool_choice": z.enum(["none","auto","required"]).describe("\`none\` means the model will not call any tools and instead generates a message. \`auto\` means the model can pick between generating a message or calling one or more tools. \`required\` means the model must call one or more tools before responding to the user.\n").nullable().describe("Controls which (if any) tool is called by the model.\n\`none\` means the model will not call any tools and instead generates a message.\n\`auto\` is the default value and means the model can pick between generating a message or calling one or more tools.\n\`required\` means the model must call one or more tools before responding to the user.\nSpecifying a particular tool like \`{\"type\": \"file_search\"}\` or \`{\"type\": \"function\", \"function\": {\"name\": \"my_function\"}}\` forces the model to call that tool.\n").optional(),
  "parallel_tool_calls": z.boolean().describe("Whether to enable [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling) during tool use.").optional(),
  "response_format": z.enum(["none","auto"]).describe("\`auto\` is the default value\n").nullable().describe("Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since \`gpt-3.5-turbo-1106\`.\n\nSetting to \`{ \"type\": \"json_object\" }\` enables JSON mode, which guarantees the message the model generates is valid JSON.\n\n**Important:** when using JSON mode, you **must** also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly \"stuck\" request. Also note that the message content may be partially cut off if \`finish_reason=\"length\"\`, which indicates the generation exceeded \`max_tokens\` or the conversation exceeded the max context length.\n").optional()
}

{
  "thread_id": z.string().describe("The ID of the thread the run belongs to."),
  "limit": z.number().int().describe("A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.\n").optional(),
  "order": z.enum(["asc","desc"]).describe("Sort order by the \`created_at\` timestamp of the objects. \`asc\` for ascending order and \`desc\` for descending order.\n").optional(),
  "after": z.string().describe("A cursor for use in pagination. \`after\` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.\n").optional(),
  "before": z.string().describe("A cursor for use in pagination. \`before\` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.\n").optional()
}

{
  "thread_id": z.string().describe("The ID of the thread to run."),
  "assistant_id": z.string().describe("The ID of the [assistant](https://platform.openai.com/docs/api-reference/assistants) to use to execute this run."),
  "model": z.union([z.string(), z.enum(["gpt-4o","gpt-4o-2024-05-13","gpt-4-turbo","gpt-4-turbo-2024-04-09","gpt-4-0125-preview","gpt-4-turbo-preview","gpt-4-1106-preview","gpt-4-vision-preview","gpt-4","gpt-4-0314","gpt-4-0613","gpt-4-32k","gpt-4-32k-0314","gpt-4-32k-0613","gpt-3.5-turbo","gpt-3.5-turbo-16k","gpt-3.5-turbo-0613","gpt-3.5-turbo-1106","gpt-3.5-turbo-0125","gpt-3.5-turbo-16k-0613"])]).nullable().describe("The ID of the [Model](https://platform.openai.com/docs/api-reference/models) to be used to execute this run. If a value is provided here, it will override the model associated with the assistant. If not, the model associated with the assistant will be used.").optional(),
  "instructions": z.string().nullable().describe("Overrides the [instructions](https://platform.openai.com/docs/api-reference/assistants/createAssistant) of the assistant. This is useful for modifying the behavior on a per-run basis.").optional(),
  "additional_instructions": z.string().nullable().describe("Appends additional instructions at the end of the instructions for the run. This is useful for modifying the behavior on a per-run basis without overriding other instructions.").optional(),
  "additional_messages": z.array(z.object({ "role": z.enum(["user","assistant"]).describe("The role of the entity that is creating the message. Allowed values include:\n- \`user\`: Indicates the message is sent by an actual user and should be used in most cases to represent user-generated messages.\n- \`assistant\`: Indicates the message is generated by the assistant. Use this value to insert messages from the assistant into the conversation.\n"), "content": z.string().describe("The text contents of the message."), "attachments": z.array(z.object({ "file_id": z.string().describe("The ID of the file to attach to the message.").optional(), "tools": z.array(z.object({ "type": z.literal("code_interpreter").describe("The type of tool being defined: \`code_interpreter\`") })).describe("The tools to add this file to.").optional() })).nullable().describe("A list of files attached to the message, and the tools they should be added to.").optional(), "metadata": z.record(z.any()).nullable().describe("Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional() }).strict()).nullable().describe("Adds additional messages to the thread before creating the run.").optional(),
  "tools": z.array(z.object({ "type": z.literal("code_interpreter").describe("The type of tool being defined: \`code_interpreter\`") })).max(20).nullable().describe("Override the tools the assistant can use for this run. This is useful for modifying the behavior on a per-run basis.").optional(),
  "metadata": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nSet of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format. Keys can be a maximum of 64 characters long and values can be a maxium of 512 characters long.\n").optional(),
  "temperature": z.number().gte(0).lte(2).nullable().describe("What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.\n").optional(),
  "top_p": z.number().gte(0).lte(1).nullable().describe("An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n\nWe generally recommend altering this or temperature but not both.\n").optional(),
  "stream": z.boolean().nullable().describe("If \`true\`, returns a stream of events that happen during the Run as server-sent events, terminating when the Run enters a terminal state with a \`data: [DONE]\` message.\n").optional(),
  "max_prompt_tokens": z.number().int().gte(256).nullable().describe("The maximum number of prompt tokens that may be used over the course of the run. The run will make a best effort to use only the number of prompt tokens specified, across multiple turns of the run. If the run exceeds the number of prompt tokens specified, the run will end with status \`incomplete\`. See \`incomplete_details\` for more info.\n").optional(),
  "max_completion_tokens": z.number().int().gte(256).nullable().describe("The maximum number of completion tokens that may be used over the course of the run. The run will make a best effort to use only the number of completion tokens specified, across multiple turns of the run. If the run exceeds the number of completion tokens specified, the run will end with status \`incomplete\`. See \`incomplete_details\` for more info.\n").optional(),
  "truncation_strategy": z.record(z.any()).describe("[EXPANDABLE PARAMETER]:\nControls for how a thread will be truncated prior to the run. Use this to control the intial context window of the run.").optional(),
  "tool_choice": z.enum(["none","auto","required"]).describe("\`none\` means the model will not call any tools and instead generates a message. \`auto\` means the model can pick between generating a message or calling one or more tools. \`required\` means the model must call one or more tools before responding to the user.\n").nullable().describe("Controls which (if any) tool is called by the model.\n\`none\` means the model will not call any tools and instead generates a message.\n\`auto\` is the default value and means the model can pick between generating a message or calling one or more tools.\n\`required\` means the model must call one or more tools before responding to the user.\nSpecifying a particular tool like \`{\"type\": \"file_search\"}\` or \`{\"type\": \"function\", \"function\": {\"name\": \"my_function\"}}\` forces the model to call that tool.\n").optional(),
  "parallel_tool_calls": z.boolean().describe("Whether to enable [parallel function calling](https://platform.openai.com/docs/guides/function-calling/parallel-function-calling) during tool use.").optional(),
  "response_format": z.enum(["none","auto"]).describe("\`auto\` is the default value\n").nullable().describe("Specifies the format that the model must output. Compatible with [GPT-4o](https://platform.openai.com/docs/models/gpt-4o), [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4), and all GPT-3.5 Turbo models since \`gpt-3.5-turbo-1106\`.\n\nSetting to \`{ \"type\": \"json_object\" }\` enables JSON mode, which guarantees the message the model generates is valid JSON.\n\n**Important:** when using JSON mode, you **must** also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly \"stuck\" request. Also note that the message content may be partially cut off if \`finish_reason=\"length\"\`, which indicates the generation exceeded \`max_tokens\` or the conversation exceeded the max context length.\n").optional()
}