> ## Documentation Index > Fetch the complete documentation index at: https://docs.interactly.ai/llms.txt > Use this file to discover all available pages before exploring further. # Test Llm Config Inline > Run a one-shot test invocation against a fully inline LLM configuration that has not been persisted yet (the create flow), and return the model's reply. Because there is no stored record to recover secrets from, the provider api_key must be supplied in ``body.config`` (or omitted to fall back to the team's stored vendor credentials). Provider/runtime failures are returned as HTTP 200 with ``success=False`` and a clean ``error`` message, mirroring the saved-config test endpoint, so the dashboard can render them inline. ## OpenAPI ````yaml https://api-prod.interactly.ai/api-docs/workflows/api.json post /workflows/v1/llm-configs/test openapi: 3.1.0 info: title: Interactly API description: API for building interactly.ai services version: 1.1.0 contact: name: Interactly url: https://interactly.ai email: developers@interactly.ai servers: - url: https://api-prod.interactly.ai description: API Server security: [] paths: /workflows/v1/llm-configs/test: post: tags: - LLM Configs summary: Test Llm Config Inline description: >- Run a one-shot test invocation against a fully inline LLM configuration that has not been persisted yet (the create flow), and return the model's reply. Because there is no stored record to recover secrets from, the provider api_key must be supplied in ``body.config`` (or omitted to fall back to the team's stored vendor credentials). Provider/runtime failures are returned as HTTP 200 with ``success=False`` and a clean ``error`` message, mirroring the saved-config test endpoint, so the dashboard can render them inline. operationId: workflow_service_test_llm_config_inline_v1_llm_configs_test_post requestBody: content: application/json: schema: $ref: '#/components/schemas/LLMConfigTestInlineRequest' required: true responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/LLMConfigTestResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - bearer: [] components: schemas: LLMConfigTestInlineRequest: properties: system_prompt: type: string minLength: 1 title: System Prompt description: Required, non-empty system prompt. messages: items: $ref: '#/components/schemas/LLMTestMessage' type: array title: Messages description: Optional ordered conversation history to send to the model. config: oneOf: - $ref: '#/components/schemas/AzureOpenAILLMConfig-Input' - $ref: '#/components/schemas/OpenAILLMConfig-Input' - $ref: '#/components/schemas/GoogleLLMConfig-Input' - $ref: '#/components/schemas/AnthropicLLMConfig-Input' - $ref: '#/components/schemas/CustomLLMConfig-Input' - $ref: '#/components/schemas/WorkflowDefaultLLMConfig-Input' - $ref: '#/components/schemas/NoLLMConfig-Input' - $ref: '#/components/schemas/LLMGroupConfig-Input' - $ref: '#/components/schemas/LLMGroupWithBackchannelConfig-Input' title: Config description: >- Full config to test inline, before any config record is persisted (create flow). Unlike the saved-config test, there is no stored record to recover secrets from: the provider api_key must be supplied here, or omitted to fall back to the team's stored vendor credentials. discriminator: propertyName: type mapping: anthropic_llm: $ref: '#/components/schemas/AnthropicLLMConfig-Input' azure_openai_llm: $ref: '#/components/schemas/AzureOpenAILLMConfig-Input' custom_llm: $ref: '#/components/schemas/CustomLLMConfig-Input' global_default_llm: $ref: '#/components/schemas/WorkflowDefaultLLMConfig-Input' google_llm: $ref: '#/components/schemas/GoogleLLMConfig-Input' llm_group: $ref: '#/components/schemas/LLMGroupConfig-Input' llm_group_with_backchannel: $ref: '#/components/schemas/LLMGroupWithBackchannelConfig-Input' no_llm: $ref: '#/components/schemas/NoLLMConfig-Input' openai_llm: $ref: '#/components/schemas/OpenAILLMConfig-Input' type: object required: - system_prompt - config title: LLMConfigTestInlineRequest LLMConfigTestResponse: properties: success: type: boolean title: Success response: anyOf: - type: string - type: 'null' title: Response description: Concatenated model output text on success. error: anyOf: - type: string - type: 'null' title: Error description: Clean error message when success is False. provider: anyOf: - type: string - type: 'null' title: Provider model: anyOf: - type: string - type: 'null' title: Model prompt_tokens: anyOf: - type: integer - type: 'null' title: Prompt Tokens completion_tokens: anyOf: - type: integer - type: 'null' title: Completion Tokens total_tokens: anyOf: - type: integer - type: 'null' title: Total Tokens latency_ms: anyOf: - type: integer - type: 'null' title: Latency Ms winning_member: anyOf: - $ref: '#/components/schemas/LLMTestMemberInfo' - type: 'null' description: Which group member produced the main `response`. backchannel_response: anyOf: - type: string - type: 'null' title: Backchannel Response description: Backchannel text emitted before the main response, if any. backchannel_member: anyOf: - $ref: '#/components/schemas/LLMTestMemberInfo' - type: 'null' description: >- Producer of the backchannel response (null for a static backchannel). backchannel_prompt_tokens: anyOf: - type: integer - type: 'null' title: Backchannel Prompt Tokens backchannel_completion_tokens: anyOf: - type: integer - type: 'null' title: Backchannel Completion Tokens backchannel_total_tokens: anyOf: - type: integer - type: 'null' title: Backchannel Total Tokens type: object required: - success title: LLMConfigTestResponse HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError LLMTestMessage: properties: role: type: string enum: - human - ai - system title: Role description: Sender role of the message. default: human content: type: string title: Content description: Text content of the message. type: object required: - content title: LLMTestMessage AzureOpenAILLMConfig-Input: properties: logical_id: anyOf: - type: string - type: 'null' title: LLM Configuration ID description: Unique identifier for the LLM configuration input_field_disallowed: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true provider: anyOf: - $ref: '#/components/schemas/LLMProvider' - type: 'null' title: LLM Provider description: >- The selected Large Language Model provider from the available options. default: azure_openai input_field_not_visible: true streaming: type: boolean title: Enable Streaming description: Whether to enable streaming for the LLM response. default: false field_visibility_level: developer max_retries: anyOf: - type: integer - type: 'null' title: Max Retries description: Maximum number of retries for the LLM request in case of failure. default: 3 field_visibility_level: developer max_parse_retries: anyOf: - type: integer - type: 'null' title: Max Parse Retries description: >- Maximum number of times to re-invoke the LLM when a successful response cannot be parsed into the expected structured (JSON) output. Distinct from `max_retries`, which controls transport-level retries (network errors, rate limits, 5xx) inside the LLM client. default: 3 field_visibility_level: developer max_tokens: anyOf: - type: integer - type: 'null' title: Max Tokens description: Maximum number of tokens to generate in the response. temperature: anyOf: - type: number maximum: 1 minimum: 0 - type: 'null' title: Temperature description: >- Controls the randomness of the model's output. Lower values make the output more deterministic. input_field_type: number_with_slider request_timeout_ms: anyOf: - type: integer - type: 'null' title: Request Timeout (ms) description: Timeout for the request in milliseconds. seed: anyOf: - type: integer - type: 'null' title: Random Seed description: Seed for random number generation to ensure reproducibility. field_visibility_level: developer model_kwargs: additionalProperties: true type: object title: Model Keyword Arguments description: Additional keyword arguments to pass to the LLM provider's API. field_visibility_level: developer api_key: anyOf: - type: string format: password writeOnly: true - type: 'null' title: API Key description: API KEY to access the LLM provider endpoint field_visibility_level: developer input_field_type: password do_not_split_sentences: type: boolean title: Do Not Split Sentences description: Whether to avoid splitting sentences in the LLM response. default: false field_visibility_level: developer truncated_max_recent_messages: anyOf: - type: integer - type: 'null' title: Truncated Max Recent Messages description: >- Maximum number of recent messages to keep when truncating the conversation history. If not set, no truncation is applied. If set, the recent conversation history will be truncated to contain only the specified number of messages when invoking the LLM for next response. If set to 0, only the system prompt is used. model: anyOf: - $ref: '#/components/schemas/AZUREOPENAIModel' - type: 'null' title: Azure OpenAI Model description: Name of the Azure OpenAI model to use default: gpt-5-mini type: type: string const: azure_openai_llm title: Type description: >- Differentiator field that helps in identifying this particular type of config when serializing and deserializing default: azure_openai_llm field_visibility_level: developer input_field_not_visible: true endpoint: anyOf: - type: string - type: 'null' title: Azure OpenAI Endpoint description: >- Endpoint URL for Azure OpenAI API requests. Automatically inferred from env var `AZURE_OPENAI_ENDPOINT` if not provided. field_visibility_level: developer api_version: anyOf: - type: string - type: 'null' title: Azure OpenAI API Version description: >- Azure OpenAI API version. Falls back to `OPENAI_API_VERSION` env var if not provided. field_visibility_level: developer reasoning_effort: anyOf: - type: string - type: 'null' title: Reasoning Effort description: >- Reasoning effort level for GPT-5-* models. Options are 'low', 'medium', 'high'. default: low type: object title: Azure OpenAI OpenAILLMConfig-Input: properties: logical_id: anyOf: - type: string - type: 'null' title: LLM Configuration ID description: Unique identifier for the LLM configuration input_field_disallowed: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true provider: anyOf: - $ref: '#/components/schemas/LLMProvider' - type: 'null' title: LLM Provider description: >- The selected Large Language Model provider from the available options. default: openai input_field_not_visible: true streaming: type: boolean title: Enable Streaming description: Whether to enable streaming for the LLM response. default: false field_visibility_level: developer max_retries: anyOf: - type: integer - type: 'null' title: Max Retries description: Maximum number of retries for the LLM request in case of failure. default: 3 field_visibility_level: developer max_parse_retries: anyOf: - type: integer - type: 'null' title: Max Parse Retries description: >- Maximum number of times to re-invoke the LLM when a successful response cannot be parsed into the expected structured (JSON) output. Distinct from `max_retries`, which controls transport-level retries (network errors, rate limits, 5xx) inside the LLM client. default: 3 field_visibility_level: developer max_tokens: anyOf: - type: integer - type: 'null' title: Max Tokens description: Maximum number of tokens to generate in the response. temperature: anyOf: - type: number maximum: 1 minimum: 0 - type: 'null' title: Temperature description: >- Controls the randomness of the model's output. Lower values make the output more deterministic. input_field_type: number_with_slider request_timeout_ms: anyOf: - type: integer - type: 'null' title: Request Timeout (ms) description: Timeout for the request in milliseconds. seed: anyOf: - type: integer - type: 'null' title: Random Seed description: Seed for random number generation to ensure reproducibility. field_visibility_level: developer model_kwargs: additionalProperties: true type: object title: Model Keyword Arguments description: Additional keyword arguments to pass to the LLM provider's API. field_visibility_level: developer api_key: anyOf: - type: string format: password writeOnly: true - type: 'null' title: API Key description: API KEY to access the LLM provider endpoint field_visibility_level: developer input_field_type: password do_not_split_sentences: type: boolean title: Do Not Split Sentences description: Whether to avoid splitting sentences in the LLM response. default: false field_visibility_level: developer truncated_max_recent_messages: anyOf: - type: integer - type: 'null' title: Truncated Max Recent Messages description: >- Maximum number of recent messages to keep when truncating the conversation history. If not set, no truncation is applied. If set, the recent conversation history will be truncated to contain only the specified number of messages when invoking the LLM for next response. If set to 0, only the system prompt is used. model: anyOf: - $ref: '#/components/schemas/OPENAIModel' - type: 'null' title: OpenAI Model description: Name of the OpenAI model to use default: gpt-5.4-mini type: type: string const: openai_llm title: Type description: >- Differentiator field that helps in identifying this particular type of config when serializing and deserializing default: openai_llm field_visibility_level: developer input_field_not_visible: true base_url: anyOf: - type: string - type: 'null' title: OpenAI Base URL description: >- Base URL path for OpenAI API requests, leave blank if not using a proxy or service emulator. field_visibility_level: developer organization: anyOf: - type: string - type: 'null' title: OpenAI Organization ID description: >- Organization ID for OpenAI API requests, Automatically inferred from env var `OPENAI_ORG_ID` if not provided. field_visibility_level: developer reasoning_effort: anyOf: - type: string - type: 'null' title: Reasoning Effort description: >- Reasoning effort level for GPT-5-* models. Options are 'minimal', 'low', 'medium', 'high'. default: low type: object title: OpenAI GoogleLLMConfig-Input: properties: logical_id: anyOf: - type: string - type: 'null' title: LLM Configuration ID description: Unique identifier for the LLM configuration input_field_disallowed: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true provider: anyOf: - $ref: '#/components/schemas/LLMProvider' - type: 'null' title: LLM Provider description: >- The selected Large Language Model provider from the available options. default: google input_field_not_visible: true streaming: type: boolean title: Enable Streaming description: Whether to enable streaming for the LLM response. default: false field_visibility_level: developer max_retries: anyOf: - type: integer - type: 'null' title: Max Retries description: Maximum number of retries for the LLM request in case of failure. default: 3 field_visibility_level: developer max_parse_retries: anyOf: - type: integer - type: 'null' title: Max Parse Retries description: >- Maximum number of times to re-invoke the LLM when a successful response cannot be parsed into the expected structured (JSON) output. Distinct from `max_retries`, which controls transport-level retries (network errors, rate limits, 5xx) inside the LLM client. default: 3 field_visibility_level: developer max_tokens: anyOf: - type: integer - type: 'null' title: Max Tokens description: Maximum number of tokens to generate in the response. temperature: anyOf: - type: number maximum: 1 minimum: 0 - type: 'null' title: Temperature description: >- Controls the randomness of the model's output. Lower values make the output more deterministic. input_field_type: number_with_slider request_timeout_ms: anyOf: - type: integer - type: 'null' title: Request Timeout (ms) description: Timeout for the request in milliseconds. seed: anyOf: - type: integer - type: 'null' title: Random Seed description: Seed for random number generation to ensure reproducibility. field_visibility_level: developer model_kwargs: additionalProperties: true type: object title: Model Keyword Arguments description: Additional keyword arguments to pass to the LLM provider's API. field_visibility_level: developer api_key: anyOf: - type: string format: password writeOnly: true - type: 'null' title: API Key description: API KEY to access the LLM provider endpoint field_visibility_level: developer input_field_type: password do_not_split_sentences: type: boolean title: Do Not Split Sentences description: Whether to avoid splitting sentences in the LLM response. default: false field_visibility_level: developer truncated_max_recent_messages: anyOf: - type: integer - type: 'null' title: Truncated Max Recent Messages description: >- Maximum number of recent messages to keep when truncating the conversation history. If not set, no truncation is applied. If set, the recent conversation history will be truncated to contain only the specified number of messages when invoking the LLM for next response. If set to 0, only the system prompt is used. model: anyOf: - $ref: '#/components/schemas/GOOGLEModel' - type: 'null' title: Google Model description: Name of the Google model to use default: gemini-3-flash-preview type: type: string const: google_llm title: Type description: >- Differentiator field that helps in identifying this particular type of config when serializing and deserializing default: google_llm field_visibility_level: developer input_field_not_visible: true thinking_budget: anyOf: - type: integer - type: 'null' title: Thinking Budget (tokens) description: Indicates the thinking budget in tokens. By default, it is set to 0. default: 0 type: object title: Google AnthropicLLMConfig-Input: properties: logical_id: anyOf: - type: string - type: 'null' title: LLM Configuration ID description: Unique identifier for the LLM configuration input_field_disallowed: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true provider: anyOf: - $ref: '#/components/schemas/LLMProvider' - type: 'null' title: LLM Provider description: >- The selected Large Language Model provider from the available options. default: anthropic input_field_not_visible: true streaming: type: boolean title: Enable Streaming description: Whether to enable streaming for the LLM response. default: false field_visibility_level: developer max_retries: anyOf: - type: integer - type: 'null' title: Max Retries description: Maximum number of retries for the LLM request in case of failure. default: 3 field_visibility_level: developer max_parse_retries: anyOf: - type: integer - type: 'null' title: Max Parse Retries description: >- Maximum number of times to re-invoke the LLM when a successful response cannot be parsed into the expected structured (JSON) output. Distinct from `max_retries`, which controls transport-level retries (network errors, rate limits, 5xx) inside the LLM client. default: 3 field_visibility_level: developer max_tokens: anyOf: - type: integer - type: 'null' title: Max Tokens description: Maximum number of tokens to generate in the response. temperature: anyOf: - type: number maximum: 1 minimum: 0 - type: 'null' title: Temperature description: >- Controls the randomness of the model's output. Lower values make the output more deterministic. input_field_type: number_with_slider request_timeout_ms: anyOf: - type: integer - type: 'null' title: Request Timeout (ms) description: Timeout for the request in milliseconds. seed: anyOf: - type: integer - type: 'null' title: Random Seed description: Seed for random number generation to ensure reproducibility. field_visibility_level: developer model_kwargs: additionalProperties: true type: object title: Model Keyword Arguments description: Additional keyword arguments to pass to the LLM provider's API. field_visibility_level: developer api_key: anyOf: - type: string format: password writeOnly: true - type: 'null' title: API Key description: API KEY to access the LLM provider endpoint field_visibility_level: developer input_field_type: password do_not_split_sentences: type: boolean title: Do Not Split Sentences description: Whether to avoid splitting sentences in the LLM response. default: false field_visibility_level: developer truncated_max_recent_messages: anyOf: - type: integer - type: 'null' title: Truncated Max Recent Messages description: >- Maximum number of recent messages to keep when truncating the conversation history. If not set, no truncation is applied. If set, the recent conversation history will be truncated to contain only the specified number of messages when invoking the LLM for next response. If set to 0, only the system prompt is used. model: anyOf: - $ref: '#/components/schemas/ANTHROPICModel' - type: 'null' title: Anthropic Model description: Name of the Anthropic Claude model to use default: claude-sonnet-4-6 type: type: string const: anthropic_llm title: Type description: >- Differentiator field that helps in identifying this particular type of config when serializing and deserializing default: anthropic_llm input_field_not_visible: true thinking_budget: anyOf: - type: integer - type: 'null' title: Thinking Budget (tokens) description: >- Token budget for Claude's extended thinking feature. Set to 0 to disable. When enabled, Claude will show its reasoning process before providing a final answer. default: 0 type: object title: Anthropic CustomLLMConfig-Input: properties: logical_id: anyOf: - type: string - type: 'null' title: LLM Configuration ID description: Unique identifier for the LLM configuration input_field_disallowed: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true provider: anyOf: - $ref: '#/components/schemas/LLMProvider' - type: 'null' title: LLM Provider description: >- The selected Large Language Model provider from the available options. default: custom input_field_not_visible: true streaming: type: boolean title: Enable Streaming description: Whether to enable streaming for the LLM response. default: false field_visibility_level: developer max_retries: anyOf: - type: integer - type: 'null' title: Max Retries description: Maximum number of retries for the LLM request in case of failure. default: 3 field_visibility_level: developer max_parse_retries: anyOf: - type: integer - type: 'null' title: Max Parse Retries description: >- Maximum number of times to re-invoke the LLM when a successful response cannot be parsed into the expected structured (JSON) output. Distinct from `max_retries`, which controls transport-level retries (network errors, rate limits, 5xx) inside the LLM client. default: 3 field_visibility_level: developer max_tokens: anyOf: - type: integer - type: 'null' title: Max Tokens description: Maximum number of tokens to generate in the response. temperature: anyOf: - type: number maximum: 1 minimum: 0 - type: 'null' title: Temperature description: >- Controls the randomness of the model's output. Lower values make the output more deterministic. input_field_type: number_with_slider request_timeout_ms: anyOf: - type: integer - type: 'null' title: Request Timeout (ms) description: Timeout for the request in milliseconds. seed: anyOf: - type: integer - type: 'null' title: Random Seed description: Seed for random number generation to ensure reproducibility. field_visibility_level: developer model_kwargs: additionalProperties: true type: object title: Model Keyword Arguments description: Additional keyword arguments to pass to the LLM provider's API. field_visibility_level: developer api_key: anyOf: - type: string format: password writeOnly: true - type: 'null' title: API Key description: API KEY to access the LLM provider endpoint field_visibility_level: developer input_field_type: password do_not_split_sentences: type: boolean title: Do Not Split Sentences description: Whether to avoid splitting sentences in the LLM response. default: false field_visibility_level: developer truncated_max_recent_messages: anyOf: - type: integer - type: 'null' title: Truncated Max Recent Messages description: >- Maximum number of recent messages to keep when truncating the conversation history. If not set, no truncation is applied. If set, the recent conversation history will be truncated to contain only the specified number of messages when invoking the LLM for next response. If set to 0, only the system prompt is used. model: anyOf: - type: string - type: 'null' title: Custom Model Name description: >- Name of the model served behind the custom LLM gateway (free-form string, e.g. 'gpt-4.1-mini', 'llama-3-70b', 'mistral-large'). reasoning_effort: anyOf: - type: string - type: 'null' title: Reasoning Effort description: >- Reasoning effort level for GPT-5-* models. Options are 'minimal', 'low', 'medium', 'high'. type: type: string const: custom_llm title: Type description: >- Differentiator field that helps in identifying this particular type of config when serializing and deserializing default: custom_llm field_visibility_level: developer input_field_not_visible: true base_url: anyOf: - type: string - type: 'null' title: Gateway Base URL description: >- Base URL of the custom OpenAI-compatible LLM gateway (e.g. 'http://api.interactly.ai/workflows/llm-proxy'). default_headers: anyOf: - additionalProperties: true type: object - type: 'null' title: Headers description: >- Optional HTTP headers to include in every request to the gateway (e.g. routing keys, tenant IDs). rewrite_base_url: type: boolean title: Rewrite Base URL description: >- When True, uses an httpx event hook to rewrite every request URL to the exact base_url. Use this for gateways that don't accept the /chat/completions suffix the SDK appends. default: false verify_ssl: type: boolean title: Verify SSL description: >- Whether to verify SSL certificates. Set to False for gateways with self-signed certificates. default: true response_unwrap_key: anyOf: - type: string - type: 'null' title: Response Unwrap Key description: >- When set, the gateway response JSON is expected to wrap the standard OpenAI response inside this key (e.g. 'result'). The runtime will extract the nested object before passing it to the OpenAI SDK. Leave empty for gateways that already return standard format. okta_auth: anyOf: - $ref: '#/components/schemas/OktaAuthConfig' - type: 'null' title: Okta Auth description: Okta authentication configuration for bearer-token injection. use_responses_api: type: boolean title: Use Responses API description: >- Whether to use the GPT 5.x+ style Responses API, instead of Chat Completions API for this Custom LLM. See https://developers.openai.com/api/reference/resources/responses default: false type: object title: Custom LLM WorkflowDefaultLLMConfig-Input: properties: logical_id: anyOf: - type: string - type: 'null' title: LLM Configuration ID description: Unique identifier for the LLM configuration input_field_disallowed: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true provider: anyOf: - $ref: '#/components/schemas/LLMProvider' - type: 'null' title: LLM Provider description: >- The selected Large Language Model provider from the available options. default: default_provider input_field_not_visible: true streaming: type: boolean title: Enable Streaming description: Whether to enable streaming for the LLM response. default: false field_visibility_level: developer max_retries: anyOf: - type: integer - type: 'null' title: Max Retries description: Maximum number of retries for the LLM request in case of failure. default: 3 field_visibility_level: developer max_parse_retries: anyOf: - type: integer - type: 'null' title: Max Parse Retries description: >- Maximum number of times to re-invoke the LLM when a successful response cannot be parsed into the expected structured (JSON) output. Distinct from `max_retries`, which controls transport-level retries (network errors, rate limits, 5xx) inside the LLM client. default: 3 field_visibility_level: developer max_tokens: anyOf: - type: integer - type: 'null' title: Max Tokens description: Maximum number of tokens to generate in the response. temperature: anyOf: - type: number maximum: 1 minimum: 0 - type: 'null' title: Temperature description: >- Controls the randomness of the model's output. Lower values make the output more deterministic. input_field_type: number_with_slider request_timeout_ms: anyOf: - type: integer - type: 'null' title: Request Timeout (ms) description: Timeout for the request in milliseconds. seed: anyOf: - type: integer - type: 'null' title: Random Seed description: Seed for random number generation to ensure reproducibility. field_visibility_level: developer model_kwargs: additionalProperties: true type: object title: Model Keyword Arguments description: Additional keyword arguments to pass to the LLM provider's API. field_visibility_level: developer api_key: anyOf: - type: string format: password writeOnly: true - type: 'null' title: API Key description: API KEY to access the LLM provider endpoint field_visibility_level: developer input_field_type: password do_not_split_sentences: type: boolean title: Do Not Split Sentences description: Whether to avoid splitting sentences in the LLM response. default: false field_visibility_level: developer truncated_max_recent_messages: anyOf: - type: integer - type: 'null' title: Truncated Max Recent Messages description: >- Maximum number of recent messages to keep when truncating the conversation history. If not set, no truncation is applied. If set, the recent conversation history will be truncated to contain only the specified number of messages when invoking the LLM for next response. If set to 0, only the system prompt is used. type: type: string const: global_default_llm title: Type description: >- Differentiator field that helps in identifying this particular type of config when serializing and deserializing default: global_default_llm input_field_not_visible: true type: object title: Workflow Default LLM hidden_in_contexts: - workflow_settings - llm_config hide_all_fields: true NoLLMConfig-Input: properties: logical_id: anyOf: - type: string - type: 'null' title: LLM Configuration ID description: Unique identifier for the LLM configuration input_field_disallowed: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true provider: anyOf: - $ref: '#/components/schemas/LLMProvider' - type: 'null' title: LLM Provider description: >- The selected Large Language Model provider from the available options. default: default_provider input_field_not_visible: true streaming: type: boolean title: Enable Streaming description: Whether to enable streaming for the LLM response. default: false field_visibility_level: developer max_retries: anyOf: - type: integer - type: 'null' title: Max Retries description: Maximum number of retries for the LLM request in case of failure. default: 3 field_visibility_level: developer max_parse_retries: anyOf: - type: integer - type: 'null' title: Max Parse Retries description: >- Maximum number of times to re-invoke the LLM when a successful response cannot be parsed into the expected structured (JSON) output. Distinct from `max_retries`, which controls transport-level retries (network errors, rate limits, 5xx) inside the LLM client. default: 3 field_visibility_level: developer max_tokens: anyOf: - type: integer - type: 'null' title: Max Tokens description: Maximum number of tokens to generate in the response. temperature: anyOf: - type: number maximum: 1 minimum: 0 - type: 'null' title: Temperature description: >- Controls the randomness of the model's output. Lower values make the output more deterministic. input_field_type: number_with_slider request_timeout_ms: anyOf: - type: integer - type: 'null' title: Request Timeout (ms) description: Timeout for the request in milliseconds. seed: anyOf: - type: integer - type: 'null' title: Random Seed description: Seed for random number generation to ensure reproducibility. field_visibility_level: developer model_kwargs: additionalProperties: true type: object title: Model Keyword Arguments description: Additional keyword arguments to pass to the LLM provider's API. field_visibility_level: developer api_key: anyOf: - type: string format: password writeOnly: true - type: 'null' title: API Key description: API KEY to access the LLM provider endpoint field_visibility_level: developer input_field_type: password do_not_split_sentences: type: boolean title: Do Not Split Sentences description: Whether to avoid splitting sentences in the LLM response. default: false field_visibility_level: developer truncated_max_recent_messages: anyOf: - type: integer - type: 'null' title: Truncated Max Recent Messages description: >- Maximum number of recent messages to keep when truncating the conversation history. If not set, no truncation is applied. If set, the recent conversation history will be truncated to contain only the specified number of messages when invoking the LLM for next response. If set to 0, only the system prompt is used. type: type: string const: no_llm title: Type description: >- Differentiator field that helps in identifying this particular type of config when serializing and deserializing default: no_llm input_field_not_visible: true type: object title: No LLM hidden_in_contexts: - node_settings - llm_config hide_all_fields: true LLMGroupConfig-Input: properties: type: type: string const: llm_group title: Type description: Discriminator field which must always be 'llm_group' default: llm_group field_visibility_level: super_admin input_field_not_visible: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true logical_id: anyOf: - type: string - type: 'null' title: LLM Group Logical ID description: Unique identifier for the LLM Group input_field_disallowed: true llms: items: oneOf: - $ref: '#/components/schemas/AzureOpenAILLMConfig-Input' - $ref: '#/components/schemas/OpenAILLMConfig-Input' - $ref: '#/components/schemas/GoogleLLMConfig-Input' - $ref: '#/components/schemas/AnthropicLLMConfig-Input' - $ref: '#/components/schemas/CustomLLMConfig-Input' - $ref: '#/components/schemas/WorkflowDefaultLLMConfig-Input' - $ref: '#/components/schemas/NoLLMConfig-Input' discriminator: propertyName: type mapping: anthropic_llm: $ref: '#/components/schemas/AnthropicLLMConfig-Input' azure_openai_llm: $ref: '#/components/schemas/AzureOpenAILLMConfig-Input' custom_llm: $ref: '#/components/schemas/CustomLLMConfig-Input' global_default_llm: $ref: '#/components/schemas/WorkflowDefaultLLMConfig-Input' google_llm: $ref: '#/components/schemas/GoogleLLMConfig-Input' no_llm: $ref: '#/components/schemas/NoLLMConfig-Input' openai_llm: $ref: '#/components/schemas/OpenAILLMConfig-Input' type: array title: LLM Group description: List of LLM configurations, in preferred order of use. operation_mode: anyOf: - $ref: '#/components/schemas/OperationMode' - type: 'null' title: Operation Mode description: Operation mode for the LLM group default: sequential_with_proactive min_patience_time_ms: anyOf: - type: integer - type: 'null' title: Minimum Patience Time description: >- Grace period in milliseconds until which we will unconditionally wait, unless all LLMs have responded. default: 1500 field_visibility_level: developer max_patience_time_ms: anyOf: - type: integer - type: 'null' title: Maximum Patience Time description: >- Threshold milliseconds after which we will stop waiting. If no LLM has responded by this time, we will return None. field_visibility_level: developer type: object title: Group of LLMs LLMGroupWithBackchannelConfig-Input: properties: type: type: string const: llm_group_with_backchannel title: Type description: >- Discriminator field which must always be 'llm_group_with_backchannel' default: llm_group_with_backchannel field_visibility_level: developer input_field_not_visible: true named_llm_config_id: anyOf: - type: string - type: 'null' title: Named LLM Config ID description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its ID. input_field_not_visible: true named_llm_config_name: anyOf: - type: string - type: 'null' title: Named LLM Config Name description: >- If this configuration was resolved from a reusable named LLM configuration, this holds its name. input_field_not_visible: true id: anyOf: - type: string - type: 'null' title: LLM Group With Backchannel ID description: Unique identifier for the LLM Group input_field_not_visible: true main_llm_config: anyOf: - $ref: '#/components/schemas/LLMGroupConfig-Input' - type: 'null' title: Main LLM Configuration description: Configuration for the main LLM non_backchannel_response_prefix: anyOf: - type: string - type: 'null' title: Non Backchannel Response Prefix description: >- If back channel response is being sent, only the LLM output following this prefix will follow the backchannel response backchannel_llm_config: anyOf: - $ref: '#/components/schemas/LLMGroupConfig-Input' - type: 'null' title: Backchannel LLM Configuration description: Configuration for the back channel LLM backchannel_static_responses: items: type: string type: array title: Backchannel Static Responses description: >- List of static responses from which one will be selected as a backchannel response. backchannel_static_responses_selection_mode: anyOf: - $ref: '#/components/schemas/SelectionMode' - type: 'null' title: Backchannel Static Responses Selection Mode description: Selection mode for backchannel static responses default: random backchannel_min_patience_time_ms: anyOf: - type: integer - type: 'null' title: Backchannel Minimum Patience Time description: >- Grace period in milliseconds until which we will unconditionally wait for the main LLM to respond, unless it has already responded. default: 2000 type: object title: Group of regular and backchanneling LLMs LLMTestMemberInfo: properties: index: anyOf: - type: integer - type: 'null' title: Index description: Position within the group's llms list. logical_id: anyOf: - type: string - type: 'null' title: Logical Id provider: anyOf: - type: string - type: 'null' title: Provider model: anyOf: - type: string - type: 'null' title: Model type: object title: LLMTestMemberInfo description: >- Identity of the group member that produced a given response (null fields for single configs). ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError LLMProvider: type: string enum: - default_provider - azure_openai - openai - google - anthropic - bedrock - custom title: LLMProvider AZUREOPENAIModel: type: string enum: - gpt-5-chat - gpt-5-mini - gpt-5-nano - gpt-4.1-nano - gpt-4.1-mini - gpt-4.1 title: AZUREOPENAIModel OPENAIModel: type: string enum: - gpt-5.4 - gpt-5.4-mini - gpt-5.4-nano - gpt-5.2 - gpt-5.1 - gpt-5-chat-latest - gpt-5.1-chat-latest - gpt-5.2-chat-latest - gpt-5.3-chat-latest - gpt-5-nano - gpt-5-mini - gpt-5 - gpt-4.1-nano - gpt-4.1-mini - gpt-4.1 - gpt-4 - gpt-4o-mini - gpt-4o - gpt-3.5-turbo title: OPENAIModel GOOGLEModel: type: string enum: - gemini-flash-latest - gemini-flash-lite-latest - gemini-3.1-pro-preview - gemini-3.5-flash - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-flash - gemini-2.0-flash - gemini-2.0-flash-lite - gemini-1.5-pro - gemini-1.5-flash - gemini-1.5-flash-8b title: GOOGLEModel ANTHROPICModel: type: string enum: - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-1-20250805 - claude-opus-4-20250514 - claude-sonnet-4-6 - claude-sonnet-4-5-20250929 - claude-sonnet-4-20250514 - claude-haiku-4-5-20251001 title: ANTHROPICModel OktaAuthConfig: properties: integration_id: anyOf: - type: string - type: 'null' title: Okta Integration ID description: >- ID of an Okta integration to use for bearer-token authentication. When set, the runtime fetches a cached OAuth token from this integration and injects it as the Authorization header. Ignored when api_key or an Authorization default_header is already provided. type: object title: OktaAuthConfig description: Configuration for Okta-based bearer-token authentication. OperationMode: type: string enum: - parallel_select_one - sequential_with_proactive title: OperationMode SelectionMode: type: string enum: - random - sequence title: SelectionMode securitySchemes: bearer: scheme: bearer bearerFormat: Bearer type: http description: >- Retrieve your API Key from [Dashboard API Keys Section](https://dashboard.interactly.ai/api-keys). ````