AI EmbeddingStoreRetriever
CertifiedRetrieve context from an embedding store
AI EmbeddingStoreRetriever
Retrieve context from an embedding store
Builds a content retriever over the configured embedding store using a query embedding from embeddingProvider. Results are filtered by maxResults and minScore (0–1). The store is not mutated; ensure the embedding model dimension matches stored vectors.
type: io.kestra.plugin.ai.retriever.EmbeddingStoreRetrieverExamples
Use RAG with AIAgent using an embedding store content retriever. This example ingests documents into a KV embedding store and then uses an AI agent with the EmbeddingStoreRetriever to answer questions grounded in the ingested data.
id: agent_with_rag
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.KestraKVStore
drop: true
fromDocuments:
- content: Paris is the capital of France with a population of over 2.1 million people
- content: The Eiffel Tower is the most famous landmark in Paris at 330 meters tall
- id: agent
type: io.kestra.plugin.ai.agent.AIAgent
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-3.5-flash-lite
apiKey: "{{ secret('GEMINI_API_KEY') }}"
contentRetrievers:
- type: io.kestra.plugin.ai.retriever.EmbeddingStoreRetriever
embeddings:
type: io.kestra.plugin.ai.embeddings.KestraKVStore
embeddingProvider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
maxResults: 3
minScore: 0.0
prompt: What is the capital of France and how many people live there?
Use multiple embedding stores simultaneously. This demonstrates the power of the content retriever approach - you can retrieve from multiple embedding stores and other sources in a single task.
id: multi_store_rag
namespace: company.ai
tasks:
- id: agent
type: io.kestra.plugin.ai.agent.AIAgent
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-3.5-flash-lite
apiKey: "{{ secret('GEMINI_API_KEY') }}"
contentRetrievers:
- type: io.kestra.plugin.ai.retriever.EmbeddingStoreRetriever
embeddings:
type: io.kestra.plugin.ai.embeddings.Pinecone
apiKey: "{{ secret('PINECONE_API_KEY') }}"
index: technical-docs
embeddingProvider:
type: io.kestra.plugin.ai.provider.OpenAI
apiKey: "{{ secret('OPENAI_API_KEY') }}"
modelName: text-embedding-3-small
- type: io.kestra.plugin.ai.retriever.EmbeddingStoreRetriever
embeddings:
type: io.kestra.plugin.ai.embeddings.Qdrant
host: localhost
port: 6333
collectionName: business-docs
embeddingProvider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
- type: io.kestra.plugin.ai.retriever.TavilyWebSearch
apiKey: "{{ secret('TAVILY_API_KEY') }}"
prompt: What are the latest trends in data orchestration?
Properties
embeddingProvider *
Embedding model provider
Model provider used to embed the query before searching the store. It must support embedding generation, and should use the same model that was used at ingestion time. No default: this property is required.
Use Amazon Bedrock models
Invokes Bedrock-hosted chat/embedding models with AWS credentials. Supports standard AWS region/auth settings; ensure model IDs match your Bedrock region and account access.
Chat completion with Amazon Bedrock
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.AmazonBedrock
accessKeyId: "{{ secret('AWS_ACCESS_KEY') }}"
secretAccessKey: "{{ secret('AWS_SECRET_KEY') }}"
modelName: anthropic.claude-3-sonnet-20240229-v1:0
thinkingBudgetTokens: 1024
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
AWS Access Key ID
AWS access key ID used to sign Bedrock requests. Store it as a Kestra secret rather than inline. No default: this property is required.
AWS Secret Access Key
AWS secret access key paired with accessKeyId. Store it as a Kestra secret rather than inline. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
COHERECOHERETITANAmazon Bedrock Embedding Model Type
Family of the Bedrock embedding model, which selects the request/response format used for embeddings. One of COHERE or TITAN. Defaults to COHERE. Ignored for chat and image models.
io.kestra.plugin.ai.provider.AmazonBedrockio.kestra.plugin.langchain4j.provider.AmazonBedrockUse Anthropic Claude models
Provides Claude chat models only; embeddings and images are unsupported. Enforces Anthropic rules (no seed/responseFormat). Thinking mode requires max_tokens > thinking.budget_tokens; set API key and optional base URL.
Chat completion with Anthropic
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.Anthropic
apiKey: "{{ secret('ANTHROPIC_API_KEY') }}"
modelName: claude-3-haiku-20240307
configuration:
thinkingEnabled: true
thinkingBudgetTokens: 1024
returnThinking: false
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
Anthropic API key used to authenticate requests. Store it as a Kestra secret rather than inline. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Maximum Tokens
Maximum number of tokens the model may generate in its response. Not set by default, in which case the Anthropic client default applies. When thinking is enabled, this must be greater than configuration.thinkingBudgetTokens or the task fails.
Model name
io.kestra.plugin.ai.provider.Anthropicio.kestra.plugin.langchain4j.provider.AnthropicUse Azure OpenAI deployments
Targets Azure-hosted OpenAI models via the resource endpoint and deployment name. Supports API key or AAD client credentials; set apiVersion when required by the deployment.
Chat completion with Azure OpenAI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.AzureOpenAI
apiKey: "{{ secret('AZURE_API_KEY') }}"
endpoint: https://your-resource.openai.azure.com/
modelName: gpt-4o-mini
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API endpoint
Azure OpenAI resource endpoint, in the form https://{resource}.openai.azure.com/. No default: this property is required.
API Key
Azure OpenAI API key. Provide either this key or the tenantId/clientId/clientSecret trio for Entra ID authentication; the API key takes precedence when both are set. Store it as a Kestra secret rather than inline.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client ID
Microsoft Entra ID application (client) ID used for service-principal authentication. Required together with tenantId and clientSecret when apiKey is not set.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Client secret
Microsoft Entra ID application client secret used for service-principal authentication. Required together with tenantId and clientId when apiKey is not set. Store it as a Kestra secret rather than inline.
Model name
API version
Azure OpenAI REST API version to call. Not set by default, in which case the Azure SDK's latest supported version is used.
Tenant ID
Microsoft Entra ID tenant used for service-principal authentication. Required together with clientId and clientSecret when apiKey is not set.
io.kestra.plugin.ai.provider.AzureOpenAIio.kestra.plugin.langchain4j.provider.AzureOpenAIUse Cohere models
Connects to the Cohere OpenAI-compatible Compatibility API with API key and model name. Supports chat completion (including tool calling) and embeddings. Image generation is not available, and structured output follows Cohere's json_object format rather than OpenAI's strict JSON schema mode.
Chat completion with Cohere
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.Cohere
apiKey: "{{ secret('COHERE_API_KEY') }}"
modelName: command-a-03-2025
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
API key used to authenticate against the provider. Store it as a Kestra secret rather than inline. No default: this property is required.
https://api.cohere.ai/compatibility/v1API base URL
Base URL of the Cohere OpenAI-compatible API. Defaults to https://api.cohere.ai/compatibility/v1.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
Use DashScope (Qwen) models
Calls Alibaba Cloud DashScope for Qwen chat/embeddings/images with API key. Some params (timeouts, retries, stop, maxTokens) map directly to DashScope limits.
Chat completion with DashScope (Qwen)
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.DashScope
apiKey: "{{ secret('DASHSCOPE_API_KEY') }}"
modelName: qwen-plus
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
Alibaba Cloud DashScope API key used to authenticate requests. Store it as a Kestra secret rather than inline. No default: this property is required.
https://dashscope-intl.aliyuncs.com/api/v1API base URL
Base URL of the DashScope API. Use https://dashscope.aliyuncs.com/api/v1 for the China (Beijing) region and https://dashscope-intl.aliyuncs.com/api/v1 for the Singapore region. Defaults to the region inferred from the worker's system timezone.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Enable Internet search
If true, the model may use Internet search results as reference when generating text. Defaults to false.
Maximum output tokens
Maximum number of tokens returned by a single request. Not set by default, in which case the DashScope default applies.
Model name
Repetition penalty
Penalty applied to repeated sequences during generation. Higher values reduce repetition; 1.0 means no penalty. Valid range is (0, +inf). Not set by default, in which case the DashScope default applies.
Use DeepSeek models
Connects to DeepSeek’s OpenAI-compatible endpoint with API key and model name for chat/embedding tasks.
Chat completion with DeepSeek
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.DeepSeek
apiKey: "{{ secret('DEEPSEEK_API_KEY') }}"
modelName: deepseek-chat
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
API key used to authenticate against the provider. Store it as a Kestra secret rather than inline. No default: this property is required.
https://api.deepseek.com/v1API base URL
Base URL of the DeepSeek OpenAI-compatible API. Defaults to https://api.deepseek.com/v1.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
io.kestra.plugin.ai.provider.DeepSeekio.kestra.plugin.langchain4j.provider.DeepSeekUse Docker Model Runner
Routes inference to a locally running Docker Model Runner instance via its OpenAI-compatible REST API.
Docker Model Runner is built into Docker Desktop and Docker Engine (Linux) and requires no separate setup.
It exposes an OpenAI-compatible API and does not require authentication — set apiKey to any non-empty value (the default not-needed works).
Base URL variants — pick the one matching where Kestra itself runs:
- Kestra in a container on Docker Desktop:
http://model-runner.docker.internal/engines/v1 - Kestra in a container on Docker Engine (Linux):
http://172.17.0.1: 12434/engines/v1 - Kestra directly on the host (default):
http://localhost: 12434/engines/v1
The default suits a host installation. Most deployments run Kestra in a bridge-networked container, where localhost is the Kestra container itself rather than the Docker Model Runner host — set baseUrl explicitly in that case.
Image generation routes to the Diffusers endpoint (/engines/diffusers/v1) automatically; use a diffuser-capable model such as ai/stable-diffusion. Docker Model Runner does not advertise which models are diffuser-capable and does not reject a chat model, so passing one makes the request hang until it times out. The first image generation also downloads the Diffusers backend, which can take several minutes.
Pair this provider with io.kestra.plugin.docker.model.Pull (plugin-docker) to manage model lifecycle in the same flow.
Chat completion with Docker Model Runner
id: docker_model_chat
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: pull_model
type: io.kestra.plugin.docker.model.Pull
model: ai/smollm2
- id: ask
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.DockerModel
modelName: ai/smollm2
messages:
- type: USER
content: "{{ inputs.prompt }}"
Chat completion (container-internal base URL)
id: docker_model_chat_container
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: ask
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.DockerModel
modelName: ai/smollm2
baseUrl: http://model-runner.docker.internal/engines/v1
messages:
- type: USER
content: "{{ inputs.prompt }}"
not-neededAPI Key
Placeholder credential: Docker Model Runner requires no authentication and accepts any non-empty value. Defaults to not-needed.
http://localhost:12434/engines/v1API base URL
Base URL of the Docker Model Runner OpenAI-compatible API. Pick the variant matching where Kestra itself runs: http://model-runner.docker.internal/engines/v1 for Kestra in a container on Docker Desktop, http://172.17.0.1: 12434/engines/v1 for Kestra in a container on Docker Engine (Linux), and the default http://localhost: 12434/engines/v1 for Kestra directly on the host. The model-runner.docker.internal alias exists only inside containers on Docker Desktop.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
Configure ordered model providers for fallback
Tries chat providers in configured order, falling back on connection failures, timeouts, and retriable provider errors such as HTTP 5xx and 429. Each provider attempt receives the full configured timeout independently, so multiple timeouts can make total duration approach N times that timeout, plus overhead; this is not guaranteed because providers differ in how they apply timeouts. It does not fall back on authentication, invalid request, guardrail, tool execution, or JSON/response-processing failures. Image and embedding operations are unsupported.
AI agent with ordered provider fallback
id: ai_agent_with_fallback
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: ai_agent
type: io.kestra.plugin.ai.agent.AIAgent
provider:
type: io.kestra.plugin.ai.provider.Fallback
providers:
- type: io.kestra.plugin.ai.provider.OpenAI
modelName: gpt-5-mini
apiKey: "{{ secret('OPENAI_API_KEY') }}"
baseUrl: https://api.openai.com/v1
- type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-2.5-flash
apiKey: "{{ secret('GEMINI_API_KEY') }}"
systemMessage: You are a helpful assistant. Answer clearly and concisely.
prompt: "{{ inputs.prompt }}"
Providers
Model providers in the order they should be considered for fallback.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
Use GitHub Models via Azure AI Inference
Calls GitHub Models through the Azure AI Inference API with a GitHub token. Supports chat and embeddings; response format options map to Azure AI Inference capabilities.
Chat completion with GitHub Models
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.GitHubModels
gitHubToken: "{{ secret('GITHUB_TOKEN') }}"
modelName: gpt-4o-mini
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely.
- type: USER
content: "{{ inputs.prompt }}"
GitHub Token
GitHub Personal Access Token (PAT) used to access GitHub Models. Store it as a Kestra secret rather than inline. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
Use Google Gemini models
Supports Gemini chat and embeddings (image generation is not currently supported). Tools do not support JSON Schema anyOf, and tools cannot be combined with responseFormat; configure either but not both.
Thinking models (e.g. gemini-3.5-flash) attach a thought_signature to every function-call part. This provider automatically captures those signatures (returnThinking defaults to true) and re-attaches them to the conversation history for every follow-up request (sendThinking is always enabled), preventing the 400 INVALID_ARGUMENT – Function call is missing a thought_signature error.
In addition, on Gemini 2.x models thinking is disabled by default (thinkingBudget = 0) to reduce token usage, unless thinkingEnabled: true or thinkingBudgetTokens > 0 is explicitly set. Gemini 3 models cannot have thinking turned off, so no thinking budget is sent for them by default and the model applies its own; set thinkingBudgetTokens to cap it.
Chat completion with Google Gemini
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
apiKey: "{{ secret('GOOGLE_API_KEY') }}"
modelName: gemini-3.5-flash-lite
configuration:
thinkingEnabled: true
thinkingBudgetTokens: 1024
returnThinking: true
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
Chat completion with Google Gemini with a local base URL + PEM certificates
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-3.5-flash-lite
clientPem: "{{ secret('CLIENT_PEM') }}"
caPem: "{{ secret('CA_PEM') }}"
baseUrl: "https://internal.gemini.company.com/endpoint"
configuration:
thinkingEnabled: true
thinkingBudgetTokens: 1024
returnThinking: true
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
Google AI Studio API key used to authenticate requests. Store it as a Kestra secret rather than inline. Required unless certificate-based authentication is configured through clientPem, optionally with caPem.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Embedding model configuration
Settings applied when this provider is used to generate embeddings rather than chat completions. Not set by default, in which case the Gemini client defaults apply.
io.kestra.plugin.ai.provider.GoogleGemini-EmbeddingModelConfiguration
Maximum retries
Number of times a failed embedding request is retried before the task fails. Not set by default, in which case the Gemini client default applies.
Output embedding size
Length the embedding vectors are truncated to, which trades a little accuracy for smaller storage. It must match the dimensionality already used in the embedding store. Not set by default (the model's full dimensionality).
RETRIEVAL_QUERYRETRIEVAL_DOCUMENTSEMANTIC_SIMILARITYCLASSIFICATIONCLUSTERINGQUESTION_ANSWERINGFACT_VERIFICATIONEmbedding task type
Downstream use the embeddings are optimized for, such as RETRIEVAL_DOCUMENT, RETRIEVAL_QUERY, SEMANTIC_SIMILARITY or CLASSIFICATION. Use the matching pair at ingestion and query time. Not set by default, in which case the Gemini default applies.
Request timeout
Maximum time to wait for each embedding request. Not set by default, in which case the Gemini client default applies.
Document title metadata key
Metadata key whose value is passed to the model as the document's title, which improves retrieval quality by giving the document context. Not set by default.
Model name
io.kestra.plugin.ai.provider.GoogleGeminiio.kestra.plugin.langchain4j.provider.GoogleGeminiUse Google Vertex AI models
Calls Vertex AI Gemini chat, embeddings, or images using project, location, and endpoint settings. Authenticates like the GCP plugin: serviceAccount when set, otherwise Application Default Credentials, optionally impersonating impersonatedServiceAccount; ensure response formats are supported by the selected model/region. Tool calls work on Gemini 3 models: the thought_signature returned with each function call is sent back automatically on the following turns. Thinking is left to the model default unless thinkingEnabled or thinkingBudgetTokens is set.
Chat completion with Google Vertex AI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.GoogleVertexAI
modelName: gemini-3.5-flash-lite
location: your-google-cloud-region
projectId: your-google-cloud-project-id
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
Chat completion with Google Vertex AI using a service account
id: chat_completion_service_account
namespace: company.ai
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.GoogleVertexAI
modelName: gemini-3.5-flash-lite
location: your-google-cloud-region
projectId: your-google-cloud-project-id
serviceAccount: "{{ secret('GCP_SERVICE_ACCOUNT_JSON') }}"
messages:
- type: USER
content: What is the capital of France?
Project location
Google Cloud region hosting the Vertex AI model, or global to use the global endpoint aiplatform.googleapis.com. No default: this property is required for chat models.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Endpoint URL
Vertex AI API endpoint for image and embedding models, as host: port or a full https:// URL. Not set by default, in which case the endpoint is derived from location (aiplatform.googleapis.com for global). Must not be set for chat models, which always use Gemini.
Service account to impersonate
Email of the service account to impersonate. Not supported for image models.
Model name
Project ID (deprecated)
Deprecated, use projectId instead. Only used when projectId is not set.
Project ID
Google Cloud project ID that owns the Vertex AI resources. When not set, it is taken from the service account credentials.
["https://www.googleapis.com/auth/cloud-platform"]OAuth scopes
OAuth scopes requested for the credentials. Defaults to https://www.googleapis.com/auth/cloud-platform. Not supported for image models, which always use Application Default Credentials with their default scopes.
Service account JSON key
Content of a Google Cloud service account JSON key. When not set, Application Default Credentials are used. Not supported for image models, which always use Application Default Credentials.
io.kestra.plugin.ai.provider.GoogleVertexAIio.kestra.plugin.langchain4j.provider.GoogleVertexAIUse Hugging Face Inference endpoints
Routes requests to Hugging Face Inference Endpoints via the OpenAI-compatible gateway (default router.huggingface.co). Requires an API token and deployment model name.
Chat completion with HuggingFace
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.HuggingFace
apiKey: "{{ secret('HUGGING_FACE_API_KEY') }}"
modelName: HuggingFaceTB/SmolLM3-3B:hf-inference
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
API key used to authenticate against the provider. Store it as a Kestra secret rather than inline. No default: this property is required.
https://router.huggingface.co/v1API base URL
Base URL of the Hugging Face router's OpenAI-compatible API. Defaults to https://router.huggingface.co/v1.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
Use Langdock models
Connects to Langdock's Completion API, which exposes OpenAI/Azure OpenAI-backed models on the OPENAI route and Claude models on the ANTHROPIC route. Set modelFamily to match the modelName you use: Claude models are only reachable when modelFamily is ANTHROPIC. Use the langdock.ListModels task with the matching family to discover valid modelName values.
Embeddings always go through the OpenAI route (only text-embedding-ada-002 is supported there). If your key is refused for embeddings, use a workspace API key with the Embedding API scope. Image generation is not offered by the Completion API.
For a dedicated deployment, set baseUrl to the route root that matches the operation you are using this provider for, e.g. https://acme.langdock.com/api/public/openai/eu/v1 for chat/embeddings on the OpenAI route, or https://acme.langdock.com/api/public/anthropic/eu/v1/ for chat on the Anthropic route; it then takes precedence over region.
Chat completion with a Langdock-hosted OpenAI model
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.Langdock
apiKey: "{{ secret('LANGDOCK_API_KEY') }}"
modelFamily: OPENAI
modelName: gpt-5.4-mini
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
AI agent using a Langdock-hosted Claude model with a Kestra tool
id: agent_with_tool
namespace: company.ai
inputs:
- id: prompt
type: STRING
defaults: "Log the message 'Hello from Langdock!'"
tasks:
- id: agent
type: io.kestra.plugin.ai.agent.AIAgent
provider:
type: io.kestra.plugin.ai.provider.Langdock
apiKey: "{{ secret('LANGDOCK_API_KEY') }}"
modelFamily: ANTHROPIC
modelName: claude-sonnet-4-6-default
prompt: "{{ inputs.prompt }}"
tools:
- type: io.kestra.plugin.ai.tool.KestraTask
tasks:
- id: log
type: io.kestra.plugin.core.log.Log
message: "..."
Ingest documents into a KV embedding store using Langdock embeddings
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.Langdock
apiKey: "{{ secret('LANGDOCK_WORKSPACE_API_KEY') }}"
modelName: text-embedding-ada-002
embeddings:
type: io.kestra.plugin.ai.embeddings.KestraKVStore
drop: true
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/README.md
API Key
Langdock API key used to authenticate requests. Store it as a Kestra secret rather than inline. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
OPENAIOPENAIANTHROPICModel family
Langdock Completion API route serving the request: OPENAI for OpenAI and Azure OpenAI-backed models, or ANTHROPIC, which is required to reach Claude models. Defaults to OPENAI. Ignored for embeddings, which always use the OpenAI route.
Model name
EUEUUSRegion
Langdock region that serves the request: EU or US. Defaults to EU. Ignored when baseUrl points at a dedicated deployment.
Use LocalAI OpenAI-compatible server
Targets a self-hosted LocalAI instance via its OpenAI-compatible API for chat/embeddings/images. Set baseUrl if your server is not on the default.
Chat completion with LocalAI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.LocalAI
modelName: gemma-3-1b-it
baseUrl: http://localhost:8080/v1
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API base URL
Base URL of the LocalAI server's OpenAI-compatible API. No default: this property is required.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
io.kestra.plugin.ai.provider.LocalAIio.kestra.plugin.langchain4j.provider.LocalAIUse Mistral models
Calls Mistral chat/embedding APIs with an API key. topK is not supported; chat configuration must respect model limits.
Chat completion with Mistral AI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.MistralAI
apiKey: "{{ secret('MISTRAL_API_KEY') }}"
modelName: mistral-small-latest
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
Mistral AI API key used to authenticate requests. Store it as a Kestra secret rather than inline. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
io.kestra.plugin.ai.provider.MistralAIio.kestra.plugin.langchain4j.provider.MistralAIUse OCI Generative AI models
Calls Oracle Cloud GenAI chat/embedding models with compartment OCID and region. Auth is handled via OCI SDK provider (config file or instance principals). Ensure the model is permitted in the chosen compartment.
Chat completion with OciGenAI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.OciGenAI
region: "{{ secret('OCI_GENAI_MODEL_REGION_PROPERTY') }}"
compartmentId: "{{ secret('OCI_GENAI_COMPARTMENT_ID_PROPERTY') }}"
authProvider: "{{ secret('OCI_GENAI_CONFIG_PROFILE_PROPERTY') }}"
modelName: cohere.command-r-plus
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
Compartment OCID
OCID of the OCI compartment holding the generative AI model. No default: this property is required.
OCI region
OCI region the client connects to, which must offer the OCI Generative AI service. No default: this property is required.
OCI config profile name
Name of the profile in your OCI config file used to authenticate the SDK client. Defaults to DEFAULT.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
Use local Ollama models
Calls an Ollama server for chat/embeddings using the given endpoint and model name. Ideal for self-hosted/local models; ensure the Ollama daemon is reachable.
Chat completion with Ollama
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.Ollama
modelName: llama3
endpoint: http://localhost:11434
configuration:
thinkingEnabled: true
returnThinking: true
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
Model endpoint
Base URL of the Ollama server exposing the model. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
io.kestra.plugin.ai.provider.Ollamaio.kestra.plugin.langchain4j.provider.OllamaCompute embeddings in-process with an ONNX model
Runs a BERT-style sentence-embedding model exported to ONNX (for example all-MiniLM, BGE or E5) inside the Kestra worker with ONNX Runtime on CPU: no API key and no model server. Provide the .onnx model file and its Hugging Face tokenizer.json; modelName is only used as a label, and baseUrl, clientPem and caPem are ignored since nothing is called over the network. Embeddings only: chat completion and image generation are not supported. A loaded model stays in the worker's memory and is reused by every task that provides the same model and tokenizer files; it is unloaded after about 10 minutes without use. Each worker keeps at most 2 models loaded between embedding calls (each distinct model, tokenizer and poolingMode combination counts as one): loading another one unloads the least recently used model as soon as no running call uses it, and that model is loaded again on its next use. Administrators can change the limit with the max-loaded-models plugin configuration of io.kestra.plugin.ai.provider.Onnx; set it to at least the number of distinct models used at the same time, otherwise they keep unloading and reloading each other. Models are loaded and unloaded on a single worker thread so that reloading them reuses the same native memory: a worker loads one model at a time, so loading a large model delays the first use of another model, while calls using already loaded models are not affected. On glibc-based workers that also run other plugins allocating native memory from many threads, setting the MALLOC_ARENA_MAX=2 environment variable on the worker further limits native memory growth.
Ingest and search documents with all-MiniLM-L6-v2 downloaded from Hugging Face
id: onnx_rag
namespace: company.ai
tasks:
- id: model
type: io.kestra.plugin.core.http.Download
uri: https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2/resolve/main/onnx/model.onnx
- id: tokenizer
type: io.kestra.plugin.core.http.Download
uri: https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2/resolve/main/tokenizer.json
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.Onnx
modelName: all-MiniLM-L6-v2
modelUri: "{{ outputs.model.uri }}"
tokenizerUri: "{{ outputs.tokenizer.uri }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.KestraKVStore
drop: true
fromDocuments:
- content: Kestra is an open-source orchestration platform for data and AI workflows.
- content: PostgreSQL is a relational database.
- id: search
type: io.kestra.plugin.ai.rag.Search
provider:
type: io.kestra.plugin.ai.provider.Onnx
modelName: all-MiniLM-L6-v2
modelUri: "{{ outputs.model.uri }}"
tokenizerUri: "{{ outputs.tokenizer.uri }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.KestraKVStore
query: Which tool orchestrates workflows?
maxResults: 1
minScore: 0.3
fetchType: FETCH
Use a BGE model stored as namespace files
id: onnx_bge_search
namespace: company.ai
inputs:
- id: query
type: STRING
tasks:
- id: search
type: io.kestra.plugin.ai.rag.Search
provider:
type: io.kestra.plugin.ai.provider.Onnx
modelName: bge-small-en-v1.5
modelUri: nsfile:///models/bge-small-en-v1.5/model.onnx
tokenizerUri: nsfile:///models/bge-small-en-v1.5/tokenizer.json
poolingMode: CLS
embeddings:
type: io.kestra.plugin.ai.embeddings.KestraKVStore
query: "{{ inputs.query }}"
maxResults: 3
minScore: 0.5
fetchType: FETCH
ONNX model file
URI of the .onnx model file: a namespace file (nsfile:///path), an internal storage file (kestra:///..., e.g. the output of io.kestra.plugin.core.http.Download), or a local file (file:///path, must be allowed by kestra.local-files.allowed-paths).
Tokenizer file
URI of the Hugging Face tokenizer.json published with the model, using the same schemes as modelUri. A tokenizer from another model fails, or silently produces wrong embeddings when its token ids happen to fit the model's vocabulary.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
MEANCLSMEANPooling mode
How token vectors are combined into one embedding. MEAN suits sentence-transformers models such as all-MiniLM and E5; BGE models expect CLS. A mismatch does not fail but degrades search quality. Defaults to MEAN.
Use OpenAI models
Connects to OpenAI-compatible endpoints (defaults to api.openai.com) for chat, embeddings, or images. Requires API key; override baseUrl for Azure-compatible or proxy setups.
Chat completion with OpenAI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.OpenAI
apiKey: "{{ secret('OPENAI_API_KEY') }}"
modelName: gpt-5-mini
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
API key used to authenticate against the provider. Store it as a Kestra secret rather than inline. No default: this property is required.
https://api.openai.com/v1API base URL
Base URL of the OpenAI-compatible API. Override it to target Azure OpenAI, a proxy, or a self-hosted gateway. Defaults to https://api.openai.com/v1.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
io.kestra.plugin.ai.provider.OpenAIio.kestra.plugin.langchain4j.provider.OpenAIUse OpenRouter models
Routes requests through OpenRouter’s multi-model API using your API key. Supports chat/embedding/image models exposed by OpenRouter; honor provider-specific safety/usage limits.
Chat completion with OpenRouter
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.OpenRouter
apiKey: "{{ secret('OPENROUTER_API_KEY') }}"
baseUrl: https://openrouter.ai/api/v1
modelName: openai/gpt-4o-mini
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
OpenRouter API key used to authenticate requests. Store it as a Kestra secret rather than inline. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
io.kestra.plugin.ai.provider.OpenRouterio.kestra.plugin.langchain4j.provider.OpenRouterUse IBM watsonx.ai models
Calls IBM watsonx.ai chat/embedding endpoints with API key and project ID. Ensure the selected model ID is available in the configured project.
Chat completion with Watsonx AI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.WatsonxAI
apiKey: "{{ secret('WATSONX_API_KEY') }}"
projectId: "{{ secret('WATSONX_PROJECT_ID') }}"
modelName: ibm/granite-3-3-8b-instruct
baseUrl : "https://api.eu-de.dataplatform.cloud.ibm.com/wx"
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
IBM Cloud API key used to authenticate against watsonx.ai. Store it as a Kestra secret rather than inline. No default: this property is required.
Project ID
Identifier of the watsonx.ai project the model runs under. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
Use Cloudflare Workers AI models
Invokes Workers AI chat, embedding, and image models using account ID and API key. Ensure the selected model is available in your account/region.
Chat completion with WorkersAI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.WorkersAI
accountId: "{{ secret('WORKERS_AI_ACCOUNT_ID') }}"
apiKey: "{{ secret('WORKERS_AI_API_KEY') }}"
modelName: "@cf/meta/llama-2-7b-chat-fp16"
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
Account Identifier
Cloudflare account ID that owns the Workers AI deployment. No default: this property is required.
API Key
Cloudflare API token with Workers AI access. Store it as a Kestra secret rather than inline. No default: this property is required.
Base URL
Custom base URL overriding the provider's default endpoint. Useful for enterprise gateways, proxies, self-hosted deployments, or test doubles such as WireMock. Defaults to the provider's public endpoint.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Model name
io.kestra.plugin.ai.provider.WorkersAIio.kestra.plugin.langchain4j.provider.WorkersAIUse ZhiPu AI models
Calls ZhiPu’s OpenAI-compatible chat/embedding/image APIs with API key and model name. Supports stop tokens, retry count, and max tokens per request.
Chat completion with ZhiPu AI
id: chat_completion
namespace: company.ai
inputs:
- id: prompt
type: STRING
tasks:
- id: chat_completion
type: io.kestra.plugin.ai.completion.ChatCompletion
provider:
type: io.kestra.plugin.ai.provider.ZhiPuAI
apiKey: "{{ secret('ZHIPU_API_KEY') }}"
modelName: glm-4.5-flash
messages:
- type: SYSTEM
content: You are a helpful assistant, answer concisely, avoid overly casual language or unnecessary verbosity.
- type: USER
content: "{{ inputs.prompt }}"
API Key
ZhiPu AI API key used to authenticate requests. Store it as a Kestra secret rather than inline. No default: this property is required.
https://open.bigmodel.cn/API base URL
Base URL of the ZhiPu AI API. Defaults to https://open.bigmodel.cn/.
CA PEM certificate content
PEM-encoded certificate authority chain, as text, used to verify the TLS certificate presented by a custom endpoint. Not set by default, in which case the JVM's default trust store is used.
Client PEM certificate content
PEM-encoded client certificate and private key, as text, used for mutual-TLS authentication against enterprise AI endpoints. Not set by default, in which case the default HTTP client is used.
Maximum retries
Number of times a failed request is retried before the task fails. Defaults to 3.
Maximum output tokens
Maximum number of tokens returned by a single request. Defaults to 512.
Model name
Stop sequences
Strings that stop generation as soon as the model is about to produce one of them. The stop sequence itself is not included in the output. Not set by default (no stop sequence).
embeddings *
Embedding store
Embedding store queried for content relevant to the user's question. No default: this property is required.
Store embeddings in Chroma
Connects to a Chroma HTTP instance using cosine distance; targets the given collection and drops it when drop=true.
Ingest documents into a Chroma embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.Chroma
baseUrl: http://localhost:8000
collectionName: embeddings
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
Database base URL
Base URL of the Chroma server. No default: this property is required.
Collection name
Chroma collection that holds the embeddings. No default: this property is required.
io.kestra.plugin.ai.embeddings.Chromaio.kestra.plugin.langchain4j.embeddings.ChromaStore embeddings in Elasticsearch
Targets an Elasticsearch 8.15+ cluster using the provided hosts/index; when drop=true the index is deleted. Supports basic auth, custom headers, path prefix, and trust-all TLS for self-signed certs.
Ingest documents into an Elasticsearch embedding store (requires Elasticsearch 8.15+)
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.Elasticsearch
connection:
hosts:
- http://localhost:9200
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
Connection
Elasticsearch connection settings: hosts, authentication, and TLS options. No default: this property is required.
io.kestra.plugin.ai.embeddings.Elasticsearch-ElasticsearchConnection
1Elasticsearch HTTP servers
URLs of the Elasticsearch nodes to connect to, each including scheme, host and port. No default: this property is required and must not be empty.
Basic authorization
Username and password used for HTTP basic authentication. Not set by default (anonymous access).
io.kestra.plugin.ai.embeddings.Elasticsearch-ElasticsearchConnection-BasicAuth
Basic authorization password
Password of the Elasticsearch user. Store it as a Kestra secret rather than inline. No default: this property is required inside basicAuth.
Basic authorization username
User authenticating against Elasticsearch. No default: this property is required inside basicAuth.
HTTP headers sent with every request
Extra HTTP headers added to each request, each written as a key: value string. Not set by default.
Path prefix for all HTTP requests
Prefix prepended to every request path, so /my/path turns each call into /my/path/ + endpoint. Use it only when Elasticsearch sits behind a proxy that serves it under a base path. Not set by default.
Strict deprecation mode
If true, responses carrying deprecation warnings are treated as failures. Defaults to false.
8Target Elasticsearch server major version
Major version advertised in the compatible-with media-type headers (Accept and Content-Type). The bundled elasticsearch-java 9.x client would otherwise negotiate compatible-with=9, which Elasticsearch 8 rejects. Use 8 for an Elasticsearch 8 cluster or 9 for Elasticsearch 9. Defaults to 8.
Trust all SSL CA certificates
If true, accept any TLS certificate presented by the server, which is sometimes needed for self-signed certificates. Defaults to false. WARNING: enabling this disables both certificate chain validation and hostname verification, exposing connections to man-in-the-middle attacks. Prefer supplying a custom CA certificate instead, and use this only in trusted, controlled environments.
Index name
Elasticsearch index that stores the embeddings. No default: this property is required.
io.kestra.plugin.ai.embeddings.Elasticsearchio.kestra.plugin.langchain4j.embeddings.ElasticsearchPrototype embeddings in Kestra KV
Stores embeddings in-memory and persists them to Kestra namespace KV on completion. Suitable for small demos; not scalable. drop=true discards any previously saved KV snapshot.
Ingest documents into a KV embedding store.\nWARNING: the KestraKVStore embeddings are for quick prototyping only; since they are stored in a KV store and loaded into memory, this won't scale with large numbers of documents.
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.KestraKVStore
drop: true
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
{{ flow.id }}-embedding-storeKV pair name
Key of the Kestra KV pair used to persist the in-memory embedding store between task runs. Defaults to {{ flow.id }}-embedding-store.
io.kestra.plugin.ai.embeddings.KestraKVStoreio.kestra.plugin.langchain4j.embeddings.KestraKVStoreStore embeddings in MariaDB
Persists embeddings to a MariaDB table; create/drop behavior is controlled by createTable and drop. Metadata storage mode defaults to COLUMN_PER_KEY (which requires column/index definitions); set COMBINED_JSON to store metadata as a JSON object in a single column. Requires valid JDBC URL and credentials.
Ingest documents into a MariaDB embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.MariaDB
username: "{{ secret('MARIADB_USERNAME') }}"
password: "{{ secret('MARIADB_PASSWORD') }}"
databaseUrl: "{{ secret('MARIADB_DATABASE_URL') }}"
tableName: embeddings
fieldName: id
createTable: true
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
Create table if missing
If true, the embeddings table is created when it does not already exist. Defaults to false.
Database URL
JDBC URL of the MariaDB database holding the embeddings. No default: this property is required.
ID column name
Name of the column used as the unique identifier of each embedding. Defaults to an empty value, which lets the MariaDB store use its own column name.
Password
Password of the database user. Store it as a Kestra secret rather than inline. No default: this property is required.
Table name
Name of the table where embeddings are stored. No default: this property is required.
Username
User connecting to the MariaDB database. No default: this property is required.
Metadata column definitions
SQL column definitions for metadata fields, one per metadata key. Required only when metadataStorageMode is COLUMN_PER_KEY. Not set by default.
Metadata index definitions
SQL index definitions for the metadata columns. Used only when metadataStorageMode is COLUMN_PER_KEY. Not set by default (no extra index).
COLUMN_PER_KEYMetadata storage mode
How document metadata is persisted. COLUMN_PER_KEY stores each metadata key in its own column and requires columnDefinitions and indexes; COMBINED_JSON stores all metadata as a single JSON column. Defaults to COLUMN_PER_KEY.
Store embeddings in Milvus
Connects via URI or host/port with token-based auth; creates the target collection if missing. Metric/index/consistency options map to Milvus settings; drop=true clears the collection. Use defaults for host=localhost, port=19530, secure gRPC unless overridden.
Ingest documents into a Milvus embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.Milvus
# Use either `uri` or `host`/`port`:
# For gRPC (typical): milvus://localhost:19530
# For HTTP: http://localhost:9091
uri: "http://localhost:9091"
token: "{{ secret('MILVUS_TOKEN') }}"
collectionName: embeddings
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
Token
Milvus authentication token. Store it as a Kestra secret rather than inline. No default: this property is required.
Auto flush on delete
Intended to flush the collection after every delete. Note: this property currently has no effect, as it is not passed to the Milvus client when the store is built.
Auto flush on insert
If true, flush the collection after every insert so new vectors are immediately searchable. Setting it to false improves ingestion throughput. Defaults to false.
Collection name
Collection that stores the embeddings. Not set by default, in which case the Milvus client's own default collection name applies.
Consistency level
Read/write consistency level applied to the collection: STRONG, BOUNDED, SESSION or EVENTUALLY. Defaults to EVENTUALLY.
Database name
Logical Milvus database holding the collection. Not set by default, in which case the server's default database is used.
Host
Hostname of the Milvus server, used when uri is not set. Not set by default, in which case the Milvus client's own default applies.
ID field name
Collection field holding the document ID. Not set by default, in which case the collection schema's own field name is used.
Index type
Vector index built on the collection, such as FLAT, IVF_FLAT, IVF_SQ8, HNSW, DISKANN or AUTOINDEX. The best choice depends on the deployment and dataset size. Defaults to FLAT.
Metadata field name
Collection field holding the document metadata. Not set by default, in which case the collection schema's own field name is used.
Metric type
Similarity metric used to compare vectors: L2, IP, COSINE, HAMMING or JACCARD. It should match the metric the embedding model was trained for. Defaults to COSINE.
Password
Password of the Milvus user. Required only when authentication or TLS is enabled. Store it as a Kestra secret rather than inline. Not set by default.
Port
Port of the Milvus server, used when uri is not set. Typically 19530 for gRPC or 9091 for HTTP. Not set by default, in which case the Milvus client's own default applies.
Retrieve embeddings on search
If true, search results also carry the stored embedding vectors. Defaults to false.
Text field name
Collection field holding the original text segment. Not set by default, in which case the collection schema's own field name is used.
io.kestra.plugin.ai.embeddings.Milvusio.kestra.plugin.langchain4j.embeddings.MilvusURI
Full connection URI of the Milvus server, such as milvus://host: 19530 for gRPC or http://host: 9091 for HTTP. Set either this property or host/port, not both. Not set by default.
Username
User authenticating against Milvus. Required only when authentication or TLS is enabled; see https://milvus.io/docs/authenticate.md. Not set by default.
Vector field name
Collection field holding the embedding vector. It must match the index definition and the embedding dimensionality. Not set by default, in which case the collection schema's own field name is used.
Store embeddings in MongoDB Atlas
Uses MongoDB Atlas vector search with the provided collection/index; can optionally create the index and wait for readiness (up to 1 minute). Supply scheme/host and credentials; drop=true removes stored vectors.
Ingest documents into a MongoDB Atlas embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.MongoDBAtlas
scheme: mongodb+srv
username: "{{ secret('MONGODB_ATLAS_USERNAME') }}"
password: "{{ secret('MONGODB_ATLAS_PASSWORD') }}"
host: "{{ secret('MONGODB_ATLAS_HOST') }}"
database: "{{ secret('MONGODB_ATLAS_DATABASE') }}"
collectionName: embeddings
indexName: embeddings
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
Collection name
Collection that stores the embedding documents. No default: this property is required.
Database name
Name of the database holding the embeddings collection. No default: this property is required.
Host
Hostname of the MongoDB cluster, optionally with a port for the mongodb scheme. No default: this property is required.
Index name
Name of the Atlas Vector Search index used to query the collection. No default: this property is required.
Connection scheme
Scheme of the MongoDB connection string: mongodb+srv for Atlas clusters, mongodb for a direct connection. No default: this property is required.
Create the index
If true, create the Atlas Vector Search index when it does not already exist. Defaults to false, which assumes the index is provisioned beforehand.
Metadata field names
Metadata keys to map into the vector search index so they can be filtered on. Not set by default, in which case no metadata index mapping is created.
Connection string options
Extra options appended to the MongoDB connection string as query parameters. Not set by default.
Password
Password of the database user. Store it as a Kestra secret rather than inline. Not set by default; omit it along with username for an unauthenticated connection.
io.kestra.plugin.ai.embeddings.MongoDBAtlasio.kestra.plugin.langchain4j.embeddings.MongoDBAtlasUsername
User connecting to the MongoDB Atlas cluster. Not set by default; omit it along with password for an unauthenticated connection.
Store embeddings with pgvector
Uses the PostgreSQL pgvector extension to persist embeddings in the given table. drop=true recreates the table; optional IVF index (useIndex) defaults to false. Ensure pgvector extension is installed and the user can create tables/indexes.
Ingest documents into a PGVector embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.PGVector
host: localhost
port: 5432
user: "{{ secret('POSTGRES_USER') }}"
password: "{{ secret('POSTGRES_PASSWORD') }}"
database: postgres
table: embeddings
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
Database name
Name of the PostgreSQL database holding the embeddings table. No default: this property is required.
Database server host
Hostname or IP address of the PostgreSQL server running the pgvector extension. No default: this property is required.
Database password
Password of the database user. Store it as a Kestra secret rather than inline. No default: this property is required.
Database server port
TCP port the PostgreSQL server listens on. No default: this property is required.
Table name
Table that stores the embeddings. No default: this property is required. When drop is requested by the ingestion task, the table is recreated.
Database user
User connecting to the PostgreSQL database. No default: this property is required.
io.kestra.plugin.ai.embeddings.PGVectorio.kestra.plugin.langchain4j.embeddings.PGVectorfalseUse an IVFFlat index
If true, build an IVFFlat index on the embedding column. IVFFlat divides vectors into lists and searches only the lists closest to the query vector: it builds faster and uses less memory than HNSW, at the cost of a worse speed-recall tradeoff. Defaults to false.
Store embeddings in Pinecone
Creates or connects to a serverless Pinecone index in the given cloud/region; namespace defaults to Pinecone’s default. Requires an API key; drop=true clears the index contents.
Ingest documents into a Pinecone embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.Pinecone
apiKey: "{{ secret('PINECONE_API_KEY') }}"
cloud: AWS
region: us-east-1
index: embeddings
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
API Key
Pinecone API key used to authenticate requests. Store it as a Kestra secret rather than inline. No default: this property is required.
Cloud provider
Cloud provider hosting the Pinecone serverless index. No default: this property is required.
Index name
Pinecone index that stores the embeddings. It is created as a serverless index if it does not exist. No default: this property is required.
Cloud provider region
Region of the cloud provider hosting the serverless index. No default: this property is required.
Namespace
Namespace that partitions vectors inside the index. Not set by default, in which case Pinecone's default namespace is used.
io.kestra.plugin.ai.embeddings.Pineconeio.kestra.plugin.langchain4j.embeddings.PineconeStore embeddings in Qdrant
Uses the Qdrant gRPC client with API key, host, and port. Targets the specified collection; drop=true removes its contents. Distance metric follows the Qdrant collection defaults.
Ingest documents into a Qdrant embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.Qdrant
apiKey: "{{ secret('QDRANT_API_KEY') }}"
host: localhost
port: 6334
collectionName: embeddings
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
API Key
Qdrant API key used to authenticate requests. Store it as a Kestra secret rather than inline. No default: this property is required.
Collection name
Qdrant collection that holds the embeddings. No default: this property is required.
Database server host
Hostname or IP address of the Qdrant server. No default: this property is required.
Database server port
gRPC port the Qdrant server listens on. No default: this property is required.
io.kestra.plugin.ai.embeddings.Qdrantio.kestra.plugin.langchain4j.embeddings.QdrantStore embeddings in Redis
Backs an embedding index with Redis (jedis). Uses indexName (defaults to embedding-index); drop=true clears stored vectors. Ensure Redis deployment supports vector search modules for production use.
Ingest documents into a Redis embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.Redis
host: localhost
port: 6379
indexName: embeddings
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
Database server host
Hostname or IP address of the Redis server (Redis Stack, with the search module enabled). No default: this property is required.
Database server port
TCP port the Redis server listens on. No default: this property is required.
embedding-indexIndex name
Name of the Redis search index that stores the embeddings. Defaults to embedding-index.
Store embeddings in Alibaba Tablestore
Connects to Tablestore using access keys and writes embeddings with cosine distance. Uses the configured instance/endpoint; metadata schemas are optional. drop=true is not supported—manage cleanup in Tablestore.
Ingest documents into a Tablestore embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.Tablestore
endpoint: "{{ secret('TABLESTORE_ENDPOINT') }}"
instanceName: "{{ secret('TABLESTORE_INSTANCE_NAME') }}"
accessKeyId: "{{ secret('TABLESTORE_ACCESS_KEY_ID') }}"
accessKeySecret: "{{ secret('TABLESTORE_ACCESS_KEY_SECRET') }}"
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
Access Key ID
Alibaba Cloud access key ID used to authenticate against Tablestore. Store it as a Kestra secret rather than inline. No default: this property is required.
Access Key Secret
Alibaba Cloud access key secret paired with accessKeyId. Store it as a Kestra secret rather than inline. No default: this property is required.
Endpoint URL
Base URL of the Tablestore instance endpoint. No default: this property is required.
Instance name
Name of the Tablestore instance holding the embeddings. No default: this property is required.
Metadata schema list
Tablestore field schemas describing the metadata columns to index, so they can be filtered on at search time. Not set by default (no metadata indexed).
com.alicloud.openservices.tablestore.model.search.FieldSchema
SingleWordMaxWordMinWordSplitFuzzycom.alicloud.openservices.tablestore.model.search.analysis.AnalyzerParameter
LONGDOUBLEBOOLEANKEYWORDTEXTNESTEDGEO_POINTDATEVECTORFUZZY_KEYWORDIPJSONFLATTENEDUNKNOWNDOCSFREQSPOSITIONSOFFSETSFLATTENNESTEDOBJECTcom.alicloud.openservices.tablestore.model.search.FieldSchema
SingleWordMaxWordMinWordSplitFuzzyLONGDOUBLEBOOLEANKEYWORDTEXTNESTEDGEO_POINTDATEVECTORFUZZY_KEYWORDIPJSONFLATTENEDUNKNOWNDOCSFREQSPOSITIONSOFFSETSFLATTENNESTEDOBJECTcom.alicloud.openservices.tablestore.model.search.vector.VectorOptions
EUCLIDEANCOSINEDOT_PRODUCTStore embeddings in Weaviate
Connects to a Weaviate cluster (HTTP + optional gRPC) using the given host/scheme. apiKey, host, port, and objectClass are required. Defaults: scheme "https", avoidDups true, consistency QUORUM, secured gRPC true. drop=true clears the class contents.
Ingest documents into a Weaviate embedding store
id: document_ingestion
namespace: company.ai
tasks:
- id: ingest
type: io.kestra.plugin.ai.rag.IngestDocument
provider:
type: io.kestra.plugin.ai.provider.GoogleGemini
modelName: gemini-embedding-001
apiKey: "{{ secret('GEMINI_API_KEY') }}"
embeddings:
type: io.kestra.plugin.ai.embeddings.Weaviate
apiKey: "{{ secret('WEAVIATE_API_KEY') }}"
scheme: https # http | https (defaults to https)
host: your-cluster-id.weaviate.network # no protocol
port: 443 # required (e.g. 443 for https, 80 for http)
objectClass: Documents # required; must start with an uppercase letter
drop: true
fromExternalURLs:
- https://raw.githubusercontent.com/kestra-io/docs/refs/heads/main/content/blogs/release-0-24.md
API key
Weaviate API key used to authenticate against the cluster. Store it as a Kestra secret rather than inline. No default: this property is required.
Host
Cluster hostname, without protocol or port. No default: this property is required.
Object class
Weaviate class that stores the embedded objects. It must start with an uppercase letter. No default: this property is required.
Port
Port of the Weaviate HTTP endpoint, typically 443 for https and 80 or 8080 for http. No default: this property is required.
Avoid duplicates
If true, each object ID is derived from a hash of its text segment, so re-ingesting the same text overwrites the existing object instead of duplicating it. If false, a random ID is assigned. Defaults to true.
ONEQUORUMALLConsistency level
Write consistency applied to each object: ONE, QUORUM or ALL. Defaults to QUORUM.
gRPC port
Port of the Weaviate gRPC endpoint. Required when useGrpcForInserts is true. Not set by default.
Metadata field name
Property used to store document metadata on the object. Not set by default, in which case the Weaviate client's own default field name applies.
Metadata keys
Metadata keys to persist alongside each object. Defaults to an empty list, meaning no metadata is stored.
Scheme
Protocol used to reach the cluster: https (recommended) or http. Defaults to https.
Secure gRPC
Whether the gRPC connection uses TLS. Defaults to true. Only relevant when useGrpcForInserts is true.
io.kestra.plugin.ai.embeddings.Weaviateio.kestra.plugin.langchain4j.embeddings.WeaviateUse gRPC for batch inserts
If true, batch inserts go over gRPC, which is faster for large ingestions; searches still use HTTP. Requires grpcPort. Defaults to false.
maxResults integerstring
3Maximum results
Number of matching segments returned by the embedding store for each query. Defaults to 3.
minScore numberstring
0.0Minimum similarity score
Similarity threshold a match must reach to be returned, from 0.0 (keep everything) to 1.0 (exact match only). Defaults to 0.0.
Metrics
ai.embedding.store.calls counter
callsNumber of times an embedding store is used, tagged by store class name
ai.provider.calls counter
callsNumber of times an embedding model is obtained from a provider, tagged by provider class name
