Connect
Also in Cloud

openai_chat_completion

Generates responses to messages in a chat conversation, using the OpenAI API and external tools.

Introduced in version 4.32.0.

  • Common

  • Advanced

processor:
  label: ""
  openai_chat_completion:
    server_address: https://api.openai.com/v1
    api_key: "" # No default (required)
    model: "" # No default (required)
    prompt: "" # No default (optional)
    system_prompt: "" # No default (optional)
    history: "" # No default (optional)
    image: "" # No default (optional)
    max_tokens: 0 # No default (optional)
    temperature: 0 # No default (optional)
    user: "" # No default (optional)
    response_format: text
    json_schema:
      name: "" # No default (required)
      schema: "" # No default (required)
    tools: [] # No default (required)
processor:
  label: ""
  openai_chat_completion:
    server_address: https://api.openai.com/v1
    api_key: "" # No default (required)
    model: "" # No default (required)
    prompt: "" # No default (optional)
    system_prompt: "" # No default (optional)
    history: "" # No default (optional)
    image: "" # No default (optional)
    max_tokens: 0 # No default (optional)
    temperature: 0 # No default (optional)
    user: "" # No default (optional)
    response_format: text
    json_schema:
      name: "" # No default (required)
      description: "" # No default (optional)
      schema: "" # No default (required)
    schema_registry:
      url: "" # No default (required)
      name_prefix: schema_registry_id_
      subject: "" # No default (required)
      refresh_interval: "" # No default (optional)
      tls:
        skip_cert_verify: false
        enable_renegotiation: false
        root_cas: ""
        root_cas_file: ""
        client_certs: []
      oauth:
        enabled: false
        consumer_key: ""
        consumer_secret: ""
        access_token: ""
        access_token_secret: ""
      basic_auth:
        enabled: false
        username: ""
        password: ""
      jwt:
        enabled: false
        private_key_file: ""
        signing_method: ""
        claims: {}
        headers: {}
    top_p: 0 # No default (optional)
    frequency_penalty: 0 # No default (optional)
    presence_penalty: 0 # No default (optional)
    seed: 0 # No default (optional)
    stop: [] # No default (optional)
    tools: [] # No default (required)

This processor sends user prompts to the OpenAI API, and the specified large language model (LLM) generates responses using all available context, including supplementary data provided by external tools. By default, the processor submits the entire payload of each message as a string, unless you use the prompt configuration field to customize it.

To learn more about chat completion, see the OpenAI API documentation, and Examples.

Fields

api_key

The API secret key for OpenAI API.

This field contains sensitive information that usually shouldn’t be added to a configuration directly. For more information, see Secrets.

Type: string

frequency_penalty

Specify a number between -2.0 and 2.0. Positive values penalize new tokens based on the frequency of their appearance in the text so far. This decreases the model’s likelihood to repeat the same line verbatim.

Requires version 4.34.0 or later.

Type: float

history

Include messages from a prior conversation. You must use a Bloblang query to create an array of objects in the form of [{"role": "user", "content": "<text>"}, {"role":"assistant", "content":"<text>"}] where:

  • role is the sender of the original messages, either system, user, or assistant.

  • content is the text of the original messages.

Requires version 4.51.0 or later.

Type: string

image

An optional image to submit along with the prompt. The result of the Bloblang mapping must be a byte array.

Requires version 4.38.0 or later.

Type: string

# Examples:
image: 'root = this.image.decode("base64") # decode base64 encoded image'

json_schema

The JSON schema used by the model when generating responses in json_schema format. To learn more about supported JSON schema features, see the OpenAI documentation.

Requires version 4.35.0 or later.

Type: object

json_schema.description

An optional description, which helps the model understand the schema’s purpose.

Type: string

json_schema.name

The name of the JSON schema to use.

Type: string

json_schema.schema

The JSON schema for the model to use when generating the output.

Type: string

max_tokens

The maximum number of tokens to generate for chat completion.

Requires version 4.34.0 or later.

Type: int

model

The name of the OpenAI model to use.

Type: string

# Examples:
model: gpt-4o

# ---

model: gpt-4o-mini

# ---

model: gpt-4

# ---

model: gpt4-turbo

presence_penalty

Specify a number between -2.0 and 2.0. Positive values penalize new tokens if they have appeared in the text so far. This increases the model’s likelihood to talk about new topics.

Requires version 4.34.0 or later.

Type: float

prompt

The user prompt you want to generate a response for. By default, the processor submits the entire payload of each message as a string.

This field supports interpolation functions.

Type: string

response_format

Specify the output format of the configured model.

If you choose the json_schema option, you must also configure a json_schema or schema_registry.

Requires version 4.35.0 or later.

Type: string

Default: text

Options: text, json, json_schema

schema_registry

The schema registry to dynamically load schemas for model responses in json_schema format. Schemas must be in JSON format. To learn more about supported JSON schema features, see the OpenAI documentation.

Requires version 4.35.0 or later.

Type: object

schema_registry.basic_auth

Configure basic authentication for requests from this component.

Type: object

schema_registry.basic_auth.enabled

Whether to use basic authentication in requests.

Type: bool

Default: false

schema_registry.basic_auth.password

The password to use for authentication. Used together with username for basic authentication.

This field contains sensitive information that usually shouldn’t be added to a configuration directly. For more information, see Secrets.

Type: string

Default: ""

schema_registry.basic_auth.username

The username of the account credentials to authenticate as. Used together with password for basic authentication.

Type: string

Default: ""

schema_registry.jwt

Beta

Configure JSON Web Token (JWT) authentication. This feature is in beta and may change in future releases. JWTs provide secure, stateless authentication between services.

Type: object

schema_registry.jwt.claims

A map of claims to include in the JWT. Claims pass the identity of the authenticated entity to the service provider.

Type: object

Default: {}

schema_registry.jwt.enabled

Whether to use JWT authentication in requests.

Type: bool

Default: false

schema_registry.jwt.headers

Additional key-value pairs to include in the JWT header (optional). These headers provide extra metadata for JWT processing.

Type: object

Default: {}

schema_registry.jwt.private_key_file

Path to a file containing the PEM-encoded private key using PKCS#1 or PKCS#8 format. The private key must be compatible with the algorithm specified in the signing_method field.

Type: string

Default: ""

schema_registry.jwt.signing_method

The cryptographic algorithm used to sign the JWT. Supported algorithms are RS256, RS384, RS512, and EdDSA. This algorithm must be compatible with the private key specified in the private_key_file field.

Type: string

Default: ""

schema_registry.name_prefix

A prefix to add to the schema registry name. To form the complete schema registry name, the schema ID is appended as a suffix.

Type: string

Default: schema_registry_id_

schema_registry.oauth

Configure OAuth version 1.0 authentication for secure API access.

Type: object

schema_registry.oauth.access_token

The value used to gain access to the protected resources on behalf of the user.

Type: string

Default: ""

schema_registry.oauth.access_token_secret

The secret that establishes ownership of the access_token in OAuth 1.0 authentication.

This field contains sensitive information that usually shouldn’t be added to a configuration directly. For more information, see Secrets.

Type: string

Default: ""

schema_registry.oauth.consumer_key

The value used to identify this component or client to the service provider.

Type: string

Default: ""

schema_registry.oauth.consumer_secret

The secret that establishes ownership of the consumer key in OAuth 1.0 authentication.

This field contains sensitive information that usually shouldn’t be added to a configuration directly. For more information, see Secrets.

Type: string

Default: ""

schema_registry.oauth.enabled

Whether to enable OAuth version 1.0 authentication for requests.

Type: bool

Default: false

schema_registry.refresh_interval

How frequently to poll the schema registry for the latest schema. If not specified, the schema does not refresh.

Type: string

schema_registry.subject

The subject name used to fetch the schema from the schema registry.

Type: string

schema_registry.tls

Configure Transport Layer Security (TLS) settings to secure network connections. This includes options for standard TLS as well as mutual TLS (mTLS) authentication where both client and server authenticate each other using certificates. Key configuration options include client_certs for mTLS authentication, root_cas/root_cas_file for custom certificate authorities, and skip_cert_verify for development environments.

Type: object

schema_registry.tls.client_certs[]

A list of client certificates for mutual TLS (mTLS) authentication. Configure this field to enable mTLS, authenticating the client to the server with these certificates.

Certificate pairing rules: For each certificate item, provide either:

  • Inline PEM data using both cert and key or

  • File paths using both cert_file and key_file.

Mixing inline and file-based values within the same item is not supported.

Type: array<object>

Default: []

# Examples:
client_certs:
  - cert: foo
    key: bar


# ---

client_certs:
  - cert_file: ./example.pem
    key_file: ./example.key

schema_registry.tls.client_certs[].cert

The plaintext certificate to use for TLS authentication. Must be paired with the corresponding private key in the key field when using inline PEM data for mTLS client certificates.

Type: string

Default: ""

schema_registry.tls.client_certs[].cert_file

The path to a file containing the certificate to use for TLS authentication. Must be paired with the corresponding private key file in the key_file field when using file-based configuration for mTLS client certificates.

Type: string

Default: ""

schema_registry.tls.client_certs[].key

Private key for mTLS client certificate as inline PEM data. Must correspond to the client certificate specified in the cert field. Use this field together with cert when providing certificate data inline rather than through files.

This field contains sensitive information that usually shouldn’t be added to a configuration directly. For more information, see Secrets.

Type: string

Default: ""

schema_registry.tls.client_certs[].key_file

Path to private key file for mTLS client certificate in PEM format. Must correspond to the client certificate specified in the cert_file field. Use this field together with cert_file when loading certificate data from files.

Type: string

Default: ""

schema_registry.tls.client_certs[].password

The password to use for the private key (specified in the key or key_file fields), if it is password-protected. The PKCS#1 and PKCS#8 formats are supported. Supports environment variable interpolation for secure password management.

The pbeWithMD5AndDES-CBC algorithm is obsolete and not supported for the PKCS#8 format. This algorithm does not authenticate the ciphertext, making it vulnerable to padding oracle attacks that can let an attacker recover the plaintext.

This field contains sensitive information that usually shouldn’t be added to a configuration directly. For more information, see Secrets.

Type: string

Default: ""

# Examples:
password: foo

# ---

password: ${KEY_PASSWORD}

schema_registry.tls.enable_renegotiation

Whether to allow the remote server to repeatedly request renegotiation. Enable this option if you’re seeing the error message local error: tls: no renegotiation.

Type: bool

Default: false

schema_registry.tls.root_cas

Specify a root certificate authority to use (optional). This is a string that represents a certificate chain from the parent-trusted root certificate, through possible intermediate signing certificates, to the host certificate. Use either this field for inline certificate data or root_cas_file for file-based certificate loading.

This field contains sensitive information that usually shouldn’t be added to a configuration directly. For more information, see Secrets.

Type: string

Default: ""

# Examples:
root_cas: |-
  -----BEGIN CERTIFICATE-----
  ...
  -----END CERTIFICATE-----

schema_registry.tls.root_cas_file

Specify the path to a root certificate authority file (optional). This is a file, often with a .pem extension, which contains a certificate chain from the parent-trusted root certificate, through possible intermediate signing certificates, to the host certificate. Use either this field for file-based certificate loading or root_cas for inline certificate data.

Type: string

Default: ""

# Examples:
root_cas_file: ./root_cas.pem

schema_registry.tls.skip_cert_verify

Whether to skip server-side certificate verification. Set to true only for testing environments as this reduces security by disabling certificate validation. When using self-signed certificates or in development, this may be necessary, but should never be used in production. Consider using root_cas or root_cas_file to specify trusted certificates instead of disabling verification entirely.

Type: bool

Default: false

schema_registry.url

The base URL of the schema registry service.

Type: string

seed

When set to a specific number, the model makes a best effort to generate consistent responses for requests that use the same prompt, seed, and parameters. Determinism is not guaranteed.

Requires version 4.34.0 or later.

Type: int

server_address

The OpenAI API endpoint to which the processor sends requests. Update the default value to use a different OpenAI-compatible service.

Type: string

Default: https://api.openai.com/v1

stop[]

Specify up to four stop sequences to use. When the model encounters a stop pattern, it stops generating text and returns the final response.

Requires version 4.34.0 or later.

Type: array<string>

system_prompt

The system prompt to submit along with the user prompt.

This field supports interpolation functions.

Type: string

temperature

Choose a sampling temperature between 0 and 2:

  • Higher values, such as 0.8 make the output more random.

  • Lower values, such as 0.2 make the output more focused and deterministic.

Redpanda recommends adding a value for this field or top_p, but not both.

Requires version 4.34.0 or later.

Type: float

tools[]

External tools the model can invoke, such as functions, APIs, or web browsing. You can build subpipelines of processors that include definitions of these tools, and the specified model can choose when to invoke them to help answer a prompt.

If you don’t want to use external tools, enter an empty array tools: [].

Requires version 4.50.0 or later.

Type: array<object>

tools[].description

A description of this tool. The LLM uses it to decide whether to invoke the tool.

Type: string

tools[].name

The name of this tool.

Type: string

tools[].parameters

The parameters the LLM needs to provide to invoke this tool.

Type: object

Default: []

tools[].parameters.properties

The parameters the LLM can provide when it invokes this tool, keyed by parameter name.

Type: object

tools[].parameters.properties.description

A description of this parameter.

Type: string

tools[].parameters.properties.enum[]

The values this parameter is limited to. Leave empty to accept any value.

Type: array<string>

Default: []

tools[].parameters.properties.type

The type of this parameter.

Type: string

tools[].parameters.required[]

The names of the parameters the LLM must provide when it invokes this tool.

Type: array<string>

Default: []

tools[].processors[]

The processors to run when the LLM invokes this tool. They receive a message whose payload is the tool call arguments as a JSON object, and their output is returned to the LLM as the tool result.

Type: array<processor>

top_p

An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. For example, a top_p of 0.1 means only the tokens comprising the top 10% probability mass are sampled.

Redpanda recommends adding a value for this field or temperature, but not both.

Requires version 4.34.0 or later.

Type: float

user

A unique identifier that represents the end-user generating the prompt. This value can help OpenAI monitor and detect platform abuse.

This field supports interpolation functions.

Requires version 4.34.0 or later.

Type: string

Examples

Use GPT-4o analyze an image

This example fetches image URLs from stdin and has GPT-4o describe the image.

input:
  stdin:
    scanner:
      lines: {}
pipeline:
  processors:
    - http:
        verb: GET
        url: "${!content().string()}"
    - openai_chat_completion:
        model: gpt-4o
        api_key: TODO
        prompt: "Describe the following image"
        image: "root = content()"
output:
  stdout:
    codec: lines

Provide historical chat history

This pipeline provides a historical chat history to GPT-4o using a cache.

input:
  stdin:
    scanner:
      lines: {}
pipeline:
  processors:
    - mapping: |
        root.prompt = content().string()
    - branch:
        processors:
          - cache:
              resource: mem
              operator: get
              key: history
          - catch:
            - mapping: 'root = []'
        result_map: 'root.history = this'
    - branch:
        processors:
        - openai_chat_completion:
            model: gpt-4o
            api_key: TODO
            prompt: "${!this.prompt}"
            history: 'root = this.history'
        result_map: 'root.response = content().string()'
    - mutation: |
        root.history = this.history.concat([
          {"role": "user", "content": this.prompt},
          {"role": "assistant", "content": this.response},
        ])
    - cache:
        resource: mem
        operator: set
        key: history
        value: '${!this.history}'
    - mapping: |
        root = this.response
output:
  stdout:
    codec: lines

cache_resources:
  - label: mem
    memory: {}

Use GPT-4o to call a tool

This example asks GPT-4o to respond with the weather by invoking an HTTP processor to get the forecast.

input:
  generate:
    count: 1
    mapping: |
      root = "What is the weather like in Chicago?"
pipeline:
  processors:
    - openai_chat_completion:
        model: gpt-4o
        api_key: "${OPENAI_API_KEY}"
        prompt: "${!content().string()}"
        tools:
          - name: GetWeather
            description: "Retrieve the weather for a specific city"
            parameters:
              required: ["city"]
              properties:
                city:
                  type: string
                  description: the city to look up the weather for
            processors:
              - http:
                  verb: GET
                  url: 'https://wttr.in/${!this.city}?T'
                  headers:
                    User-Agent: curl/8.11.1 # Returns a text string from the weather website
output:
  stdout: {}