Chat

POST 
/api/v1/pipelines/:pipeline_id/chat

Make a retrieval query + chat completion for a managed pipeline.

Request

Path Parameters

pipeline_id uuidrequired

Cookie Parameters

session any

application/json

Body

required

messages

object[]

Array [

id Id (string)required

ID of the message, if any. a UUID.

role MessageRole (string)required

Message role.

Possible values: [system, user, assistant, function, tool, chatbot, model]

content Content (string)required

data

object

Additional data to be stored with the message.

anyOf

MOD1

object

class_name Class Name (string)

Default value: base_component

]

data

object

retrieval_parameters

object

Schema for the search params for an retrieval execution that can be preset for a pipeline.

dense_similarity_top_k

object

Number of nodes for dense retrieval.

anyOf

MOD1

integer

Possible values: >= 1 and <= 100

dense_similarity_cutoff

object

Minimum similarity score wrt query for retrieval

anyOf

MOD1

number

Possible values: <= 1

sparse_similarity_top_k

object

Number of nodes for sparse retrieval.

anyOf

MOD1

integer

Possible values: >= 1 and <= 100

enable_reranking

object

Enable reranking for retrieval

anyOf

MOD1

boolean

rerank_top_n

object

Number of reranked nodes for returning.

anyOf

MOD1

integer

Possible values: >= 1 and <= 100

alpha

object

Alpha value for hybrid retrieval to determine the weights between dense and sparse retrieval. 0 is sparse retrieval and 1 is dense retrieval.

anyOf

MOD1

number

Possible values: <= 1

search_filters

object

Search filters for retrieval.

anyOf

MetadataFilters

Metadata filters for vector stores.

filters

object[]

required

Array [

anyOf

MetadataFilter

Comprehensive metadata filter for vector stores to support more operators.

Value uses Strict* types, as int, float and str are compatible types and were all converted to string before.

See: https://docs.pydantic.dev/latest/usage/types/#strict-types

key Key (string)required

value

object

required

anyOf

MOD1

integer

MOD2

number

MOD3

string

MOD4

Array [

string

]

MOD5

Array [

number

]

MOD6

Array [

integer

]

operator FilterOperator (string)

Vector store filter operator.

Possible values: [==, >, <, !=, >=, <=, in, nin, any, all, text_match, contains, is_empty]

Default value: ==

]

condition

object

anyOf

FilterCondition

Vector store filter conditions to combine different filters.

string

Possible values: [and, or]

files_top_k

object

Number of files to retrieve (only for retrieval mode files_via_metadata and files_via_content).

anyOf

MOD1

integer

Possible values: >= 1 and <= 5

retrieval_mode RetrievalMode (string)

The retrieval mode for the query.

Possible values: [chunks, files_via_metadata, files_via_content, auto_routed]

Default value: auto_routed

retrieve_image_nodes Retrieve Image Nodes (boolean)

Whether to retrieve image nodes.

Default value: false

class_name Class Name (string)

Default value: base_component

llm_parameters

object

anyOf

LLMParameters

model_name SupportedLLMModelNames (string)

The name of the model to use for LLM completions.

Possible values: [GPT_3_5_TURBO, GPT_4, GPT_4_TURBO, GPT_4O, GPT_4O_MINI, AZURE_OPENAI]

Default value: GPT_3_5_TURBO

system_prompt

object

The system prompt to use for the completion.

anyOf

MOD1

string

temperature

object

The temperature value for the model.

anyOf

MOD1

number

class_name Class Name (string)

Default value: base_component

class_name Class Name (string)

Default value: base_component
class_name Class Name (string)
Default value: base_component

Responses

200

Successful Response

application/json

Schema

Schema

any

422

Validation Error

application/json

Schema

Schema

detail

object[]

Array [

loc

object[]

required

Array [

anyOf

MOD1

string

MOD2

integer

]

msg Message (string)required

type Error Type (string)required

]

Example (from schema)

{
  "detail": [
    {
      "loc": [
        "string",
        0
      ],
      "msg": "string",
      "type": "string"
    }
  ]
}

Loading...