Skip to main content
PATCH
Update Knowledge Base

Path Parameters

kb_id
string
required

Knowledge Base ID.

Body

application/json

Request body for updating a knowledge base. Every field is optional and nullable; omitting a field leaves it unchanged. embedding_model and vector_backend are absent by design - both are fixed at creation and cannot be changed afterwards.

name
string | null

New name for the knowledge base. Omit to leave it unchanged.

description
string | null

New description. Omit to leave it unchanged.

document_language
string | null

New primary language. A free-form string, not an enum: an unrecognised value is accepted and stored rather than rejected.

chunking_strategy
enum<string> | null

Strategy used to split source material into chunks before embedding.

Available options:
fixed_size,
semantic,
sentence,
paragraph,
recursive
chunk_size
integer | null

New chunk size. Applied at index time; nothing already indexed is re-chunked.

Required range: 100 <= x <= 4096
chunk_overlap
integer | null

New chunk overlap. Applied at index time.

Required range: 0 <= x <= 500
assistant_persona
string | null

New persona. Because omitting a field means 'leave unchanged', null is indistinguishable from not sending the field at all - send an empty string to reset the persona to its name-derived default.

Maximum string length: 280

Response

The updated knowledge base.

Response model for a knowledge base.

id
string
required

Unique identifier of the knowledge base.

name
string
required

The name of the knowledge base.

created_at
string<date-time>
required

UTC timestamp indicating when this knowledge base was created.

updated_at
string<date-time>
required

UTC timestamp indicating when this knowledge base was last modified.

description
string | null

Description of the knowledge base's purpose and contents.

document_language
string
default:en

Primary language of the material in this knowledge base.

document_count
integer
default:0

Number of documents stored in this knowledge base.

chunking_strategy
enum<string>
default:recursive

Chunking strategy currently in effect for new material.

Available options:
fixed_size,
semantic,
sentence,
paragraph,
recursive
vector_backend
enum<string>
default:pinecone

Vector database backing this knowledge base's index.

Available options:
pinecone,
qdrant,
weaviate,
milvus,
pgvector
embedding_model
enum<string>
default:text-embedding-3-small

Embedding model in use. Note the asymmetry with the create request, which defaults to text-embedding-3-large: this schema's default is text-embedding-3-small because knowledge bases created before the field existed are on small. Read the value off the response rather than assuming it.

Available options:
text-embedding-3-small,
text-embedding-3-large
chunk_size
integer
default:512

Target chunk size for new material.

chunk_overlap
integer
default:50

Overlap between consecutive chunks.

assistant_persona
string | null

How an assistant built on this knowledge base introduces itself.

status
enum<string> | null

Aggregate processing state of the knowledge base's documents.

Available options:
processing,
completed,
failed
workspace_id
string | null

Workspace this knowledge base belongs to.

organization_id
string | null

Organization this knowledge base belongs to, resolved from your token.

user_id
string | null

User this knowledge base belongs to, resolved from your token.