Updates a knowledge space.
Try it now
Test
RAM authorization
Request syntax
POST HTTP/1.1Request parameters
|
Parameter |
Type |
Required |
Description |
Example |
| RegionId |
string |
Yes |
The region ID. |
cn-beijing |
| KnowledgeSpaceId |
string |
Yes |
The unique identifier of the knowledge space. |
pks-xxxxxx |
| Name |
string |
No |
The name of the knowledge space. The name must be 1 to 128 characters in length. |
testName |
| Description |
string |
No |
The description of the knowledge space. The description can be up to 512 characters in length. |
testDesc |
| RerankModel |
string |
No |
The name of the reranking model. |
qwen3-rerank |
| LLMModel |
string |
No |
The name of the large language model. |
qwen3.6-plus |
| ShardingStrategyConfig |
object |
No |
The default chunking strategy configuration for the knowledge space. Both simple strategies and composite strategies that match by content type are supported. |
|
| DefaultStrategy |
object |
No |
The default chunking strategy. This strategy is used when no rule is matched. |
|
| Type |
string |
No |
The type of the default chunking strategy. Valid values: hybrid or hierarchical. |
hybrid |
| Parameters |
object |
No |
The parameters of the default chunking strategy. MaxTokens and MergePeers take effect only when Type is set to hybrid. |
|
| MaxTokens |
integer |
No |
The maximum number of tokens per chunk. The value must be a positive integer. |
512 |
| MergePeers |
boolean |
No |
Specifies whether to merge adjacent small chunks under the same heading. |
true |
| Rules |
array<object> |
No |
The list of override rules that are matched in order. Currently, a maximum of one exact-match rule with ContentType set to table is supported. |
|
|
array<object> |
No |
The rule details. |
||
| Match |
object |
No |
The match condition of the rule. Currently, only exact matching of table content by content type is supported. |
|
| ContentType |
string |
No |
The content type. Currently, only table is supported. |
table |
| Strategy |
object |
No |
The chunking strategy to use when the rule is matched. |
|
| Type |
string |
No |
The type of the chunking strategy to use when the rule is matched. Valid values: hybrid or hierarchical. |
hierarchical |
| Parameters |
object |
No |
The chunking strategy parameters for the override rule. MaxTokens takes effect only when Type is set to hybrid. MarkdownTables supports auto, on, or off. |
|
| MaxTokens |
integer |
No |
The maximum number of tokens per chunk for matched content. The value must be a positive integer. |
512 |
| MarkdownTables |
string |
No |
The processing mode for Markdown tables. Valid values: auto, on, or off. |
auto |
Response elements
|
Element |
Type |
Description |
Example |
|
object |
Schema of Response |
||
| RequestId |
string |
Id of the request |
CD35F3-F3-44CA-AFFF-BAF869****** |
| KnowledgeSpaceId |
string |
The unique identifier of the knowledge space. |
pks-xxxxxx |
| TaskId |
integer |
The ID of the asynchronous workflow task. This value is empty if no actual model change occurs. |
52*****03 |
Examples
Success response
JSON format
{
"RequestId": "CD35F3-F3-44CA-AFFF-BAF869******",
"KnowledgeSpaceId": "pks-xxxxxx",
"TaskId": 0
}Error codes
See Error Codes for a complete list.
Release notes
See Release Notes for a complete list.