-
Notifications
You must be signed in to change notification settings - Fork 69
Description
When I tried to use Roo code extension's codebase indexing in RunVSAgent, I came across an error that request was over LLM's batch size limit.
Error - Failed during initial scan: Indexing failed: Failed to process batch after 3 attempts: Failed to create embeddings after 3 attempts: HTTP 413 - 413 litellm.APIError: APIError: OpenAIException - Error code: 413 - {'message': 'batch size 60 > maximum allowed batch size 32', 'code': 413, 'type': 'Validation'}. Received Model Group=BAAI/bge-m3 Available Model Group Fallbacks=None
I searched for help on Roo's github issue. They said user could configure this limit in VSCode setting.
{ "roo-code.codeIndex.embeddingBatchSize": 10 }
RooCodeInc/Roo-Code#7464
Could you please tell me how can I configure VSCode setting params in RunVSAgent environment?