Config: replace disable_output_chunking flag with output_chunking

This commit is contained in:
turboderp
2025-10-14 02:47:52 +02:00
parent 7eee3924c7
commit 8abdfe7b13
4 changed files with 11 additions and 10 deletions

View File

@@ -250,8 +250,8 @@ class ExllamaV3Container(BaseModelContainer):
self.chunk_size = self.adjust_chunk_size(user_chunk_size)
# Output chunking
disable_output_chunking = unwrap(kwargs.get("disable_output_chunking"), False)
self.max_rq_tokens = None if disable_output_chunking else self.chunk_size
output_chunking = unwrap(kwargs.get("output_chunking"), True)
self.max_rq_tokens = self.chunk_size if output_chunking else None
# Template setup
self.prompt_template = await find_prompt_template(