fix: tolerate spotify version suffixes and surface rerank failures
This commit is contained in:
parent
ce652d3114
commit
0ccc9a5d4e
6 changed files with 92 additions and 29 deletions
|
|
@ -41,8 +41,10 @@ class Settings(BaseSettings):
|
|||
llm_model: str = "claude-sonnet-5"
|
||||
intent_effort: str = "low"
|
||||
rerank_effort: str = "medium"
|
||||
intent_max_tokens: int = 8192
|
||||
rerank_max_tokens: int = 4096
|
||||
# Ceilings include adaptive thinking tokens, which is why they sit far
|
||||
# above the size of the structured output itself.
|
||||
intent_max_tokens: int = 16384
|
||||
rerank_max_tokens: int = 16384
|
||||
|
||||
# Pipeline shape. candidate_count is the main call-1 latency lever and
|
||||
# the hallucination budget: at "new to you" familiarity a large share of
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue