Update max output tokens to 16K (#1110)

This commit is contained in:
Shuchang Zheng 2024-11-02 21:46:55 -07:00 committed by GitHub
parent 3e0639ba38
commit bf06fcfeb7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 19 additions and 6 deletions

View file

@ -34,6 +34,7 @@ class LLMConfigBase:
@dataclass(frozen=True)
class LLMConfig(LLMConfigBase):
litellm_params: Optional[LiteLLMParams] = field(default=None)
max_output_tokens: int = SettingsManager.get_settings().LLM_CONFIG_MAX_TOKENS
@dataclass(frozen=True)
@ -69,6 +70,7 @@ class LLMRouterConfig(LLMConfigBase):
allowed_fails: int | None = None
allowed_fails_policy: AllowedFailsPolicy | None = None
cooldown_time: float | None = None
max_output_tokens: int = SettingsManager.get_settings().LLM_CONFIG_MAX_TOKENS
class LLMAPIHandler(Protocol):