mirror of
https://github.com/theroyallab/tabbyAPI.git
synced 2026-03-15 00:07:28 +00:00
OAI: Clarify types in docs
Adding field descriptions show which parameters are used solely for OAI compliance and not actually parsed in the model code. Signed-off-by: kingbri <bdashore3@proton.me>
This commit is contained in:
@@ -19,13 +19,13 @@ class CommonCompletionRequest(BaseModel):
|
|||||||
model: Optional[str] = None
|
model: Optional[str] = None
|
||||||
|
|
||||||
# Extra OAI request stuff
|
# Extra OAI request stuff
|
||||||
best_of: Optional[int] = None
|
best_of: Optional[int] = Field(description = "Not parsed. Only used for OAI compliance.", default = None)
|
||||||
echo: Optional[bool] = False
|
echo: Optional[bool] = Field(description = "Not parsed. Only used for OAI compliance.", default = False)
|
||||||
logit_bias: Optional[Dict[str, float]] = None
|
logit_bias: Optional[Dict[str, float]] = Field(description = "Not parsed. Only used for OAI compliance.", default = None)
|
||||||
logprobs: Optional[int] = None
|
logprobs: Optional[int] = Field(description = "Not parsed. Only used for OAI compliance.", default = None)
|
||||||
n: Optional[int] = 1
|
n: Optional[int] = Field(description = "Not parsed. Only used for OAI compliance.", default = 1)
|
||||||
suffix: Optional[str] = None
|
suffix: Optional[str] = Field(description = "Not parsed. Only used for OAI compliance.", default = None)
|
||||||
user: Optional[str] = None
|
user: Optional[str] = Field(description = "Not parsed. Only used for OAI compliance.", default = None)
|
||||||
|
|
||||||
# Generation info
|
# Generation info
|
||||||
# seed: Optional[int] = -1
|
# seed: Optional[int] = -1
|
||||||
@@ -36,7 +36,7 @@ class CommonCompletionRequest(BaseModel):
|
|||||||
max_tokens: Optional[int] = 150
|
max_tokens: Optional[int] = 150
|
||||||
|
|
||||||
# Aliased to repetition_penalty
|
# Aliased to repetition_penalty
|
||||||
frequency_penalty: Optional[float] = 0.0
|
frequency_penalty: Optional[float] = Field(description = "Aliased to Repetition Penalty", default = 0.0)
|
||||||
|
|
||||||
# Sampling params
|
# Sampling params
|
||||||
token_healing: Optional[bool] = False
|
token_healing: Optional[bool] = False
|
||||||
|
|||||||
@@ -27,7 +27,7 @@ class ModelList(BaseModel):
|
|||||||
class DraftModelLoadRequest(BaseModel):
|
class DraftModelLoadRequest(BaseModel):
|
||||||
draft_model_name: str
|
draft_model_name: str
|
||||||
draft_rope_scale: Optional[float] = 1.0
|
draft_rope_scale: Optional[float] = 1.0
|
||||||
draft_rope_alpha: Optional[float] = None
|
draft_rope_alpha: Optional[float] = Field(description = "Automatically calculated if not present", default = None)
|
||||||
|
|
||||||
# TODO: Unify this with ModelCardParams
|
# TODO: Unify this with ModelCardParams
|
||||||
class ModelLoadRequest(BaseModel):
|
class ModelLoadRequest(BaseModel):
|
||||||
@@ -36,7 +36,7 @@ class ModelLoadRequest(BaseModel):
|
|||||||
gpu_split_auto: Optional[bool] = True
|
gpu_split_auto: Optional[bool] = True
|
||||||
gpu_split: Optional[List[float]] = Field(default_factory=list)
|
gpu_split: Optional[List[float]] = Field(default_factory=list)
|
||||||
rope_scale: Optional[float] = 1.0
|
rope_scale: Optional[float] = 1.0
|
||||||
rope_alpha: Optional[float] = None
|
rope_alpha: Optional[float] = Field(description = "Automatically calculated if not present", default = None)
|
||||||
no_flash_attention: Optional[bool] = False
|
no_flash_attention: Optional[bool] = False
|
||||||
# low_mem: Optional[bool] = False
|
# low_mem: Optional[bool] = False
|
||||||
cache_mode: Optional[str] = "FP16"
|
cache_mode: Optional[str] = "FP16"
|
||||||
|
|||||||
Reference in New Issue
Block a user