phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 03:04:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

(fix) update docstring for get_max_tokens

Browse source
This commit is contained in:
ishaan-jaff 2023-10-18 09:16:33 -07:00
parent fa488e29e0
commit 0c090e3675
1 changed files with 17 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

22
litellm/utils.py
View file

@ -1608,20 +1608,32 @@ def get_api_key(llm_provider: str, dynamic_api_key: Optional[str]):
def get_max_tokens(model: str): def get_max_tokens(model: str):
""" """
Get the maximum tokens (context window) for a given model. Get a dict for the maximum tokens (context window),
input_cost_per_token, output_cost_per_token for a given model.
Parameters: Parameters:
model (str): The name of the model. model (str): The name of the model.
Returns: Returns:
int: The maximum tokens for the given model. dict: A dictionary containing the following information:
- max_tokens (int): The maximum number of tokens allowed for the given model.
- input_cost_per_token (float): The cost per token for input.
- output_cost_per_token (float): The cost per token for output.
- litellm_provider (str): The provider of the model (e.g., "openai").
- mode (str): The mode of the model (e.g., "chat" or "completion").
Raises: Raises:
Exception: If the model is not mapped yet. Exception: If the model is not mapped yet.
Example: Example:
>>> get_max_tokens("gpt-3.5-turbo") >>> get_max_tokens("gpt-4")
4096 {
"max_tokens": 8192,
"input_cost_per_token": 0.00003,
"output_cost_per_token": 0.00006,
"litellm_provider": "openai",
"mode": "chat"
}
""" """
try: try:
return litellm.model_cost[model] return litellm.model_cost[model]