phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 19:24:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

(feat) proxy: pass model_info as kwargs

Browse source
This commit is contained in:
ishaan-jaff 2023-12-08 14:07:31 -08:00
parent de6880dc09
commit 0740728be0
1 changed files with 4 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

4
litellm/router.py
View file

@ -187,6 +187,7 @@ class Router:
deployment = self.get_available_deployment(model=model, messages=messages, specific_deployment=kwargs.pop("specific_deployment", None)) deployment = self.get_available_deployment(model=model, messages=messages, specific_deployment=kwargs.pop("specific_deployment", None))
kwargs.setdefault("metadata", {}).update({"deployment": deployment["litellm_params"]["model"]}) kwargs.setdefault("metadata", {}).update({"deployment": deployment["litellm_params"]["model"]})
data = deployment["litellm_params"].copy() data = deployment["litellm_params"].copy()
kwargs["model_info"] = deployment.get("model_info", {})
for k, v in self.default_litellm_params.items(): for k, v in self.default_litellm_params.items():
if k not in data: # prioritize model-specific params > default router params if k not in data: # prioritize model-specific params > default router params
data[k] = v data[k] = v
@ -234,6 +235,7 @@ class Router:
original_model_string = None # set a default for this variable original_model_string = None # set a default for this variable
deployment = self.get_available_deployment(model=model, messages=messages, specific_deployment=kwargs.pop("specific_deployment", None)) deployment = self.get_available_deployment(model=model, messages=messages, specific_deployment=kwargs.pop("specific_deployment", None))
kwargs.setdefault("metadata", {}).update({"deployment": deployment["litellm_params"]["model"]}) kwargs.setdefault("metadata", {}).update({"deployment": deployment["litellm_params"]["model"]})
kwargs["model_info"] = deployment.get("model_info", {})
data = deployment["litellm_params"].copy() data = deployment["litellm_params"].copy()
for k, v in self.default_litellm_params.items(): for k, v in self.default_litellm_params.items():
if k not in data: # prioritize model-specific params > default router params if k not in data: # prioritize model-specific params > default router params
@ -303,6 +305,7 @@ class Router:
# pick the one that is available (lowest TPM/RPM) # pick the one that is available (lowest TPM/RPM)
deployment = self.get_available_deployment(model=model, input=input, specific_deployment=kwargs.pop("specific_deployment", None)) deployment = self.get_available_deployment(model=model, input=input, specific_deployment=kwargs.pop("specific_deployment", None))
kwargs.setdefault("metadata", {}).update({"deployment": deployment["litellm_params"]["model"]}) kwargs.setdefault("metadata", {}).update({"deployment": deployment["litellm_params"]["model"]})
kwargs["model_info"] = deployment.get("model_info", {})
data = deployment["litellm_params"].copy() data = deployment["litellm_params"].copy()
for k, v in self.default_litellm_params.items(): for k, v in self.default_litellm_params.items():
if k not in data: # prioritize model-specific params > default router params if k not in data: # prioritize model-specific params > default router params
@ -329,6 +332,7 @@ class Router:
deployment = self.get_available_deployment(model=model, input=input, specific_deployment=kwargs.pop("specific_deployment", None)) deployment = self.get_available_deployment(model=model, input=input, specific_deployment=kwargs.pop("specific_deployment", None))
kwargs.setdefault("metadata", {}).update({"deployment": deployment["litellm_params"]["model"]}) kwargs.setdefault("metadata", {}).update({"deployment": deployment["litellm_params"]["model"]})
data = deployment["litellm_params"].copy() data = deployment["litellm_params"].copy()
kwargs["model_info"] = deployment.get("model_info", {})
for k, v in self.default_litellm_params.items(): for k, v in self.default_litellm_params.items():
if k not in data: # prioritize model-specific params > default router params if k not in data: # prioritize model-specific params > default router params
data[k] = v data[k] = v