highperfocused/danswer
1
0
Fork 0
mirror of https://github.com/danswer-ai/danswer.git synced 2025-04-01 00:18:18 +02:00
Code Issues Packages Projects Releases Wiki Activity

Handle models like openai/bedrock/claude-3.5-... (#2869)

Browse Source 
* Handle models like openai/bedrock/claude-3.5-...

* Fix log statement
This commit is contained in:
Chris Weaver 2024-10-21 22:27:26 -07:00 committed by GitHub
parent e4779c29a7
commit 6e9b6a1075
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 18 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
backend/danswer/llm/utils.py
View File

@ -342,12 +342,26 @@ def get_llm_max_tokens(
try:
model_obj = model_map.get(f"{model_provider}/{model_name}")
if not model_obj:
model_obj = model_map[model_name]
logger.debug(f"Using model object for {model_name}")
else:
if model_obj:
logger.debug(f"Using model object for {model_provider}/{model_name}")
if not model_obj:
model_obj = model_map.get(model_name)
if model_obj:
logger.debug(f"Using model object for {model_name}")
if not model_obj:
model_name_split = model_name.split("/")
if len(model_name_split) > 1:
model_obj = model_map.get(model_name_split[1])
if model_obj:
logger.debug(f"Using model object for {model_name_split[1]}")
if not model_obj:
raise RuntimeError(
f"No litellm entry found for {model_provider}/{model_name}"
)
if "max_input_tokens" in model_obj:
max_tokens = model_obj["max_input_tokens"]
logger.info(