Skip to content

fix: add support for capturing usage details with ChatBedRock #1267

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open

fix: add support for capturing usage details with ChatBedRock #1267

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
a834039
fix: add support for capturing usage details with ChatBedRock
vignesh-pattalam-zoomrx Jul 21, 2025
d17dd95
Merge branch 'main' into fix-chatbedrock-usage-details
pbvignesh Jul 25, 2025
9923873
Merge branch 'main' into fix-chatbedrock-usage-details
pbvignesh Jul 26, 2025
d601c83
Merge branch 'main' into fix-chatbedrock-usage-details
pbvignesh Jul 29, 2025
ef12a9b
Merge branch 'main' into fix-chatbedrock-usage-details
pbvignesh Jul 29, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 13 additions & 4 deletions langfuse/langchain/CallbackHandler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -821,6 +821,9 @@ def _parse_usage_model(usage: typing.Union[pydantic.BaseModel, dict]) -> Any:
("input_tokens", "input"),
("output_tokens", "output"),
("total_tokens", "total"),
# ChatBedrock API follows a separate format compared to ChatBedrockConverse API
("prompt_tokens", "input"),
("completion_tokens", "output"),
# https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/get-token-count
("prompt_token_count", "input"),
("candidates_token_count", "output"),
Expand All @@ -837,10 +840,16 @@ def _parse_usage_model(usage: typing.Union[pydantic.BaseModel, dict]) -> Any:
usage_model = cast(Dict, usage.copy()) # Copy all existing key-value pairs

# Skip OpenAI usage types as they are handled server side
if not all(
openai_key in usage_model
for openai_key in ["prompt_tokens", "completion_tokens", "total_tokens"]
):
openai_keys = [
"prompt_tokens",
"completion_tokens",
"total_tokens",
# We add the below two since OpenAI keys look similar to ChatBedRock
# keys except for the below two keys
"completion_token_details",
"prompt_token_details",
]
if not all(openai_key in usage_model for openai_key in openai_keys):
for model_key, langfuse_key in conversion_list:
if model_key in usage_model:
captured_count = usage_model.pop(model_key)
Expand Down