fix(core): correct Langfuse import path to fix tracing issue #1805
+276
−10
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…/openai' - Change import from 'langfuse' to '@langfuse/openai' package - Update documentation to reflect correct package installation - Fix issue #1677 where Langfuse tracing was not working The observeOpenAI function is exported from @langfuse/openai package, not the base langfuse package. This fixes the integration issue where users couldn't see traces in Langfuse dashboard.
✅ Deploy Preview for midscene ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
- Automatically initialize OpenTelemetry SDK when Langfuse is enabled - Remove manual initialization requirement from documentation - Add error handling with clear installation instructions - Track initialization state to prevent duplicate initialization This simplifies the Langfuse setup process - users no longer need to manually initialize OpenTelemetry in their application entry file.
yuyutaotao
approved these changes
Jan 20, 2026
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
This PR fixes issue #1677 where Langfuse LLM call tracing was not working.
Root Causes Fixed
1. Incorrect Import Path
The code was importing
observeOpenAIfrom the wrong package:import { observeOpenAI } from 'langfuse'import { observeOpenAI } from '@langfuse/openai'According to Langfuse's official documentation, the OpenAI SDK integration should use the
@langfuse/openaipackage, not the baselangfusepackage.2. Missing OpenTelemetry Initialization
Langfuse's tracing is built on top of OpenTelemetry, which requires initialization before it can work. Previously, users had to manually add this initialization code to their application entry file, which was easy to miss.
Changes
1. Fixed Import Path (
packages/core/src/ai-model/service-caller/index.ts:188)'langfuse'to'@langfuse/openai'2. Auto-Initialize OpenTelemetry (
packages/core/src/ai-model/service-caller/index.ts:190-214)3. Documentation Updates
apps/site/docs/zh/model-config.mdxapps/site/docs/en/model-config.mdxnpm install @langfuse/openai @langfuse/otel @opentelemetry/sdk-nodeBenefits
Testing
For Users Affected by #1677
If you're experiencing Langfuse tracing issues:
Update packages:
Remove manual initialization (if you added it):
NodeSDKorLangfuseSpanProcessorinitialization code from your applicationRestart your application
You should see these logs:
Langfuse tracing should now work correctly!
Fixes #1677