langchain-ai/langsmith-sdk
1
0
Fork 0
mirror of https://github.com/langchain-ai/langsmith-sdk.git synced 2026-07-21 01:35:25 -04:00
Code Issues 125 Packages Projects Releases 16 Wiki Activity

[PR #2076] [CLOSED] fix: support response API stream object parse #2056

New Issue
Closed
opened 2026-02-15 23:16:21 -05:00 by yindo · 0 comments
yindo commented 2026-02-15 23:16:21 -05:00
Owner
Copy Link
Copy Source

📋 Pull Request Information

Original PR: https://github.com/langchain-ai/langsmith-sdk/pull/2076
Author: @EugeneJinXin
Created: 10/15/2025
Status: ❌ Closed

Base: main ← Head: xin/fix-openai-wrapper-streaming-parser

📝 Commits (1)

  • 3ab8ea7 fix: support response API stream object parse

📊 Changes

1 file changed (+18 additions, -2 deletions)

View changed files

📝 python/langsmith/wrappers/_openai.py (+18 -2)

📄 Description

OpenAI Streaming Response Serialization Bug
When using OpenAI's streaming APIs (both chat.completions.create(stream=True) and responses.create(stream=True))
with LangSmith tracing via wrap_openai(), the trace output was being serialized as a string representation of the Stream object: "<openai.Stream object at 0x113355010>" instead of the actual response data.

The stream isn't really a pydantic model and won't have model_dump(), so Exception is caught, and {"output": response} is returned, When serialized to JSON for LangSmith, Python converts the Stream object to its string representation: "<openai.Stream object at 0x...>"

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/langchain-ai/langsmith-sdk/pull/2076 **Author:** [@EugeneJinXin](https://github.com/EugeneJinXin) **Created:** 10/15/2025 **Status:** ❌ Closed **Base:** `main` ← **Head:** `xin/fix-openai-wrapper-streaming-parser` --- ### 📝 Commits (1) - [`3ab8ea7`](https://github.com/langchain-ai/langsmith-sdk/commit/3ab8ea79cbeb539ebe2d7a4fb5cf909ddb8e5a81) fix: support response API stream object parse ### 📊 Changes **1 file changed** (+18 additions, -2 deletions) <details> <summary>View changed files</summary> 📝 `python/langsmith/wrappers/_openai.py` (+18 -2) </details> ### 📄 Description OpenAI Streaming Response Serialization Bug When using OpenAI's streaming APIs (both chat.completions.create(stream=True) and responses.create(stream=True)) with LangSmith tracing via wrap_openai(), the trace output was being serialized as a string representation of the Stream object: "<openai.Stream object at 0x113355010>" instead of the actual response data. The stream isn't really a pydantic model and won't have model_dump(), so Exception is caught, and {"output": response} is returned, When serialized to JSON for LangSmith, Python converts the Stream object to its string representation: "<openai.Stream object at 0x...>" --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
yindo added the pull-request label 2026-02-15 23:16:21 -05:00
yindo closed this issue 2026-02-15 23:16:21 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Annotation Queue
Bulk Export
Datasets
Evaluators
Experiments
Feature
Migrated
Perf
Playground
Polish
Prompt Hub
TypeScript
User Feature Request
backend
bug
documentation
enhancement
evaluation
frontend
goalie
insights
integrations
integrations
javascript
multi-modal
needs-design
needs-investigation
needs-scoping
nit
opentelemetry
oss
otel
pull-request
Mirrored from GitHub Pull Request
 python
sdk
token tracking
tracing + observability
ui
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/langsmith-sdk#2056
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?