✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
analytics
autogen
evaluation
langchain
large-language-models
llama-index
+12 more
llm
llm-evaluation
llm-observability
llmops
monitoring
observability
openai
playground
prompt-engineering
prompt-management
self-hosted
ycombinator
Summary
AI summaryUpdates Chores, trace, and web across a mixed release.
Full changelog
What's Changed
Features
- feat(ingestion): persist SDK attribution by @maxdeichmann in https://github.com/langfuse/langfuse/pull/14593
- feat(trace-ui): collapse tool calls when there are many by @nimarb in https://github.com/langfuse/langfuse/pull/14718
- feat(trace): custom graph renderer + trace playback (LFE-10577) by @nkabardin in https://github.com/langfuse/langfuse/pull/14659
Fixes
- fix(web): open evaluator delete confirmation as centered dialog by @wochinge in https://github.com/langfuse/langfuse/pull/14729
- fix(trace): decode unicode escapes in remaining JSON viewers by @toyayuto in https://github.com/langfuse/langfuse/pull/14668
- fix(worker): raise ClickHouse data export timeout default to 60 minutes (LFE-10648) by @niklassemmler in https://github.com/langfuse/langfuse/pull/14753
Chores
- ci: Setup knip for eslint-plugin by @bezbac in https://github.com/langfuse/langfuse/pull/14734
- chore(deps): Remove unused
bracesdependency by @bezbac in https://github.com/langfuse/langfuse/pull/14752 - refactor(web): move experiments table selection into zustand store by @wochinge in https://github.com/langfuse/langfuse/pull/14755
Full Changelog: https://github.com/langfuse/langfuse/compare/v3.203.3...v3.204.0
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About langfuse
🪢 Open source LLM engineering platform: LLM Observability, metrics, evals, prompt management, playground, datasets. Integrates with OpenTelemetry, Langchain, OpenAI SDK, LiteLLM, and more. YC W23
Related context
Related tools
Beta — feedback welcome: [email protected]