Skip to content

langfuse

v3.209.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 17d Tracing
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ 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 summary

Updates Chores, web, and blobstorage across a mixed release.

Changes in this release

Feature Low

feat(web): revamp dark theme with surface elevation, text hierarchy, semantic status colors

feat(web): revamp dark theme with surface elevation, text hierarchy, semantic status colors

Source: llm_adapter@2026-07-16

Confidence: high

Feature Low

feat(web): add system table view preset category chips

feat(web): add system table view preset category chips

Source: llm_adapter@2026-07-16

Confidence: high

Feature Low

feat(blobstorage): disable export on deterministic customer‑config faults

feat(blobstorage): disable export on deterministic customer‑config faults

Source: llm_adapter@2026-07-16

Confidence: high

Dependency Low

chore(fetch-llm-completion): add remaining AI SDK adapters

chore(fetch-llm-completion): add remaining AI SDK adapters

Source: llm_adapter@2026-07-16

Confidence: high

Bugfix Medium

fix(evals): stop offering new trace‑level evals to v4‑toggle cohort on Cloud

fix(evals): stop offering new trace‑level evals to v4‑toggle cohort on Cloud

Source: llm_adapter@2026-07-16

Confidence: high

Bugfix Medium

fix(datasets): show empty state for filtered run compare view

fix(datasets): show empty state for filtered run compare view

Source: llm_adapter@2026-07-16

Confidence: high

Bugfix Medium

fix(otel): map input/output from gen_ai.client.inference.operation.details span events

fix(otel): map input/output from gen_ai.client.inference.operation.details span events

Source: llm_adapter@2026-07-16

Confidence: high

Refactor Low

refactor(events): update SDK metadata extraction for OTel compatibility

refactor(events): update SDK metadata extraction for OTel compatibility

Source: llm_adapter@2026-07-16

Confidence: high

Refactor Low

chore(shared): document ms‑aligned start_time assumption in v2 observations cursor (LFE-10405)

chore(shared): document ms‑aligned start_time assumption in v2 observations cursor (LFE-10405)

Source: llm_adapter@2026-07-16

Confidence: high

Other Low

chore: pipeline blob storage cleanup deletes and batch tombstone inserts

chore: pipeline blob storage cleanup deletes and batch tombstone inserts

Source: llm_adapter@2026-07-16

Confidence: high

Full changelog

What's Changed

Features

  • feat(blobstorage): disable export on deterministic customer-config faults by @niklassemmler in https://github.com/langfuse/langfuse/pull/14628
  • feat(web): dark theme revamp — surface elevation, text hierarchy, semantic status colors by @nmtrang29 in https://github.com/langfuse/langfuse/pull/14873
  • feat(web): system table view preset category chips by @nmtrang29 in https://github.com/langfuse/langfuse/pull/14774

Fixes

  • fix(evals): stop offering new trace-level evals to v4-toggle cohort on Cloud by @marliessophie in https://github.com/langfuse/langfuse/pull/14924
  • fix(datasets): show empty state for filtered run compare by @marliessophie in https://github.com/langfuse/langfuse/pull/14893
  • fix(otel): map input/output from gen_ai.client.inference.operation.details span events by @hassiebp in https://github.com/langfuse/langfuse/pull/14930

Chores

  • chore: pipeline blob storage cleanup deletes and batch tombstone inserts by @sumerman in https://github.com/langfuse/langfuse/pull/14910
  • docs(readme): add hiring message by @jannikmaierhoefer in https://github.com/langfuse/langfuse/pull/14919
  • refactor(events): update SDK metadata extraction for OTel compatibility by @marliessophie in https://github.com/langfuse/langfuse/pull/14917
  • chore(fetch-llm-completion): add remaining AI SDK adapters by @hassiebp in https://github.com/langfuse/langfuse/pull/14854
  • chore(shared): document ms-aligned start_time assumption in v2 observations cursor (LFE-10405) by @niklassemmler in https://github.com/langfuse/langfuse/pull/14926

Full Changelog: https://github.com/langfuse/langfuse/compare/v3.208.0...v3.209.0

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

Track langfuse

Get notified when new releases ship.

Sign up free

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

All releases →

Related context

Beta — feedback welcome: [email protected]