Releasesby IbisLangfuse 4.58.0
UpdateyesterdayUnverifiedAdded Oct 14, 2026
Features - feat(web): require Turnstile on password-reset emails - feat(ci): bump dependabot to opus5.5 - feat(otel): extract media before TypeScript parsing - feat(media): add upload pool with concurrency and byte limits - feat(worker): per-project trace idle time in tracing_readiness_settings - feat(topics): configure Topics on the project's own LLM...
Topics: LLMs, Observability, Developer tools
More Langfuse releases
Every Langfuse releaseLangfuse 4.57.0
Features - feat(ee): refuse FIPS mode with storage options that compute MD5 - feat(worker): add optional startup jitter before queue registration - feat(auth): hash passwords with PBKDF2 in FIPS mode - feat(web): restore optional Cloudflare Turnstile captcha on credential auth - feat(web): carry the typed email over to the password reset form -...
Langfuse 4.56.0
Features - feat(web): design-system CodeSection replaces CodeView - feat(sessions): Add syntax highlighting to session tool previews - feat(billing): send spend alerts for ClickHouse-billed organizations - feat(web): ask new users what they will use Langfuse for - feat(topics): add adaptive map exploration and compact workspace - feat(evals): revamp...
Langfuse 4.55.0
Features - feat(docker): create zero vuln base images and add fips mode - feat(topics): rework the summary system prompt and default facets - feat(web): show ClickHouse organization id in org debug information - feat(topics): summarize all facets of a trace in one call - feat(skills): import skills from local files and ZIPs - feat(skills): add rendered file...
Langfuse 4.54.0
Features - feat(topics): classify traces into buckets - feat: add external media storage integration - feat(scores): compile experiment-item and prompt-event reads through the query AST - feat(openai): add structured review output and enhance customer response fixtures - feat(analytics): track user-visible error toasts - feat(sessions): Make session view...
Langfuse 4.53.0
Features - feat(ai-gateway): expose resolution context in response headers, spans and logs Fixes / Improvements - fix(web): zinc neutral tokens, surface tokens, observation line and fill colours - fix(evals): skip Bedrock URL media interpolation - fix(web): filled badge variant, score and tag chips built on it - fix(web): tab bar labels at 14px with a...
Langfuse 4.52.0
Features - feat(trace-batch): bound batches by estimated weight and exclude oversized traces - feat(web): show the organization usage breakdown on every deployment - feat(auth): remember the last enterprise SSO email on the sign-in form Fixes - fix(llm): drop retired and superseded models from the Vertex AI default list - fix(web): unify card surface across...
Langfuse 4.51.0
Features - feat(mcp): support decision-model evaluators - feat(experiments): annotate directly from comparison grid - feat(audit-logs): log actor info for audited actions and deletions - feat(web): pin evaluator actions during table scrolling - feat(ai-gateway): correlate requests with a gateway request ID - feat(skills): unify file and folder uploads -...
Langfuse 4.50.0
Features - feat(web): show error id in tRPC error toast and include it in issue reports - feat(evals): add facet type, built-in protection, and rule idle time Fixes / Improvements - feat(evals): show seven-day execution health in evaluator status - feat(web): show view owner name or email on avatar hover - fix(web): show full metadata JSONPath values in the...
Also shipped on Oct 13, 2026
Langfuse in October 2026What's included:
- Three training methods: Supervised fine-tuning (SFT) , direct preference optimization (DPO) , and reinforcement learning with group relative policy optimization (GRPO) . Each method trains a low-rank adapter (LoRA) or the full model weights. Custom loss functions : Train with any objective you compute in PyTorch from the model's log-probabilities.
The AI.AGG function is generally available (GA)
The AI.AGG function is generally available (GA). You can use this function to semantically aggregate unstructured input data based on natural language instructions.
BigQuery generative AI functions now support the gemini-4-argon model
BigQuery generative AI functions now support the gemini-4-argon model. This model is generally available .
Gemini 4 Argon in CodeMender (Preview)
CodeMender now supports Gemini 4 Argon. To use it, pass --model gemini-4-argon to cm find , cm verify , or cm fix . Gemini 3.8 Flash remains the default model. For more information, see Specifying the model .
Gemini 4 Argon and Gemini 4 Argon Cyber Permissive
Gemini 4 Argon ( gemini-4-argon ) is available in General Availability (GA) . Gemini 4 Argon is Google's frontier model that excels at deep work in complex, real-world tasks across domains including coding, finance, legal, cybersecurity, and science.
On-demand state repartitioning is now generally available
On-demand state repartitioning is now generally available. You can resize the number of shuffle partitions for a stateful Structured Streaming query or Lakeflow pipeline without losing checkpoint state. See On-demand state repartitioning for stateful streaming queries .