Releasesby GlyphLangfuse 4.41.0
Features feat: encode v4 native blocks and event bytes in Rust feat(automations): filter prompt events by tags Fixes refactor(normalized-io): enhance additional input handling fix(alerts): link score alerts to filtered scores fix(dashboards): Avoid nested scrollbars for scores widget fix(web): group trace graph steps without recursive stack overflow by...
Also shipped Sep 21, 2026Langfuse in September 2026 →
BigQuery Studio query results pane shows recent run history
The Query results pane now shows a short history of recent runs for a query, including multi-statement queries, without switching to the Job history tab. It is generally available.
The query history system table is now generally available
The system.query.history system table is now generally available. Use it to access account-wide records for queries run on SQL warehouses or serverless compute, including execution status, duration metrics, and query source. See Query history system table reference .
Cortex AI Function Evaluation for measuring quality (Public Preview)
Snowflake Cortex AI Function Evaluation is now available in public preview, enabling customers to measure the output quality, cost, and token usage of custom AI Functions and Cortex AI calls against labeled datasets.
Cortex AI Function Optimization for more efficient AI implementations (Public Preview)
Snowflake Cortex AI Function Optimization is now available in public preview, enabling customers to automatically search across prompts and models for improved implementations of custom AI Functions.
Cortex Agents Compact API (Preview)
The Cortex Agents Compact API is now available in preview. Use the agent:compact endpoint to summarize a conversation and pass the compact representation to subsequent agent:run requests. Compacting conversation history reduces token consumption and helps conversations stay within the model context window.
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes. This feature is generally available ( GA ). It accelerates vector search performance and increases queries per second (QPS) for vector workloads. For more information, see Accelerate queries with the Columnar Engine .
More from LangfuseEvery Langfuse release →
Langfuse 4.40.0
Features feat(trace-batching): link transcript spans to source traces Fixes fix(auth): evict the API key cache after deleting the row Choes refactor(web): thin remaining fat pages into feature Page components docs: remove duplicated 'override' in table columns README chore(agents): use AGENTS.md directly for folder instructions refactor(web): thin leftover...
Langfuse 4.39.0
Features - feat(web): publish event-loop delay for autoscaling - feat(automations): filter prompt events by label - feat(ai-gateway): add phase spans so request traces have no gaps - feat(ai-gateway): record Codex client metadata as agent metadata - feat(auth): authorize every ingestion event through the policy core - feat(otel): simulate worker admission...
Langfuse 4.43.0
Features feat(members): filter project and organization members by role feat(scores): Allow creating scores from annotation sidebar feat(feature-flags): introduce internal view mode and related components feat(evals): route TypeSafe decision-model connections through OpenRouter or Vercel AI Gateway Fixes fix(scim): scope Users/{id} reads and mutations to...
Langfuse 4.42.0
- refactor(web): reroute leftover feature deep imports through index.ts fix(web): trace page panels fill the window width refactor(web): reroute remaining feature deep imports through index.ts feat: trace messages internal preview fix(web): antialiased body text refactor(web): reroute leftover feature deep imports through index.ts fix(web): remove unused...
Langfuse 4.45.0
- fix(web): use tuple key for confusion matrix lookup to avoid delimiter collisions refactor(web): Migrate exports settings table to design system feat(billing): send the organization name on ClickHouse Billing checkout refactor(web): Create design-system StackedBarChart refactor(web): Create design-system AreaChart docs(self-hosting): document Docker log...
Langfuse 4.44.0
Features feat(llm-connections): send extra headers on Vertex AI Gemini requests feat(worker): add isolated OTEL replay integration harness feat(evals): add conversation-signals decision-model gallery template feat(trace-batching): enhance transcript assembly with timing metrics and token estimates feat(ai-gateway): relay Anthropic Messages with usage...
Langfuse 4.46.0
Features - feat(experiments): streamline side-by-side comparison - feat(web): Add formatted json to the dataset item - feat(skills): basic skill management - feat(web): add 100-row page size option to tracing tables Fixes - fix(web): header metrics as plain text on trace, observation and session headers - fix(billing): accept string event-bus timestamps on...
Langfuse 4.38.0
Features - feat(api)!: reject legacy reads for new organizations - feat(ci): deploy AI Gateway automatically to staging - feat(api): attribute legacy GET rejections to clients - feat(web): persistent trace summary strip with tags - feat(filters): metadata key presence operators; coerce legacy empty-substring filters - feat(ai-gateway): isolate caller...