---
type: "post"
slug: "langfuse-v4-cli-and-evaluator-templates"
title: "Langfuse v4, CLI and evaluator templates"
description: "Langfuse v4 makes every LLM call, tool execution and agent step directly queryable, with table loads dropping from seconds to milliseconds and a hard Cloud cutover in November."
retrieval_nugget: "Langfuse v4 makes every LLM call, tool execution and agent step directly queryable, with table loads dropping from seconds to milliseconds and a hard Cloud cutover in November."
published_at: "2026-08-31"
updated_at: "2026-09-12"
record_date: "2026-08-17"
date_kind: "published_at"
topics: ["agent-integrations","agents","ai"]
entities: ["Langfuse"]
source_url: "https://langfuse.com/changelog/2026-08-17-langfuse-v4"
source_title: "Langfuse v4 is live: faster at scale, with more ways to search, monitor, and evaluate - Langfuse"
source_domain: "langfuse.com"
source_terms: ["Langfuse","evaluator","templates","faster","Cloud"]
summary_word_count: 199
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:langfuse-v4-cli-and-evaluator-templates"
status: "published"
source_urls: ["https://langfuse.com/changelog/2026-08-17-langfuse-v4"]
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:6bae11d52029569242772551e5fbc7f94b714f49c2c216871ae18d64292ad717","reviewed_at":"2026-09-12T10:00:00Z","source_evidence_count":1,"verified_claim_count":2,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":2,"implication_count":1,"watch_condition_count":1,"related_record_count":3}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Langfuse v4 went live on Langfuse Cloud and generally available for self-hosted deployments on 17 August.","observed_facts":[{"text":"Langfuse v4 is live on Langfuse Cloud and generally available for self-hosted deployments.","source_urls":["https://langfuse.com/changelog/2026-08-17-langfuse-v4"]},{"text":"Langfuse Cloud becomes v4-only on 16 November 2026, with a Migration Assistant listing per-project actions.","source_urls":["https://langfuse.com/changelog/2026-08-17-langfuse-v4"]}],"mechanism":"every LLM call, tool execution and agent step becomes directly queryable and evaluable, rather than reachable only through a trace view: initial table loads over large datasets drop from seconds to milliseconds, and dashboards over longer time ranges load at least ten times faster in large projects","why_now":"Langfuse Cloud becomes v4-only on 16 November 2026; most projects need no migration, but a Migration Assistant surfaces per-project actions that must be completed before that date, and self-hosted operators are told to plan the upgrade and review infrastructure requirements first.","implications":["Making the agent step the queryable unit is what turns observability into a diagnostic rather than a record"],"evidence_boundary":"The performance figures are the vendor's own and stated as ranges for large projects rather than a published benchmark.","watch_conditions":["The condition to watch is whether step-level querying makes evaluator results comparable across versions, because faster dashboards over the same ambiguous step boundaries would improve latency without improving diagnosis."],"related_records":[{"url":"https://newruntime.com/posts/evals-move-from-leaderboards-to-operating-diagnostics","relation":"Prior coverage of evaluation becoming an operating diagnostic."},{"url":"https://newruntime.com/posts/agent-operations-are-converging-on-connectors-memory-observability-dashboard","relation":"Prior coverage of the agent operations control plane."},{"url":"https://newruntime.com/topics/agents","relation":"Topic hub for agent observability."}]}
routes: {"html":"https://newruntime.com/posts/langfuse-v4-cli-and-evaluator-templates/","markdown":"https://newruntime.com/posts/langfuse-v4-cli-and-evaluator-templates.md","json":"https://newruntime.com/posts/langfuse-v4-cli-and-evaluator-templates.json"}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
---

# Langfuse v4, CLI and evaluator templates

## Retrieval answer

Langfuse v4 makes every LLM call, tool execution and agent step directly queryable, with table loads dropping from seconds to milliseconds and a hard Cloud cutover in November.

Langfuse v4 went live on Langfuse Cloud and generally available for self-hosted deployments on 17 August. The change is that every LLM call, tool execution and agent step becomes directly queryable and evaluable, rather than reachable only through a trace view: initial table loads over large datasets drop from seconds to milliseconds, and dashboards over longer time ranges load at least ten times faster in large projects.

There is a date attached, which is the part teams will feel. Langfuse Cloud becomes v4-only on 16 November 2026; most projects need no migration, but a Migration Assistant surfaces per-project actions that must be completed before that date, and self-hosted operators are told to plan the upgrade and review infrastructure requirements first.

Making the agent step the queryable unit is what turns observability into a diagnostic rather than a record, which is the transition described in [evals moving from leaderboards to operating diagnostics](https://newruntime.com/posts/evals-move-from-leaderboards-to-operating-diagnostics/) and part of the control plane in [agent operations converging on connectors, memory and observability](https://newruntime.com/posts/agent-operations-are-converging-on-connectors-memory-observability-dashboard/). The [agents hub](https://newruntime.com/topics/agents/) tracks that layer.

The performance figures are the vendor's own and stated as ranges for large projects rather than a published benchmark. The condition to watch is whether step-level querying makes evaluator results comparable across versions, because faster dashboards over the same ambiguous step boundaries would improve latency without improving diagnosis.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Agent Integrations](https://newruntime.com/topics/agent-integrations.md): Explore the agent-integrations topic hub.
- topic hub: [Agents](https://newruntime.com/topics/agents.md): Explore the agents topic hub.
- topic hub: [Ai](https://newruntime.com/topics/ai.md): Explore the ai topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)