- Major overhaul of Role-Based Access Control with unified permission APIs, workspace USE permission for creating experiments and registered models, prompt promoted to first-class RBAC resource_type, and new Admin UI with account widget and Platform Admin pages
- Claude Code, OpenClaw, Ollama, and OpenAI Codex integrated as first-class assistant providers in the AI Gateway with Claude Code TypeScript plugin and setup wizard
- End-to-end trace archival across the tracking stack with archival configuration models, OTLP and artifact helpers, and archive-aware retrieval fallback
- Helm charts for deploying MLflow to Kubernetes clusters with production-ready configuration, ingress, and persistence
- mlflow.genai.test_agent API for automated agent stress-testing with adversarial input generation and trace review
- OpenTelemetry Span Links support via LiveSpan.add_link() to connect causally related spans across traces
- Database replica routing in SQL tracking store for reader/writer instance routing to enable horizontal scaling
- Legacy per-resource permission tables collapsed into unified role_permissions table
We're excited to announce MLflow 3.13.0rc0, which deepens agent observability, tightens permissions, and broadens deployment options:
Major New Features:
-
RBAC + Admin UI: Major overhaul of MLflow's Role-Based Access Control — legacy per-resource permission tables collapsed into
role_permissions, unified per-user permission APIs under/mlflow/users/permissions/*, workspaceUSEpermission lets users create experiments and registered models, default roles are seeded on workspace creation,promptis promoted to a first-class RBACresource_type, and a new 4-page Admin UI (account widget,/accountpage, Platform Admin pages, backend auth endpoints) opens to workspace managers scoped per their workspace. (#22855, #22857, #22859, #22928, #22929, #22941, #22973, #23086, #23247, #23248, #23337, #23379, @PattaraS) -
Coding-Agent Tracing as Plugins: Claude Code, OpenClaw, Ollama, and OpenAI Codex are now wired into the AI Gateway as first-class assistant providers, plus a Claude Code TypeScript plugin with a setup wizard and
settings.local.jsonsupport. The legacy Python autolog hook formlflow autolog claudeis replaced by the new official plugin, and a coding-agent endpoint creation flow is now available directly in the AI Gateway UI. (#20414, #22098, #22566, #22717, #23218, #23285, #23339, #23430, #23517, @B-Step62, @joelrobin18, @Gkrumbach07, @SuperSonnix71, @TomeHirata) -
Trace Archival: End-to-end trace archival across the tracking stack. Includes archival configuration models, OTLP and artifact helpers, SQLAlchemy archival passes, archive-aware retrieval fallback, plus workspace/experiment/server-level archival settings in the UI. Read archived traces back seamlessly. (#23359, @mprahl)
-
Helm Charts for Kubernetes Deployment: First-class Helm chart for deploying MLflow to Kubernetes clusters — production-ready configuration, ingress, persistence, and
appVersionwired to the released MLflow image. Get fromhelm installto a running tracking server without writing your own manifests. (#21973, @WeichenXu123) -
mlflow.genai.test_agentfor Automated Agent Stress-Testing: New API for stress-testing GenAI agents — generate adversarial inputs, replay them through your agent, and review the resulting traces in MLflow. Wires into the existing evaluation flow and assessment APIs. (#22990, @serena-ruan) -
OpenTelemetry Span Links: Tracing now supports the OpenTelemetry
Linkentity viaLiveSpan.add_link(), letting you connect causally related spans across traces. (#22797, @khaledsulayman) -
Database Replica Routing: The SQL tracking store now supports reader/writer instance routing for database replicas, so read-heavy MLflow deployments can scale horizontally without overloading the primary. (#22910, @ravidarbha)
Stay tuned for the full release, which will include even more features and bug fixes.
To try out this release candidate, please run:
pip install mlflow==3.13.0rc0