Per-provider model catalog files. Updated weekly by CI.

MLflow 3.13.0 includes several major features and improvements

Major New Features

• 🔐 Role-Based Access Control & Admin UI: A full RBAC system with reusable roles and workspace-scoped grants, plus a new web Admin UI for managing users, roles, and permissions on self-hosted MLflow.
• 🗄️ Trace Retention & Auto Archival: Automatically move aged trace span data out of your SQL backend into object storage (e.g. S3) while keeping every trace fully readable in the UI and APIs.
• 🤖 One-click observability & governance for coding agents: Onboard Claude Code, OpenAI Codex, or Gemini CLI to the AI Gateway in one click for tracing, usage tracking, budgets, and guardrails.
• ✨ New engines for MLflow Assistant: Run MLflow Assistant on a local Ollama model, the OpenAI Codex CLI, or any MLflow AI Gateway endpoint, in addition to Claude Code.
• ☸️ Helm chart for Kubernetes: An official, production-ready Helm chart for deploying the MLflow tracking server to any Kubernetes cluster.
• 🌐 Hermes Agent support: Route the Hermes Agent runtime through the AI Gateway and capture its end-to-end traces in MLflow over OpenTelemetry.
• 🪵 Span log levels: Python-logging-style severity levels on spans, with a "Minimum log level" filter in the trace UI to hide low-level noise.

Breaking Changes

• The permission system has been overhauled into a unified Role-Based Access Control model. The legacy per-resource permission tables, REST endpoints, and client methods are removed and replaced by roles backed by role_permissions, default_permission now acts as a floor rather than an override, and a workspace USE grant is sufficient to create experiments and registered models. Code that relied on the old per-resource permission APIs must migrate to the new role-based APIs. (#22855, #22859, #22941, #23337, #23379, @PattaraS)
• MLServer is no longer available as a pyfunc serving backend. The previously deprecated enable_mlserver option has been removed, so mlflow models serve always uses the built-in scoring server. (#23356, @harupy)
• mlflow autolog claude no longer installs the old Python autolog hook; Claude Code tracing is now provided by the official Claude plugin, which must be installed separately. (#23339, @B-Step62)
• The default optimizer used by judge.align() is now MemAlign, so existing alignment workflows may produce different judges than before unless an optimizer is passed explicitly. (#23254, @veronicalyu320)
• Pointing the tracking or model registry store at a local file-system path now raises an error by default; set MLFLOW_ALLOW_FILE_STORE=true to keep using a file-based store. (#22773, @harupy)

Other Assorted Features & Improvements:

• [Gateway] Support AI Gateway as a backend of MLflow Assistant (#23559, @B-Step62)
• [UI] Make admin pickers target the workspace they're granting into (#23543, @PattaraS)
• [UI] Bring direct-grant picker to parity with role picker (#23420, @PattaraS)
• [UI] Cherry-pick: Add OpenAI Codex CLI as assistant provider (#22566) (#23517, @B-Step62)
• [Tracing] Support settings.local.json for Claude Code tracing config (#23285, @Gkrumbach07)
• [UI] Add coding-agent endpoint creation flow in AI Gateway UI (#23430, @TomeHirata)
• [Tracking / UI] Unified per-user permission APIs: grant / revoke / get / list under /mlflow/users/permissions/* (#23247, @PattaraS)
• [Evaluation / Tracing] Add mlflow.genai.test_agent for automated agent stress-testing (#22990, @serena-ruan)
• [Gateway] Add /gateway/proxy/{endpoint_name}/{path} raw proxy endpoint (#23330, @TomeHirata)
• [UI] Add Ollama as assistant provider (#22098, @SuperSonnix71)
• [Model Registry / Tracking] feat(tracking): Add reader/writer instance routing for database replicas (#22910, @ravidarbha)
• [Tracing / Tracking] Add workspace trace archival configuration plumbing (#22164, @HumairAK)
• [Prompts / Tracking] Promote prompt to a first-class RBAC resource_type (#23248, @PattaraS)
• [Tracing] Publish claude code integration as a plugin and add a setup wizard. (#23218, @B-Step62)
• [UI] Forward Usage tab MetricsFilter metric filters to chart View traces links by translating to Traces page URL DSL (#23239, @aaronteo-db)
• [Tracing] Claude Code Agent Typescript (#20414, @joelrobin18)
• [UI] Adding metric filter component for Experiment Usage tab (#23120, @aaronteo-db)
• [Tracing] Add Link entity and LiveSpan.add_link() for OpenTelemetry Span Links (#22797, @khaledsulayman)
• [Docs / Evaluation] Add Google ADK and third-party scorers (#22299, @debu-sinha)
• [UI] Open /admin to workspace managers (scoped per their workspace) (#23086, @PattaraS)
• [Tracing] Trace Runner.run_streamed() in OpenAI Agents SDK autolog (#22962, @ktrk115)
• [Tracking] Feature/sagemaker build network option (#22996, @pdifranc)
• [Docs / Tracing / UI] Add log levels for Trace Spans with UI switch to filter (#23017, @rrtheonlyone)
• [Tracing / UI] Improve trace page empty state onboarding with setup and code blocks (#22533, @vivian-xie-db)
• [Tracing / Tracking] Add UC traces upsell message for set_experiment calls on Databricks (#23038, @xsh310)
• [Server-infra / UI] [Admin-UI-3/4] Add Platform Admin pages (#22929, @PattaraS)
• [UI] [Admin-UI-2/4] Add /account page and bottom-left account widget (#22973, @PattaraS)
• [Build] Add Helm charts for deploying mlflow to kubernetes cluster (#21973, @WeichenXu123)
• [Tracking] Fix Databricks unified auth support when MLFLOW_ENABLE_DB_SDK=true (#20599, @vb-dbrks)
• [Tracking] [Admin-UI-1/4] Add backend auth endpoints (#22928, @PattaraS)- [Evaluation] Support multiple assessments per trace in MemAlign optimizer (#22846, @veronicalyu320)
• [Docs / Model Registry / Prompts] Include workspace in webhook delivery envelopes when workspaces are enabled (#22873, @copilot-swe-agent)
• [Server-infra] Seed default RBAC roles and grant creator on workspace creation (#22857, @PattaraS)

Bug fixes:

• [UI] Warn on submit with an unsaved direct-grant draft (#23612, @PattaraS)
• [Tracing] Clear archive-now requests for non-archivable leftovers (#23655, @HumairAK)
• [Tracking] Forward MLflow client telemetry from inside Databricks workloads (#23483, @smoorjani)
• [UI] Drop the vestigial directPermissions parallel pass; hide synthetic __user_<id>__ roles on Account/UserDetail (#23578, @PattaraS)
• [Gateway] Tighten response format JSON schema type (#23290, @fenil210)
• [Tracking] fix(tracking): return <console> for mlflow.source.name when sys.argv[0] is empty (#23352, @xodn348)
• [Scoring] Fix UnicodeEncodeError on artifact download with non-ASCII filename (#23241, @1fanwang)
• [Artifacts / UI] Preserve pdfjs-dist bundles in webpack build (craco.config.js) (#23349, @B-Step62)
• [Tracking] Improve misleading DB SDK auth error (#23374, @B-Step62)
• [Tracing] Make mlflow.get_trace V4 retry policy configurable (#23443, @artjen)
• [Tracking] Release _post_import_hooks_lock before firing hooks (#23466, @harupy)
• [Model Registry / Prompts] RBAC: extend prompt resource_type to after-request handlers (#23426, @PattaraS)
• [Evaluation] Surface mlflow version mismatch when deserializing scorers (#23215, @smoorjani)
• [Tracing] Ship compiled dist/ in @mlflow/mlflow-openclaw so openclaw plugins install works (#23220, @B-Step62)
• [Tracing] Fix ended LiveSpan state mutation (#23152, @SahilKumar75)
• [Gateway] Fix AmazonBedrockProvider._build_converse_kwargs tool-call history and validation for Bedrock Converse (#23223, @copilot-swe-agent)
• [UI] fenil-fix: experiment name error (#23199, @fenil210)
• [Tracking] Add workspace isolation on scorers when creating a guardrail (#23115, @mprahl)
• [Evaluation] [Security] Add MLFLOW_ALLOW_PICKLE_DESERIALIZATION guard to PickleEvaluationArtifact (#23183, @TomeHirata)
• [UI] Fix getExperimentNameValidator showing incorrect "deleted state" error for active experiments (#23169, @copilot-swe-agent)
• [Tracking] Fix runs:/<run_id>/<model_name> loading by resolving logged-model artifacts via models:/<model_id> (#23130, @copilot-swe-agent)
• [Gateway] Fix Vertex AI gateway to use Anthropic API format for Claude models (#23175, @TomeHirata)
• [Tracing] Fix invalid stop-hook command when using pixi environment manager (#23030, @copilot-swe-agent)
• [Gateway] Fix MySQL-incompatible NULLS LAST syntax in list_endpoint_guardrail_configs (#23168, @copilot-swe-agent)
• [Evaluation] gateway: honor Anthropic api_base from secret auth_config (#23167, @copilot-swe-agent)
• [Gateway] Fix nested array items being stripped from function tool schemas (#23053, @shyamspr)
• [Tracing / Tracking] Fix OTLP trace ingestion: double-encoded request ID and missing trace tags (#23067, @sairavuri-sudo)
• [Gateway] Add per-image/video/audio pricing to amazon.nova-2-multimodal-embeddings-v1:0 in Bedrock catalog (#23117, @copilot-swe-agent)
• [Evaluation] Skip re-alignment of unchanged traces in MemAlignOptimizer (#23008, @veronicalyu320)
• [Tracing] Fix trace API authorization vulnerability (#23014, @TomeHirata)
• [Models] Fix sentence_transformers pyfunc predict for v5.4+ (#23108, @harupy)
• [Tracing / UI] Fix grouped trace session counts in the UI (#23012, @lavaFreak)
• [Tracing] Fix Azure OpenAI streaming usage tracing (#23036, @Genmin)
• [Tracing] Add session_count trace metric for grouped traces (#23011, @...

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/*, workspace USE permission lets users create experiments and registered models, default roles are seeded on workspace creation, prompt is promoted to a first-class RBAC resource_type, and a new 4-page Admin UI (account widget, /account page, 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.json support. The legacy Python autolog hook for mlflow autolog claude is 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 appVersion wired to the released MLflow image. Get from helm install to a running tracking server without writing your own manifests. (#21973, @WeichenXu123)

• mlflow.genai.test_agent for 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 Link entity via LiveSpan.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

Bump several RC TypeScript packages stable version.

MLflow 3.12.0 includes several major features and improvements

Major New Features

• 🖼️ Multimodal Tracing: Users can now store multimodal content in tracing spans as artifact attachments instead of inline binary data. We've also patched the UI to support the new mlflow-attachment:// style URI, with rich rendering available for PDFs, audio, and images.
• 🤖 Codex, Gemini, Qwen coding agent tracing support: Similar to our Claude Code tracing integration, we've now added support for the Codex, Gemini, and Qwen coding agent platforms as well!
• 🛡️ Gateway guardrails: You can now set guardrails on your gateway endpoints to prevent unsafe or non-compliant model inputs and outputs. Try it out in the MLflow UI!
• ⚡ Trace table pagination: The traces tab is now paginated, rather than fetching all traces up to a limit of 1000. This improves initial load time, and makes the page feel more responsive overall.

Breaking Changes

• [Scoring] Deprecate enable_mlserver in pyfunc serving backend (#22994, @B-Step62)

Other Assorted Features & Improvements:

• [UI] Add coding agents section to AI Gateway quick start (#23006, @TomeHirata)
• [Tracing] feat: record caller in gateway traces from request headers (#22926, @TomeHirata)
• [] Run guardrails on passthrough endpoints; skip response_format for non-chat payloads (#22856, @TomeHirata)
• [] Gateway: preserve client auth header for subscription-based CLI tools (claude-cli, Codex-Desktop, GeminiCLI) (#22915, @TomeHirata)
• [Tracing] Expose Codex and Qwen Code hooks as installable CLI binaries (#22853, @kriscon-db)
• [Tracking / UI] Add Portkey as a supported AI Gateway provider (#22830, @sairavuri-sudo)
• [Tracking] Cache successful basic-auth credential checks to eliminate per-request PBKDF2 (#22817, @PattaraS)
• [] Add last_updated_at field to model catalog entries (#22838, @copilot-swe-agent)
• [Evaluation] Enable third-party scorer registration in OSS MLflow (#22634, @smoorjani)
• [] Add platform-side telemetry fields to Gateway invocation and budget events (#22557, @PattaraS)
• [Tracing] Auto-start SQL warehouse before V4/V5 MLflow tracing calls (#22798, @artjen)
• [Tracing] Migrate Claude Code tracing to TypeScript-based plugin for simpler installation and runtime handling. (#22338, @B-Step62)
• [Tracing] Add TypeScript Qwen Code tracing via Stop hook (#22411, @kriscon-db)
• [] Support model_kwargs in DeepEval scorers for LLM parameter control (#22494, @debu-sinha)
• [] Add Application Default Credentials auth mode for Vertex AI gateway (#22754, @harupy)
• [UI] Reorganize Settings into section-based routing with sub-sidebar navigation (#22743, @serena-ruan)
• [Tracing] Add TypeScript Codex CLI tracing via notify hook (#22410, @kriscon-db)
• [Tracing / UI] Add Share button to trace detail view (#22608, @alkispoly-db)
• [Docs / Models] Add mlflow.diffusers flavor for diffusion model LoRA adapters (#22253, @Rasaboun)
• [Tracing] Add size limit for trace attachments (#22575, @kriscon-db)
• [Tracing] Add GeminiCliTranslator for Gemini CLI OTLP span type mapping (#22409, @kriscon-db)
• [Tracing] Add JSON OTLP encoding support for trace ingestion (#22408, @kriscon-db)
• [Tracing] Extract service.name from OTLP resource attributes for usage telemetry (#22407, @kriscon-db)
• [Tracing] Add MLflow tracing spans to guardrail execution in gateway (#22581, @TomeHirata)
• [Tracing / UI] Add rendering size guards for large media content (#22574, @kriscon-db)
• [Tracking] #21037 Add presigned upload URL endpoint for S3 artifact uploads (#21039, @henishborad)
• [UI] Add Guardrails tab to endpoint editor (#22360, @TomeHirata)
• [UI] Add click-to-expand modal for trace attachment images (#22461, @kriscon-db)
• [UI] Enhance Gateway quick start cards with logos, multi-model options, and compact variant (#22513, @xq-yin)
• [Tracing] Add @mlflow/vercel for better Vercel AI SDK tracing in Databricks UC (#22105, @dbrx-euirim)
• [UI] [UI] Refactor API keys page: bulk delete, inline drawer editing, consistent list pattern (#22485, @xq-yin)
• [] Add Guardrail base class and JudgeGuardrail implementation (#21964, @TomeHirata)
• [Tracing] Add multimodal demo traces with synthetic image and audio data (#22465, @kriscon-db)
• [] [FR] #22217 Add configurable provider allow/deny list to AI Gateway (#22291, @etirelli)
• [UI] [UI] Add bulk duplicate and delete actions to AI Gateway endpoints table (#22404, @xq-yin)
• [UI] Redesign endpoint details model section with unified Model card (#22376, @xq-yin)
• [] Add X-MLflow-Gateway-Duration-Ms and X-MLflow-Gateway-Overhead-Duration-Ms response headers (#22229, @PattaraS)
• [UI] Redesign AI Gateway endpoint details: Overview tab, starter code, and Try in Browser (#22284, @xq-yin)
• [] Add guardrail proto definitions, server handlers, and REST store (#21962, @TomeHirata)
• [Build] Add aiohttp as a core dependency of mlflow (#22189, @TomeHirata)
• [] Add gateway guardrail DB layer: schema, store methods, and tests (#21960, @TomeHirata)
• [UI] Add quick-start template cards to AI Gateway empty state (#22249, @xq-yin)
• [Tracing / UI] Add freeform Notes section to trace assessments pane (#21676, @smoorjani)
• [Tracing / UI] [UI] Improve traces table visual design and UX (#20424, @ridgupta26)
• [UI] Improve Image Grid chart UX for multi-run image comparison (#22118, @Rasaboun)
• [Models] Add base_model_path parameter to save PEFT adapter-only with local base model reference (#22052, @rpathade)
• [Tracking] Add mlflow db move-resources command to move resources between workspaces (#21263, @mprahl)
• [Tracking] feat(tracking): Add composite index on metrics (run_uuid, key, step) (#21727, @ravidarbha)

Bug fixes:

• [Tracing / Tracking] Redact presigned URL credentials in urllib3 retry logs (#22995, @B-Step62)
• [Tracing] Fix assessment logging silently dropped in distributed tracing (#22963, @serena-ruan)
• [] Filter _list_budget_windows results by active workspace when request is workspace-scoped (#22885, @copilot-swe-agent)
• [Models / Scoring] Fix uv custom index URLs omitted from model requirements.txt (#22921, @serena-ruan)
• [Tracing / UI] Set mlflow.message.format on gateway passthrough spans to enable Chat tab (#22916, @TomeHirata)
• [Server-infra] Add RBAC regression tests + gateway validator fix (#22862, @PattaraS)
• [Server-infra] Include role-based grants in list_accessible_workspace_names (#22864, @PattaraS)
• [Tracing] Preserve cache_read tokens in claude_code tracing for cache observability (#22683, @dgokeeffe)
• [Tracing] Fix InferenceTableSpanProcessor init for opentelemetry-sdk 1.41.0 (#22867, @harupy)
• [UI] Fix audio artifact viewer failing in non-default workspaces (#22811, @harupy)
• [Tracking] Fix O(n^2) body buffering in WSGIMiddleware for large uploads (#22729, @harupy)
• [Tracking] Propagate workspace_id as X-Databricks-Org-Id header for SPOG support (#22554, @B-Step62)
• [UI] Fix sessions pagination not auto-loading after virtualizer settles (#22808, @daniellok-db)
• [Tracing] Fix _get_token_usage dropping zero-valued token counts (#22748, @copilot-swe-agent)
• [Tracing] Fix _get_token_usage dropping cache token fields (#22818, @harupy)
• [UI] Remove rendering size guards from artifact image and video views (#22816, @kriscon-db)
• [UI] Disable Create Guardrail in AddGuardrailModal until a Guardrail Model endpoint is selected (#22766, @copilot-swe-agent)
• [UI] Disable Guardrails tab in EditEndpointFormRenderer when endpoint.experiment_id is null, refresh i18n messages, and fix JS type-check follow-ups (#22757, @copilot-swe-agent)
• [] Fix OPENAI_API_KEY requirement when guardrails use gateway:/ judge model (#22769, @TomeHirata)
• [Tracking] Fix ABFSS URI double-encoding in _parse_abfss_uri (#22759, @artjen)
• [UI] Persist immediate gateway usage-tracking updates and preserve dirty edits in useEditEndpointForm (#22734, @copilot-swe-agent)
• [] Fix Vertex AI gateway global location handling for Gemini 3 models (#22696, @harupy)
• [Tracing] Fix ValueError: Circular reference detected crash in dump_span_attribute_value (pydantic_ai autolog) (#22693, @barry3406)
• [Tracing] Add Bedrock Converse models to model catalog (#22515, @ajgassner)
• [Tracing] Fix incorrect cost calculation for Anthropic cached requests (#22516, @ajgassner)
• [UI] Disable Create button when experiment name is empty (#22649, @ssam18)
• [Tracing] Fix AsyncHttpxClientWrapper AttributeError by avoiding deepcopy in TraceJSONEncoder (#22742, @harupy)
• [Evaluation] Fix JSON parsing failure for judge model responses containing newlines (#22736, @harupy)
• [Evaluation] Fix non-ASCII characters being escaped in judge template variables (#22728, @harupy)
• [Tracking] Fix pandas 3.x datetime64 resolution compatibility in cast_df_types_according_to_schema (#22705, @copilot-swe-agent)
• [Tracing] Fix conversation simulator adding wrapper span that breaks custom scorers (#21321, @B-Step62)
• [] Fix Bedrock tool-use chat response parsing (#22677, @lavaFreak)
• [UI] Fix AI Gateway provider dropdown by flattening OpenAI / Azure OpenAI (#22701, @harupy)
• [Evaluation / Tracing] Preserve predict_fn signature in simulation turn tracing (#22610, @rogalski)
• [Tracing / Tracking] fix: Extract cached_tokens in OpenAI streaming responses to correct cost tracking (#22620, @Rishabh-git10)
• [Tracing / UI] Auto-select first endpoint and API key in issue detection modal (#22633, @serena-ruan)
• [Model Registry] Use SQL MAX() in create_model_version instead of loading all rows (#22635, @neolunar7)
• [Tracing] Add modality/tool pricing extraction to dev/update_model_catalog.py, backfill anthropic.json / gemini.json, and skip reasoning modality pricing (#22699, @copilot-swe-agent)
• [Tracing] Fix pydantic-ai run_stream_sync autologging broken by AgentSpec forwar...

We're excited to announce MLflow 3.12.0rc0, which brings powerful new capabilities for agent developers:

Major New Features:

• Automatic Tracing for more AI Coding Assistants: First-class tracing for Claude Code, Codex, Qwen Code, and Gemini CLI, now distributed as standalone TypeScript-based plugins installable as CLI binaries. Drop one into your CLI of choice and capture every prompt, tool call, and turn out-of-the-box. (#22338, #22410, #22411, #22409, #22853, @B-Step62, @kriscon-db)

• OpenClaw Tracing Plugin: New TypeScript-based tracing plugin for OpenClaw. Install via npm and get full automatic tracing of OpenClaw sessions, no SDK changes required. Captures every prompt, tool call, and response across the session, allowing you to inspect agent behavior directly in the MLflow UI. (#22717)

• AI Gateway Guardrails: Add safety checks to any agent endpoint with the new Guardrail base class and JudgeGuardrail implementation. Configure pre-LLM and post-LLM guardrails directly from the Gateway endpoint editor, with full DB persistence, REST API, and gateway-level execution. Protect production agents from unsafe inputs and outputs without writing wrapper code. Tracing spans for guardrail execution are included. (#21964, #21960, #21962, #22306, #22360, #22577, #22581, #22767, @TomeHirata)

• Multimodal Trace Attachments: Trace agents that handle images, audio, and files, and view them directly in the MLflow UI. Click-to-expand image modals, inline rendering in span Details and Timeline views, audio playback, multi-part chat normalization for Gemini and OpenAI Responses API, and size guards for large media so the UI stays responsive. (#22465, #22461, #22451, #22462, #22466, #22460, #22449, #22450, #22574, #22575, #21783, @kriscon-db)

• mlflow.diffusers Flavor: New first-class flavor for diffusion models. Save and serve LoRA adapters for image-generation pipelines with the same mlflow.<flavor>.log_model ergonomics as PyTorch or Transformers. (#22253, @Rasaboun)

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==3.12.0rc0`

Release candidate for @mlflow/vercel TypeScript package with version 0.2.0: #22105

MLflow 3.11.1 includes several major features and improvements.

Major New Features:

• 🔍 Automatic Issue Identification: Automatically identify quality issues in your agent with AI! Use the new "Detect Issues" button in the traces table to analyze selected traces and surface potential problems across categories like correctness, safety, and performance. Issues are linked directly to traces for easy investigation and debugging. Docs (#21431, #21204, #21165, #21163, #21161, @smoorjani, @serena-ruan)
• 💰 Gateway Budget Alerts & Limits: Control your AI Gateway spending with configurable budget policies! Set spending limits by time window (daily, weekly, or monthly), receive alerts before hitting limits, and prevent runaway costs with automatic request blocking. The new budget management UI lets you track spending, configure webhooks for notifications, and monitor violations across all your gateway endpoints. Docs (#21116, #21534, #21569, #21473, #21108, @TomeHirata, @copilot-swe-agent)
• 📊 Trace Graph View: Visualize complex trace hierarchies with an interactive graph view! Navigate multi-level trace structures, understand parent-child relationships at a glance, and debug complex systems more effectively with a visual representation of your trace topology. Docs (#20607, @joelrobin18)
• 🌐 Native OpenTelemetry GenAI Convention Support: MLflow now natively supports the OpenTelemetry GenAI Semantic Conventions for trace export! When exporting traces via OTLP with MLFLOW_ENABLE_OTEL_GENAI_SEMCONV enabled, MLflow automatically translates them to follow the OTel GenAI semantic conventions, enabling seamless integration with OTel-compatible observability platforms while preserving GenAI-specific metadata. Docs (#21494, #21495, @B-Step62)
• 🔧 OpenCode Tracing Integration: Debug smarter with OpenCode CLI integration! Track and analyze code execution flows directly from your development workflow, making it easier to identify performance bottlenecks and trace issues back to specific code paths. Docs (#20133, @joelrobin18)
• ⚡ Native UV Support for Model Dependencies: Automatic dependency inference now supports UV! MLflow automatically detects UV projects and captures exact, locked dependencies from your lockfile when logging models, ensuring reproducible environments. Docs (#20344, #20935, @debu-sinha)
• 🔒 Pickle-Free Model Serialization: Enhance security with pickle-free model formats! MLflow now supports safer model serialization using torch.export and skops formats, with improved controls when MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False. Comprehensive documentation guides you through migrating existing models to pickle-free formats for production deployments. Docs (#21404, #21188, #20774, @WeichenXu123)

Breaking Changes:

• ⚠️ TypeScript SDK Package Renaming: The MLflow TypeScript SDK packages have been renamed to use npm organization scoping. If you're using the TypeScript SDK, update your package.json dependencies and import statements: mlflow-tracing → @mlflow/core, mlflow-openai → @mlflow/openai, mlflow-anthropic → @mlflow/anthropic, mlflow-gemini → @mlflow/gemini. All packages are now at version 0.2.0. (#20792, @B-Step62)
• Remove MLFLOW_ENABLE_INCREMENTAL_SPAN_EXPORT environment variable (#22182, @PattaraS)
• Remove litellm and gepa from genai extras (#22059, @TomeHirata)
• Block / and : in Registered Model names (#21458, @Bhuvan-08)

Features:

• [Evaluation] Allow MetaPromptOptimizer to work without litellm (#22233, @TomeHirata)
• [Tracking] Update Databricks API calls to use new gRPC APIs instead of py4j APIs (#22205, @WeichenXu123)
• [Build] Add aiohttp as a core dependency of mlflow (#22189, @TomeHirata)
• [Evaluation] Extend _get_provider_instance with groq, deepseek, xai, openrouter, ollama, databricks, vertex_ai (#22148, @kriscon-db)
• [UI] Move native providers to non-LiteLLM in gateway UI (#22203, @TomeHirata)
• [Tracing / Tracking] Add trace_location parameter to create_experiment (#22075, @dbrx-euirim)
• [Gateway] Complete Bedrock provider with Converse API support (#21999, @TomeHirata)
• [Gateway] Add native Vertex AI gateway provider (#21998, @TomeHirata)
• [Gateway] Add native Databricks gateway provider (#21997, @TomeHirata)
• [Gateway] Add native Ollama gateway provider (#21995, @TomeHirata)
• [Gateway] Add native xAI (Grok) gateway provider (#21993, @TomeHirata)
• [Tracing] Use bulk upsert in log_spans() to eliminate per-span ORM overhead (#21954, @harupy)
• [Tracing] Add builtin cost_per_token to remove litellm dependency for cost tracking (#22046, @TomeHirata)
• [Evaluation] Remove LiteLLM hard dependency from the discovery pipeline and judge adapters (#21739, @harupy)
• [Evaluation] Add pipelined predict-score execution for mlflow.genai.evaluate (#20940, @alkispoly-db)
• [Tracing / Tracking] Default trace location table_prefix to experiment ID in set_experiment (#21815, @danielseong1)
• [Tracking] Add default uvicorn log config with timestamps (#21838, @harupy)
• [Tracing / UI] Add Session ID filter to GenAI traces table filter dropdown (#21794, @daniellok-db)
• [Evaluation / UI] Add Default Credential Chain auth mode for Bedrock/SageMaker in AI Gateway (#21061, @timsolovev)
• [UI] Add multi metric bar chart support (#21258, @RenzoMXD)
• [Tracking] Add TCP keepalive to HTTP sessions to detect stale connections and reduce timeout hangs (#21514, @mobaniha)
• [Evaluation] Add proxy URL support for make_judge (#21185, @yukimori)
• [UI] Improve run group filter to use grouping criteria instead of run IDs (#21072, @daniellok-db)
• [UI] Add tool selector to Tool Calls charts and fix dark mode/sizing (#20865, @B-Step62)
• [UI] Graph View Traces + OpenAI (#20607, @joelrobin18)
• [UI] Show run description in chart tooltip (#21580, @KaushalVachhani)
• [Evaluation / Tracing / UI] Add bulk judge execution from traces table toolbar with status feedback (#21270, @PattaraS)
• [Gateway] Add Redis-backed BudgetTracker for distributed gateway deployments (#21504, @TomeHirata)
• [Tracing / Tracking] Add trace location param to set_experiment (#21385, @danielseong1)
• [Build / Tracking] Add azure extra for Azure Blob Storage support in full Docker image (#21582, @harupy)
• [UI] Add budget violation indicator to gateway budget list page (#21569, @copilot-swe-agent)
• [Evaluation] [5/5] Add discover_issues() pipeline and public API (#21431, @smoorjani)
• [UI] Add Structured Output (JSON Schema) Support to the MLflow Prompts UI (#21394, @kennyvoo)
• [Tracing] Auto-inject tracing context headers in autologging (#21490, @TomeHirata)
• [UI] Add budget alert webhooks UI and fix budgets table borders (#21534, @TomeHirata)
• [Model Registry / Prompts / UI] Add webhooks management UI to settings page (#21483, @TomeHirata)
• [Tracing] Opencode CLI (#20133, @joelrobin18)
• [Models] Add uv_groups and uv_extras params for uv dependency group control (#20935, @debu-sinha)
• [Tracing] Add GenAI Semantic Convention translator for OTLP trace export (#21494, @B-Step62)
• [Tracking] Add polars dataset support to autologging (#21507, @harupy)
• [Tracing] Add mlflow.tracing.context() API for injecting metadata/tags without wrapper spans (#21318, @B-Step62)
• [UI] Add budget dates and current spending for gateway budgets (#21473, @TomeHirata)
• [Tracing / UI] Improve DSPy trace chat view readability (#21296, @B-Step62)
• [UI] Add Kubernetes request auth provider plugin (#21176, @HumairAK)
• [Tracking] Add IS NULL/IS NOT NULL support for tags and params in search_runs (#21283, @TomeHirata)
• [Tracing / UI] Display clickable gateway trace link in trace explorer (#21316, @TomeHirata)
• [UI] Add session selection support with checkbox, actions, and row alignment (#21324, @B-Step62)
• [Models] Add UV package manager support for automatic dependency inference (#20344, @debu-sinha)
• [Evaluation / UI] Add feature flag to control evaluation runs issues panel visibility (#21406, @serena-ruan)
• [Tracing / UI] Add cached tokens display to Token Usage chart (#21295, @TomeHirata)
• [UI] Add budget policies management UI for AI Gateway (#21116, @TomeHirata)
• [UI] Allow multiple judge selection in Run judge on trace modal (#21322, @B-Step62)
• [Docs / Tracking] Add admin-only authorization to webhook CRUD operations (#21271, @TomeHirata)
• [Evaluation / Tracking] Add SqlIssue database table for storing experiment issues (#21165, @serena-ruan)
• [Model Registry / Prompts] Support search_prompt_versions in OSS SQLAlchemy store (#21315, @TomeHirata)
• [Evaluation / Tracing / UI] Add issue detection button to traces table toolbar with feature flag (#21204, @serena-ruan)
• [Docs / Tracing / UI] Add inline audio player for input_audio content parts in trace UI (#21302, @TomeHirata)
• [Evaluation / Tracing] Add IssueReference assessment type to store issue links with traces (#21163, @serena-ruan)
• [Evaluation / Tracing] Add issue management protos with create, update, get, and search APIs (#21161, @serena-ruan)
• [UI] Add IS NULL/IS NOT NULL operators for trace tags in search UI (#21280, @TomeHirata)
• [Docs / Tracing] Add IS NULL/IS NOT NULL support for trace tags in search_traces (#21277, @TomeHirata)
• [Tracing] Add steer message tracing support for Claude Code (#21265, @harupy)
• [Models / Tracking] Add support for transformers 5.x (#20728, @KUrushi)
• [Gateway] Add WEE...

Stripped third-party dependencies from evaluation and AI Gateway features, replacing external provider routing with built-in implementations.

We're excited to announce MLflow 3.11.0rc0, which includes several notable updates:

Major New Features:

• 🔍 Automatic Issue Identification: Automatically identify quality issues in your agent with AI! Use the new "Detect Issues" button in the traces table to analyze selected traces and surface potential problems across categories like correctness, safety, and performance. Issues are linked directly to traces for easy investigation and debugging. (#21431, #21204, #21165, #21163, #21161, @smoorjani, @serena-ruan)
• 💰 Gateway Budget Alerts & Limits: Control your AI Gateway spending with configurable budget policies! Set spending limits by time window (daily, weekly, or monthly), receive alerts before hitting limits, and prevent runaway costs with automatic request blocking. The new budget management UI lets you track spending, configure webhooks for notifications, and monitor violations across all your gateway endpoints. (#21116, #21534, #21569, #21473, #21108, @TomeHirata, @copilot-swe-agent)
• 📊 Trace Graph View: Visualize complex trace hierarchies with an interactive graph view! Navigate multi-level trace structures, understand parent-child relationships at a glance, and debug complex systems more effectively with a visual representation of your trace topology. (#20607, @joelrobin18)
• 🌐 Native OpenTelemetry GenAI Convention Support: MLflow now natively supports the OpenTelemetry GenAI Semantic Conventions for trace export! When exporting traces via OTLP with MLFLOW_ENABLE_OTEL_GENAI_SEMCONV enabled, MLflow automatically translates them to follow the OTel GenAI semantic conventions, enabling seamless integration with OTel-compatible observability platforms while preserving GenAI-specific metadata. (#21494, #21495, @B-Step62)
• 🔧 Opencode Tracing Integration: Debug smarter with Opencode CLI integration! Track and analyze code execution flows directly from your development workflow, making it easier to identify performance bottlenecks and trace issues back to specific code paths. (#20133, @joelrobin18)
• ⚡ UV Package Manager Support: Automatic dependency inference now supports UV! MLflow automatically detects UV projects and captures exact, locked dependencies from your lockfile when logging models, ensuring reproducible environments. (#20344, #20935, @debu-sinha)
• 🔒 Pickle-Free Model Serialization: Enhance security with pickle-free model formats! MLflow now supports safer model serialization using torch.export and skops formats, with improved controls when MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False. Comprehensive documentation guides you through migrating existing models to pickle-free formats for production deployments. (#21404, #21188, #20774, @WeichenXu123)

Breaking Changes:

• ⚠️ TypeScript SDK Package Renaming: The MLflow TypeScript SDK packages have been renamed to use npm organization scoping. If you're using the TypeScript SDK, update your package.json dependencies and import statements: mlflow-tracing → @mlflow/core, mlflow-openai → @mlflow/openai, mlflow-anthropic → @mlflow/anthropic, mlflow-gemini → @mlflow/gemini. All packages are now at version 0.2.0. (#20792, @B-Step62)

Stay tuned for the full release, which will be packed with even more features and bugfixes.

To try out this release candidate, please run:

pip install mlflow==3.11.0rc0