NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
PyPI · #639 most downloaded on PyPI
MLflow is an open source platform for the complete machine learning lifecycle
Last release 5 days ago
24 Sep 2026
Ships fairly regularly
a new release about every 2 weeks
Nearly every release is documented
notes for 57 of the last 60 stable releases
6 versions withdrawn
withdrawn after publishing
8 years old
189 releases · first in 2018
[Tracing] Fix trace API authorization vulnerability ( #23014 , @TomeHirata )
MLflow 3.13.0 includes several major features and improvements
logging-style severity levels on spans, with a "Minimum log level" filter in the trace UI to hide low-level noise.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)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)judge.align() is now MemAlign, so existing alignment workflows may produce different judges than before unless an optimizer is passed explicitly. (#23254, @veronicalyu320)MLFLOW_ALLOW_FILE_STORE=true to keep using a file-based store. (#22773, @harupy)grant / revoke / get / list under /mlflow/users/permissions/* (#23247, @PattaraS)mlflow.genai.test_agent for automated agent stress-testing (#22990, @serena-ruan)prompt to a first-class RBAC resource_type (#23248, @PattaraS)Link entity and LiveSpan.add_link() for OpenTelemetry Span Links (#22797, @khaledsulayman)/admin to workspace managers (scoped per their workspace) (#23086, @PattaraS)Runner.run_streamed() in OpenAI Agents SDK autolog (#22962, @ktrk115)workspace in webhook delivery envelopes when workspaces are enabled (#22873, @copilot-swe-agent)Bug fixes:
directPermissions parallel pass; hide synthetic __user_<id>__ roles on Account/UserDetail (#23578, @PattaraS)<console> for mlflow.source.name when sys.argv[0] is empty (#23352, @xodn348)craco.config.js) (#23349, @B-Step62)mlflow.get_trace V4 retry policy configurable (#23443, @artjen)_post_import_hooks_lock before firing hooks (#23466, @harupy)prompt resource_type to after-request handlers (#23426, @PattaraS)dist/ in @mlflow/mlflow-openclaw so openclaw plugins install works (#23220, @B-Step62)LiveSpan state mutation (#23152, @SahilKumar75)AmazonBedrockProvider._build_converse_kwargs tool-call history and validation for Bedrock Converse (#23223, @copilot-swe-agent)MLFLOW_ALLOW_PICKLE_DESERIALIZATION guard to PickleEvaluationArtifact (#23183, @TomeHirata)getExperimentNameValidator showing incorrect "deleted state" error for active experiments (#23169, @copilot-swe-agent)runs:/<run_id>/<model_name> loading by resolving logged-model artifacts via models:/<model_id> (#23130, @copilot-swe-agent)pixi environment manager (#23030, @copilot-swe-agent)NULLS LAST syntax in list_endpoint_guardrail_configs (#23168, @copilot-swe-agent)gateway: honor Anthropic api_base from secret auth_config (#23167, @copilot-swe-agent)amazon.nova-2-multimodal-embeddings-v1:0 in Bedrock catalog (#23117, @copilot-swe-agent)MemAlignOptimizer (#23008, @veronicalyu320)sentence_transformers pyfunc predict for v5.4+ (#23108, @harupy)session_count trace metric for grouped traces (#23011, @lavaFreak)gateway_adapter not forwarding workspace header to judge endpoints (#23047, @sairavuri-sudo)TaskContext.artifactDir to get the correct unpacked artifacts directory (#22969, @WeichenXu123)MLFLOW_SKIP_PIP_REQUIREMENTS_CHECK env var to bypass pip validation in air-gapped environments (#22920, @copilot-swe-agent)ScrollableTooltip when many data series are shown (#22917, @copilot-swe-agent)delete_user FK constraint failure when user has dependent rows (#22922, @PattaraS)@mlflow/claude-code TypeScript plugin for cache observability (#22906, @dgokeeffe)CreateBudgetPolicyModal submit (#22903, @PattaraS)Documentation updates:
codex.mdx to use Codex openai_base_url config (CLI + ~/.codex/config.toml) (#23272, @copilot-swe-agent)ANTHROPIC_BASE_URL example in Claude Code gateway docs (#23269, @copilot-swe-agent)basic-http-auth.mdx (#23198, @copilot-swe-agent)Small bug fixes and documentation updates:
#23637, #23405, #23181, #23673, #23659, #23293, #23440, #23441, #23127, #23182, #23179, #23180, #22992, @B-Step62; #23557, @AayushShah-904; #23594, #23592, #23583, #23496, #23417, #23415, #23414, #23413, #23412, #23410, #23409, #23408, #23407, #23406, #23399, #23398, #23452, #22861, #22933, #22888, @PattaraS;
Note truncated.
One column per quarter.
We're excited to announce MLflow 3.13.0rc0, which deepens agent observability, tightens permissions, and broadens deployment options:
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
[Scoring] Deprecate enable_mlserver in pyfunc serving backend (#22994, @B-Step62)
MLflow 3.12.0 includes several major features and improvements
enable_mlserver in pyfunc serving backend (#22994, @B-Step62)response_format for non-chat payloads (#22856, @TomeHirata)claude-cli, Codex-Desktop, GeminiCLI) (#22915, @TomeHirata)last_updated_at field to model catalog entries (#22838, @copilot-swe-agent)model_kwargs in DeepEval scorers for LLM parameter control (#22494, @debu-sinha)notify hook (#22410, @kriscon-db)mlflow.diffusers flavor for diffusion model LoRA adapters (#22253, @Rasaboun)GeminiCliTranslator for Gemini CLI OTLP span type mapping (#22409, @kriscon-db)service.name from OTLP resource attributes for usage telemetry (#22407, @kriscon-db)Guardrail base class and JudgeGuardrail implementation (#21964, @TomeHirata)X-MLflow-Gateway-Duration-Ms and X-MLflow-Gateway-Overhead-Duration-Ms response headers (#22229, @PattaraS)aiohttp as a core dependency of mlflow (#22189, @TomeHirata)base_model_path parameter to save PEFT adapter-only with local base model reference (#22052, @rpathade)mlflow db move-resources command to move resources between workspaces (#21263, @mprahl)Bug fixes:
_list_budget_windows results by active workspace when request is workspace-scoped (#22885, @copilot-swe-agent)requirements.txt (#22921, @serena-ruan)mlflow.message.format on gateway passthrough spans to enable Chat tab (#22916, @TomeHirata)list_accessible_workspace_names (#22864, @PattaraS)InferenceTableSpanProcessor init for opentelemetry-sdk 1.41.0 (#22867, @harupy)WSGIMiddleware for large uploads (#22729, @harupy)workspace_id as X-Databricks-Org-Id header for SPOG support (#22554, @B-Step62)_get_token_usage dropping zero-valued token counts (#22748, @copilot-swe-agent)_get_token_usage dropping cache token fields (#22818, @harupy)Create Guardrail in AddGuardrailModal until a Guardrail Model endpoint is selected (#22766, @copilot-swe-agent)Guardrails tab in EditEndpointFormRenderer when endpoint.experiment_id is null, refresh i18n messages, and fix JS type-check follow-ups (#22757, @copilot-swe-agent)_parse_abfss_uri (#22759, @artjen)useEditEndpointForm (#22734, @copilot-swe-agent)global location handling for Gemini 3 models (#22696, @harupy)ValueError: Circular reference detected crash in dump_span_attribute_value (pydantic_ai autolog) (#22693, @barry3406)AsyncHttpxClientWrapper AttributeError by avoiding deepcopy in TraceJSONEncoder (#22742, @harupy)datetime64 resolution compatibility in cast_df_types_according_to_schema (#22705, @copilot-swe-agent)predict_fn signature in simulation turn tracing (#22610, @rogalski)cached_tokens in OpenAI streaming responses to correct cost tracking (#22620, @Rishabh-git10)MAX() in create_model_version instead of loading all rows (#22635, @neolunar7)dev/update_model_catalog.py, backfill anthropic.json / gemini.json, and skip reasoning modality pricing (#22699, @copilot-swe-agent)run_stream_sync autologging broken by AgentSpec forward reference in _returns_sync_streamed_result (#22666, @copilot-swe-agent)spans.content scan (#22433, @harupy)StringDtype mismatch in cast_df_types_according_to_schema (#22537, @copilot-swe-agent)@mlflow.trace within @trace_disabled context (#22501, @harupy)input_file content type in Responses API chat rendering (#22466, @kriscon-db)inline_data extraction (#22453, @kriscon-db)contents in Gemini chat input normalization (#22455, @kriscon-db)role field is omitted (#22454, @kriscon-db)search_traces (#22431, @harupy)copy_model_version failure caused by get_logged_model call (#22262, @TomeHirata)rename_experiment in SqlAlchemyStore to enforce 500-char name limit (#22418, @copilot-swe-agent)mp4 from AUDIO_EXTENSIONS to fix video artifact preview (#22350, @copilot-swe-agent)MetaPromptOptimizer failing on prompts with no template variables (#22301, @alkispoly-db)PromptsListFilters (#22321, @copilot-swe-agent)make_judge with bool/numeric types not reporting aggregated metrics in evaluate() (#22302, @alkispoly-db)start_trace when merging traces with existing metrics (#22257, @alkispoly-db)_all_tables_exist to derive expected_tables from Base.metadata dynamically (#22128, @copilot-swe-agent)T | None as top-level feedback_value_type in make_judge (#22201, @copilot-swe-agent)--static-prefix not applied to /api/ REST routes (#22159, @TomeHirata)Documentation updates:
Pre-LLM / Post-LLM across guardrail_utils.py, Gateway API, UI, docs, and guardrailValidation.ts (#22767, @copilot-swe-agent)EvaluationDataset type references in GenAI datasets docs (#22761, @harupy)docs/docs/genai/governance/ai-gateway/legacy/* and redirect legacy AI Gateway routes to ai-gateway/index (#22579, @copilot-swe-agent)invocation() function example in manual tracing docs (#22524, @prithvipal)uv run --frozen flag for offline/no-network usage in CLAUDE.md (#22505, @copilot-swe-agent)Small bug fixes and documentation updates:
#22993, #22978, #21482, #21321, #22559, #21319, #22483, #22346, #22091, @B-Step62; #22919, #22925, #22854, #22821, #22786, #22820, #22615, #22819, #22781, #22703, #22565, #22200, #22193, @serena-ruan; #22869, #22863, #22843, #22839, #22837, #22815, #22765, #22762, #22749, #22708, #22486, #22481, #22471, #22401, #22366, #22278, #22281, @harupy; #22848, #22775, #22764, #22763, #22772, #22527, #22435, #22358, #22357, #22356, #22355, #22261, #22259, #22202, #22234, #22243, @TomeHirata; #22824, @vinh412; #22805, #22804, #22803, #22802, #22801, #22800, #22447, #22632, #22735, #22697, #22691, #22508, #22628, #22567, #22482, #22441, #22273, #22272, #22190, #22434, #22154, #22147, #22144, #22268, #22241, #22187, #22074, @daniellok-db; #22557, #22722, #22807, #22721, #22558, #22589, #22319, #22145, #21789, @PattaraS; #22825, #22783, #22789, #22787, #22784, #22782, #22757, #22771, #22770, #22758, #22756, #22755, #22744, #22737, #22715, #22718, #22714, #22713, #22711, #22710, #22707, #22704, #22700, #22692, #22686, #22690, #22689, #22688, #22684, #22685, #22682, #22670, #22669, #22668, #22667, #22664, #22663, #22662, #22655, #22657, #22659, #22601, #22580, #22578, #22576, #22563, #22550, #22549, #22548, #22547, #22545, #22544, #22541, #22540, #22539, #22538, #22511, #22532, #22529, #22528, #22526, #22523, #22525, #22519, #22518, #22517, #22514, #22512, #22509, #22510, #22506, #22504, #22490, #22488, #22480, #22478, #22477, #22470, #22468, #22443, #22438, #22436, #21203, #22417, #22403, #22413, #22399, #22395, #22391, #22392, #22388, #22379, #22348, #22347, #22343, #22341, #22342, #22340, #22339, #22337, #22330, #22329, #22328, #22316, #22309, #22305, #22250, #22207, #22204, #22198, #22177, #22174, @copilot-swe-agent; #22795, @rollyjoel; #22629, #22730, #22614, #22573, #22334, #22502, #22448, #22196, #22263, #22186, #22185, #22184, #21956, #22183, #22246, #21955, @kriscon-db; #22627, #22238, @xsh310; #22295, #22212, @smoorjani; #22463, #22370, #22365, #22240, #22226, @xq-yin; #22473, #22439, #22437, @WeichenXu123; #22387, #21419, @mprahl; #22382, #22373, @SeldonAi; #21848, @alkispoly-db; #22160, @amotl
We're excited to announce MLflow 3.12.0rc0, which brings powerful new capabilities for agent developers:
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`
[Scoring] Fix tar path traversal vulnerability in extract_archive_to_dir (#21824, @TomeHirata)
MLflow 3.11.1 includes several major features and improvements.
Major New Features:
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)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:
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)MLFLOW_ENABLE_INCREMENTAL_SPAN_EXPORT environment variable (#22182, @PattaraS)litellm and gepa from genai extras (#22059, @TomeHirata)/ and : in Registered Model names (#21458, @Bhuvan-08)Features:
MetaPromptOptimizer to work without litellm (#22233, @TomeHirata)aiohttp as a core dependency of mlflow (#22189, @TomeHirata)_get_provider_instance with groq, deepseek, xai, openrouter, ollama, databricks, vertex_ai (#22148, @kriscon-db)log_spans() to eliminate per-span ORM overhead (#21954, @harupy)cost_per_token to remove litellm dependency for cost tracking (#22046, @TomeHirata)table_prefix to experiment ID in set_experiment (#21815, @danielseong1)SqlIssue database table for storing experiment issues (#21165, @serena-ruan)b2://) (#20731, @jeronimodeleon)Bug fixes:
DatabricksProvider to use OpenAI-compatible endpoint URLs (#22393, @TomeHirata)InferenceTableSpanProcessor alongside DatabricksUCTableSpanProcessor in model serving (#22332)" (#22362, @smurching)UCSchemaLocation destination is set in Databricks model serving (trace: null) (#22332, @smurching)tool_reference content blocks in Anthropic Chat UI parser (#22331, @B-Step62)uv version requirement from 0.5.0 to 0.6.10 (#22313, @copilot-swe-agent)use_dbconnect_artifact path in spark_udf (#22300, @franciffu723)get_provider_name() to align with model_prices_and_context_window.json (#22223, @TomeHirata)log_image with slash-containing keys: replace # with ~ as path separator (#22172, @copilot-swe-agent)_call_llm_via_gateway to handle gateway:/ URIs (#22153, @TomeHirata)polars_dataset.py to fix import failure with polars<1 (#22085, @TomeHirata)huey_consumer.py path resolution when venv bin dir is not on PATH (#22126, @copilot-swe-agent)model_provider in calculate_cost_by_model_and_token_usage (#22134, @TomeHirata)"Discarded unknown message" log in Anthropic gateway provider (#21942, @copilot-swe-agent)inference_params for built-in scorers (#21943, @debu-sinha)rate_limit_event from Claude Code CLI (#22067, @forrestmurray-db)NextMethod() S3 dispatch error in R mlflow_get_run_context (#21957, @daniellok-db)extract_archive_to_dir (#21824, @TomeHirata)mlflow/pyfunc/scoring_server/__init__.py (#21908, @copilot-swe-agent)warnings.showwarning handler (#21707, @mango766)Sequence instead of list in to_chat_completions_input (#21724, @mr-brobot)UV_PROJECT_ENVIRONMENT in run_uv_sync to install into the correct Python environment (#21750, @copilot-swe-agent)build-system in examples/uv-dependency-management/pyproject.toml (#21752, @copilot-swe-agent)batch_get_traces (#21650, @harupy)_issue_discovery_judge feedback from traces UI (#21648, @harupy)register_prompt tag behavior (#21600, @yangbaechu)_maybe_save_model for Databricks ACL-protected artifact URIs (#21602, @mohammadsubhani)MLFLOW_ALLOW_PICKLE_DESERIALIZATION is disabled (#21404, @WeichenXu123)SpanType.TOOL (#21552, @LeviLong01)run_config from span attributes (#21454, @MarkVasile)_load_pyfunc (#21480, @copilot-swe-agent)% separator (#21269, @harupy)uv run in Claude Code tracing hooks (#21327, @copilot-swe-agent)env key for Claude Code settings environment variables (#21344, @smoorjani)_deduplicate_requirements merging marker-differentiated requirements (#21098, @harupy)ExperimentListTable (#20908, @joelrobin18)Documentation updates:
uv dependency management vs MLFLOW_LOCK_MODEL_DEPENDENCIES, add uv workspace limitation (#22312, @copilot-swe-agent)--exclude-newer (#22133, @copilot-swe-agent)MLFLOW_ENABLE_ASYNC_TRACE_LOGGING docs to reflect OSS default behavior (#21731, @copilot-swe-agent)generate_signature_output in favor of input_example (#21556, @shivamshinde123).mcp.json or CLI (#21609, @copilot-swe-agent)Small bug fixes and documentation updates:
#22377, #22258, #22260, #22259, #21988, #22000, #21994, #21992, #21991, #21990, #21989, #21986, #21918, #21919, #20739, #21753, #21784, #21785, #21786, #21653, #21647, #21558, #21572, #21567, #21571, #21540, #21544, #21542, #21120, #21114, #21112, #21198, #21111, #21289, #21110, #20743, #21109, #20960, #21125, #21124, #22369, #22261, #22234, #22220 @TomeHirata; #22370, #22277, @xq-yin; #22346, #22311, #22091, #21930, #22235, #22232, #22086, #22008, #21975, #21866, #21940, #21920, #21931, #21820, #21830, #21825, #21810, #21788, #21712, #21620, #21702, #21479, #21495, #21506, #21377, #21010, @B-Step62; #22166, @SomtochiUmeh; #22283, #22083, #22200, #22248, #21317, #22242, #22173, #22239, #22193, #22197, #22224, #22225, #22222, #22138, #22175, #22176, #21932, #22053, #22039, #21860, #21268, #21863, #21833, #21917, #21864, #21804, #21854, #21803, #21840, #21837, #21802, #21831, #21835, #21797, #21821, #21758, #21793, #21747, #21796, #21746, #21718, #21756, #21787, #21745, #21741, #21734, #21719, #21715, #21713, #21716, #21661, #21701, #21407, #21589, #21655, #21703, #21664, #21700, #21663, #21662, #21623, #21622, #21619, #21618, #21546, #21654, #21597, #21625, #21596, #21595, #21594, #21613, #21593, #21592, #21591, #21590, #21588, #21579, #21578, #21577, #21575, #21536, #21531, #21510, #21509, #21492, #21500, #21363, #21499, #21498, #21532, #21497, #21502, #21491, #21361, #21409, #21360, #21408, #21437, #21189, #21187, #21167, #21398, #21209, #21208, #21166, #21207, #21286, #21284, #21367, #21366, #21365, #21206, #21164, #21162, #21297, #21303, #21261, #21090, @serena-ruan; #22179, #22191, #22178, #22007, #22004, #21852, #21799, #21832, #21759, #21717, #21657, #21624, #21541, @kriscon-db; #22374, #21503, #20795, #21372, @WeichenXu123; #22192, #22139, #22136, #22108, #22107, #22104, #22103, #22099, #22094, #22093, #22090, #22087, #22079, #22065, #22064, #22058, #22062, #22051, #22048, #22044, #22035, #22026, #22025, #22021, #21965, #21961, #21959, #21952, #21937, #21936, #21926, #21924, #21916, #21914, #21912, #21911, #21901, #21900, #21897, #21896, #21894, #21891, #21888, #21887, #21885, #21880, #21879, #21878, #21876, #21877, #21875, #21868, #21862, #21861, #21859, #21834, #21808, #21822, #21807, #21766, #21782, #21761, #21757, #21742, #21740, #21737, #21733, #21730, #21729, #21728, #21710, #21694, #21683, #21684, #21677, #21675, #21672, #21671, #21670, #21652, #21651, #21634, #21629, #21627, #21621, #21610, #21543, #21529, #21527, #21523, #21513, #21511, #21508, #21501, #21496, #21486, #21485, #21481, #21477, #21474, #21472, #21471, #21464, #21462, #21457, #21459, #21456, #21455, #21452, #21451, #21449, #21448, #21442, #21441, #21415, #21411, #21402, #21397, #21375, #21369, #21330, #21353, #21335, #21331, #21328, #21285, #21251, #21239, #21235, #21229, #21228, #21224, #21210, #21153, #21147, #21139, #21122, #21113, #21117, #21095, #21091, #21092, #21089, #21088, #21056, #21047, #21044, #21043, #21042, #21040, #21031, #21032, #21024, #21021, #21018, #21011, #21016, #20997, #20994, #20985, #20982, #20981, #20978, #20967, #20966, #20942, #20921, #20916, #20911, #20905, #20909, #20906, #20823, #20820, #20811, #20810, #20788, #20798, #20786, #20754, #20745, #20735, #20724, #20722, #20721, @copilot-swe-agent; #22140, #22180, #22171, #22014, #22009, #22005, #22001, #21921, #21858, #21780, #21640, #21612, #21643, #21563, #21537, #21525, #21275, #21145, #21358, #21272, #21068, #21066, #20789, #21026, #20790, @daniellok-db; #22131, @amotl; #22080, #22056, #22019, #22017, #22018, #22015, #22016, #22013, #22011, #22002, #21996, #21985, #21984, #21981, #21971, #21939, #21838, #21806, #21798, #21748, #21666, #21665, #21642, #21637, #21566, #21539, #21405, #21484, #21439, #21440, #21395, #21195, #21194, #21150, #21029, #21014, #20987, #20738, @harupy; #22115, @sebneira; #21641, #21632, #21614, @dbrx-euirim; #21849, #20939, #20938, #20937, @alkispoly-db; #21846, #21764, #21678, #21644, #21636, #21342, #21430, #21429, #21428, #21446, #21427, #21336, #21339, #21344, #21345, #21033, @smoorjani; #21853, @bbqiu; #21827, @kevin-lyn; #21350, #21777, #21696, #21691, #21515, #21488, #21386, @dbczumar; #21528, #21673, #21549, @debu-sinha; #21658, #21600, #21493, #21393, @yangbaechu; #21660, #21639, #21443, #21444, @PattaraS; #21581, @kennyvoo; #21219, @mprahl; #21424, #21232, #21184, #21178, @danielseong1; #21173, @nananosirova; #20388, @mdalvz0000; #21080, @ManasVardhan
Nothing published for this version
Stripped third-party dependencies from evaluation and AI Gateway features, replacing external provider routing with built-in implementations.
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:
We're excited to announce MLflow 3.11.0rc0, which includes several notable updates:
Major New Features:
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)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:
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
MLflow 3.10.1 is a patch release that contains some minor feature enhancements, bug fixes, and documentation updates.
MLflow 3.10.1 is a patch release that contains some minor feature enhancements, bug fixes, and documentation updates.
Features:
Bug fixes:
default_permission=NO_PERMISSIONS (#21220, @copilot-swe-agent)HttpArtifactRepository (#12955, @Koenkk)logged_model tables experiment_id foreign keys (#20185, @harupy)Documentation updates:
Small bug fixes and documentation updates:
#20740, #21148, #21149, #21096, @TomeHirata; #21368, #21118, @B-Step62; #21384, #21345, #21236, #21106, #21033, #21115, #21034, @smoorjani; #21326, #21133, #21036, @copilot-swe-agent; #21293, @daniellok-db; #21175, @caponetto; #21305, #21264, @serena-ruan; #21216, @justinwei-db; #21038, #21082, @bbqiu; #21143, #20733, @mprahl; #20488, @mdalvz0000; #21142, @EPgg92; #21094, @PattaraS
[Build] Validate webhook url to fix SSRF vulnerability (#20747, @TomeHirata)
We're excited to announce MLflow 3.10.0, which includes several notable updates:
Major New Features:
🏢 Organization Support in MLflow Tracking Server: MLflow now supports multi-workspace environments. Users can organize experiments, models, prompts, with a coarser level of unit and logically isolate them in a single tracking server. (#20702, #20657, @mprahl, @Gkrumbach07, @B-Step62)
💬 Multi-turn Evaluation & Conversation Simulation: MLflow now supports multi-turn evaluation, including evaluating existing conversations with session-level scorers and simulating conversations to test new versions of your agent, without the toil of regenerating conversations. Use the session-level scorers introduced in MLflow 3.8.0 and the brand new session UIs to evaluate the quality of your conversational agents and enable automatic scoring to monitor quality as traces are ingested. (#20243, #20377, #20289, @smoorjani)
💰 Trace Cost Tracking: Gain visibility into your LLM spending! MLflow now automatically extracts model information from LLM spans and calculates costs, with a new UI that renders model and cost data directly in your trace views. (#20327, #20330, @serena-ruan)
🎯 Navigation bar redesign: We've redesigned the navigation to provide a frictionless experience. A new workflow type selector in the top-level navbar lets you quickly switch between GenAI and Classical ML contexts, with streamlined sidebars that reduce visual clutter. (#20158, #20160, #20161, #20699, @ispoljari, @daniellok-db)
🎮 MLflow Demo Experiment: New to MLflow GenAI? With one click, launch a pre-populated demo and explore tracing, evaluation, and prompt management in action. No configuration, no code required. (#19994, #19995, #20046, #20047, #20048, #20162, @BenWilson2)
📊 Gateway Usage Tracking: Monitor your AI Gateway endpoints with detailed usage analytics. A new usage page shows request patterns and metrics, with trace ingestion that links gateway calls back to your experiments for end-to-end observability. (#20357, #20358, #20642, @TomeHirata)
⚡ In-UI Trace Evaluation: Users can now run custom or pre-built LLM judges directly from the traces and sessions UI. This enables quick evaluation of individual traces and individual without context switching to the python SDK. (#20360, @hubertzub-db, @danielseong1)
Features:
migrate-filestore command (#20615, @harupy)# clint: disable= comments (#20651, @copilot-swe-agent)virtualenv with python -m venv in virtualenv env_manager path (#20640, @copilot-swe-agent)sampling_ratio_override parameter to @mlflow.trace (#19784, @harupy)mlflow datasets list CLI command (#20167, @alkispoly-db)Bug fixes:
SearchTraces V2 request body on ENDPOINT_NOT_FOUND fallback (#20963, @brendanmaguire)MLFLOW_GATEWAY_RESOLVE_API_KEY_FROM_FILE flag to prevent local file inclusion in API gateway (#20965, @TomeHirata)RestException crash on null error_code and incorrect except clause (#20903, @copilot-swe-agent)experiment_id type error in gateway config resolver (#20764, @copilot-swe-agent)mlflow.pyfunc.log_model with loader_module param (#20727, @WeichenXu123)mlflow demo URL to use experiment ID instead of name (#20678, @copilot-swe-agent)_validate_max_retries and _validate_backoff_factor (#20597, @vb-dbrks)_search_runs and handle max_results=None (#20547, @copilot-swe-agent)Documentation updates:
mlflow migrate-filestore command (#20616, @harupy)use-prompts-in-apps.mdx (#20661, @PattaraS)use-prompts-in-apps.mdx with a section for prompt linking under traced method (#20593, @PattaraS)ToolAgent name formatting in ag2 documentation and examples (#20470, @Umakanth555)Small bug fixes and documentation updates:
#20959, #20915, #20986, #20956, #20912, #20955, #20943, #20919, #20776, #20826, #20781, #20767, #20761, #20760, #20763, #20762, #20687, #20746, #20682, #20667, #20658, #20578, #20559, #20495, #20497, @TomeHirata; #21006, #20980, #20707, #20777, @bbqiu; #20950, #21008, #20877, #20822, #20817, #20813, #20816, #20796, #20815, #20765, #20716, #20689, #20744, #20690, #20451, #20502, #20252, #20314, #20210, @B-Step62; #21000, #20975, #20806, #20449, #20686, #20603, #20573, #20572, #20584, #20551, #20526, #20550, #20523, #20525, #20453, #20478, #20452, #20438, #20474, #20460, #20457, #20459, #20456, #20444, #20418, #20285, #20284, #20283, #20282, #20281, #20280, #20051, @smoorjani; #21005, #21007, #20880, #20857, #20802, #20779, #20717, #20713, #20714, #20692, #20693, #20683, #20675, #20665, #20674, #20673, #20663, #20662, #20659, #20652, #20649, #20650, #20647, #20646, #20641, #20638, #20635, #20634, #20633, #20626, #20625, #20621, #20619, #20618, #20617, #20606, #20564, #20581, #20570, #20568, #20566, #20558, #20560, #20543, #20554, #20537, #20536, #20532, #20530, #20528, #20512, #20505, #20501, #20498, #20496, #20491, #20490, #20489, #20487, #20486, #20484, #20483, #20482, #20441, #20436, #20427, #20417, #20400, #20399, #20397, #20395, #20396, #20391, #20342, #20341, #20332, #20326, #20316, #20315, #20305, #20300, #20299, #20297, #20293, #20268, #20262, #20260, #20251, #20250, #20244, #20235, #20228, #20227, #20226, #20220, #20202, #20186, #20172, #20152, #20150, #19984, #20102, #20098, #20095, #20093, #20094, #20091, #20090, #20089, #20088, #20087, #20086, #20085, #20084, #20083, #20082, #20081, #20080, #20077, #20076, #20075, #20070, #20067, #20069, #20020, #20026, @copilot-swe-agent; #20793, #20791, #20768, @WeichenXu123; #20979, #20701, #20609, #20608, #20569, #20535, #20481, #20318, #20224, #20149, #20119, #20068, #20014, #20016, #20019, @harupy; #20973, @Gkrumbach07; #21003, #20936, #20730, #20041, #20381, @xsh310; #20989, #20830, #20766, #20759, #20758, #20757, #20756, #20699, #20697, #20696, #20695, #20694, #20255, #20254, #20253, #20248, #20247, #20010, #20009, #19999, #19998, #19976, #19975, #19974, #19973, #19971, @daniellok-db; #20976, @aravind-segu; #20725, #20339, #20565, #20660, #20455, #20440, #20404, #20403, #20402, #20567, #20542, #20541, #20540, #20557, #20503, #20506, #20500, #20499, #20467, #20338, #20337, #20331, #20462, #20329, #20328, #20323, @serena-ruan; #20737, @jamesbxwu; #20862, #20861, @PattaraS; #20805, #20705, #20373, @mprahl; #20773, @etirelli; #20753, @etscript; #20629, #19758, @justinwei-db; #20711, @kevin-lyn; #20576, @nisha2003; #20553, #20521, @danielseong1; #20548, @bartosz-grabowski; #20504, @smivv; #20527, @BenWilson2; #20363, #20364, @rollyjoel; #20494, @dbczumar; #20360, #20340, #20313, #20312, #20276, #20275, #20261, #20233, #19484, @hubertzub-db; #20359, @LiberiFatali; #20386, @chenmoneygithub; #20159, @ispoljari
We're excited to announce MLflow 3.10.0rc0, which includes several notable updates:
We're excited to announce MLflow 3.10.0rc0, which includes several notable updates:
Major New Features:
mlflow demo CLI command generates a fully-populated demo environment with sample traces, prompts, and evaluation data so you can explore MLflow's features hands-on without any setup. (#19994, #19995, #20046, #20047, #20048, #20162, @BenWilson2)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==3.10.0rc0
[Gateway] Deprecate Unity Catalog function integration in AI Gateway (#19457, @harupy)
We're excited to announce MLflow 3.9.0, which includes several notable updates:
Major New Features:
mlflow.tracing.distributed module (with more documentation to come soon).MemAlignOptimizer, a new algorithm that makes your judges smarter over time. It learns general guidelines from past feedback while dynamically retrieving relevant examples at runtime, giving you more accurate evaluations.Features:
generateContent and streamGenerateContent APIs (#19425, @TomeHirata)MemAlign as a new optimizer for judge alignment (#19598, @smoorjani)MemAlign guideline distillation (#20291, @veronicalyu320)GePaAlignmentOptimizer for judge instruction optimization (#19882, @alkispoly-db)Fluency scorer for evaluating text quality (#19414, @alkispoly-db)KnowledgeRetention built-in scorer (#19436, @alkispoly-db)mlflow.genai (#19614, @smoorjani)mlflow.genai.evaluate (#19760, @smoorjani)IS NULL / IS NOT NULL comparator support for trace metadata filtering (#19720, @dbczumar)mlflow.genai.optimize_prompts() (#19762, @chenmoneygithub)dspy.module.save API (#19704, @WeichenXu123)mlflow.genai.to_predict_fn for app invocation endpoints (#19779, @jennsun)log_stream API for logging binary streams as artifacts (#19104, @harupy)import_checkpoints API for databricks SGC Checkpointing with MLflow (#19839, @WeichenXu123)JupyterNotebookRunContext for Tracking local Jupyter notebook as the source (#19162, @iyashk)torch.export.save, add skops serialization format, and deprecate unsafe pickle/cloudpickle formats (#18759, #18832, #19692, #20151, @WeichenXu123)Bug fixes:
ConversationSimulator on managed (#20236, @xsh310)SparkDF trace handling in eval (#20207, @BenWilson2)MemAlign optimizer for incremental judge alignment (#20049, @veronicalyu320)DATABRICKS_CONFIG_PROFILE env var detection when fetching databricks credentials (#20112, @daniellok-db)mlflow.genai.optimize_prompts() (#19993, @chenmoneygithub)auth_mode from LiteLLMConfig (#20059, @TomeHirata)infer_code_paths to capture transitive imports of functions/classes (#19814, @copilot-swe-agent)MlflowException to HTTPException (#19728, @danielseong1)gateway_deprecated decorator - AI Gateway is not deprecated (#19821, @copilot-swe-agent)response_schema injection (#19741, @sinanshamsudheen)@overload annotations to @scorer decorator for proper type inference (#19570, @mr-brobot)catch_mlflow_exception (#19781, @harupy)KnowledgeRetention model parameter not propagating to inner scorer (#19753, @danielseong1)serve-artifacts is not enabled in docker-compose #19700 (#19701, @zjffdu)@trace_disabled decorator (#19569, @mr-brobot)StringDtype (#19518, @harupy)DeprecationWarning from generator.throw() in tracing (#19629, @mr-brobot)PyFuncOutput type alias for ResponsesAgent/ChatAgent/ChatModel (#19560, @copilot-swe-agent)enable_git_model_versioning to work from subdirectories (#19529, @copilot-swe-agent)Documentation updates:
multi_class argument from scikit-learn's LogisticRegression in docs (#20266, @SOORAJTS2001)register_prompt documentation example (#19591, @copilot-swe-agent)KnowledgeRetention scorer (#19478, @alkispoly-db)Small bug fixes and documentation updates:
#20406, #20122, #20317, #20333, #20361, #20274, #20362, #20249, #20169, #20345, #20252, #20314, #20214, #20215, #20210, #20212, #20142, #20183, #20121, #20141, #20140, #20124, #20073, #20062, #20065, #19893, #19912, #19464, #19857, #19401, #19600, #19555, #19400, #19392, #19393, @B-Step62; #20323, #20263, #19982, #20218, #20143, #20146, #20145, #20064, #20117, #20144, #20110, #20050, #20017, #20116, #20118, #19989, #19953, #19836, #19915, #19955, #19952, #19940, #19939, #19938, #19937, #19877, #19874, #19869, #19867, #19865, #19837, #19835, #19834, #19864, #19873, #19833, #19825, #19876, #19799, #19798, #19793, #19771, #19770, #19635, #19634, #19633, #19632, #19624, #19622, #19621, #19620, #19631, #19619, #19747, #19609, #19608, #19607, #19606, #19604, #19603, #19602, #19601, #19588, #19587, #19581, #19585, #19610, #19590, #19580, #19579, #19578, #19577, #19576, #19234, @serena-ruan; #20378, #20385, #20205, #20237, #20193, #20171, #20155, #20170, #20132, #20097, #20100, #20101, #19736, #19717, #19716, #19759, #19718, #19714, #19713, #19712, #19711, #19840, #19710, #19709, #19708, #19777, #19707, @dbczumar; #20387, #19981, #19964, @bbqiu; #20390, #20334, #20208, #19978, #19980, #19875, #19854, #19816, #19815, #19796, #19806, #19785, #19789, #19769, #19748, #19773, #19782, #19706, #19523, #19505, #19450, #19482, #19458, #19433, #19431, #19455, #19417, #19426, #19424, @harupy; #20355, #20245, #20120, #20229, #20114, #20053, #20012, #19972, #20002, #19991, #19990, #19977, #19986, #19985, #19967, #19957, #19960, #19954, #19945, #19941, #19934, #19917, #19916, #19905, #19904, #19903, #19900, #19899, #19897, #19894, #19892, #19890, #19888, #19887, #19861, #19828, #19818, #19803, #19802, #19791, #19788, #19795, #19790, #19786, #19783, #19767, #19768, #19746, #19735, #19733, #19732, #19726, #19561, #19549, #19544, #19543, #19510, #19486, #19487, #19463, #19871, @copilot-swe-agent; #20308, #20264, #20109, #20181, #20180, #20177, #20134, #20107, #20015, #20007, #20008, #19930, #20006, #20005, #19965, #19942, #19944, #19950, #19936, #19947, #19948, #19946, #19870, #19824, #19823, #19856, #19863, #19858, #19860, #19849, #19822, #19765, #19792, #19764, #19763, #19618, #19453, #19452, #19404, #19390, #19290, @TomeHirata; #20350, #20203, #19675, #19677, #19674, #19476, #19447, @BenWilson2; #20286, #20157, #20051, #20216, #20200, #20213, #20194, #20072, #20195, #20175, #20039, #19844, #19935, #19696, #19451, #19409, @smoorjani; #20209, #20131, #19742, #19969, #19734, #19480, #19351, @daniellok-db; #20204, #20164, #20192, #19997, #19925, #19850, #19914, #19774, #19721, #19673, #19623, #19668, #19496, #19554, #19471, @danielseong1; #20037, #19884, #19846, #19843, #19813, #19454, #19391, #19322, #19388, #19307, #19382, @xsh310; #20130, @iyashk; #20147, #20030, #19962, #19826, @kevin-lyn; #20108, #20071, #19743, #20045, #20042, #19959, #19880, @SomtochiUmeh; #20025, #19662, #19749, #19738, #19419, @WeichenXu123; #19847, @jaceklaskowski; #19820, @Abhiii47; #19800, @shreenidhi2205; #19703, #19693, #19689, #19688, #19664, #19663, #19660, #19534, #19533, #19532, #19531, @hubertzub-db; #19652, @AMRUTH-ASHOK; #19493, #19495, @alkispoly-db; #16372, @mohammadsubhani; #19522, @pmeier
We're excited to announce MLflow 3.9.0rc0, a pre-release including several notable updates:
We're excited to announce MLflow 3.9.0rc0, a pre-release including several notable updates:
Major New Features:
mlflow.tracing.distributed module (with more documentation to come soon).MemAlignOptimizer, a new algorithm that makes your judges smarter over time. It learns general guidelines from past feedback while dynamically retrieving relevant examples at runtime, giving you more accurate evaluations.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==3.9.0rc0
Please try it out and report any issues on the issue tracker.
MLflow 3.8.1 includes several bug fixes and documentation updates.
MLflow 3.8.1 includes several bug fixes and documentation updates.
Bug fixes:
Small bug fixes and documentation updates:
#19539, #19451, #19409, @smoorjani; #19493, @alkispoly-db
[] Fix SQL injection vulnerability in UC function execution (#19381, @harupy)
MLflow 3.8.0 includes several major features and improvements
get_judge API enables using DeepEval and RAGAS evaluation metrics as MLflow scorers, providing access to 20+ evaluation metrics including answer relevancy, faithfulness, and hallucination detection. (#18988, @smoorjani, #19345, @SomtochiUmeh)MLFLOW_DISABLE_TELEMETRY=true or DO_NOT_TRACK=true. If you do not manage your own server (e.g. you use a managed service or are not the admin), you can still opt out personally via the new "Settings" tab in the MLflow UI. For more information, please read the documentation on usage tracking.--max-results option to mlflow experiments search (#19359, @alkispoly-db)_search_runs for dataset filters (#19498, @fredericosantos)mlflow.genai.evaluate() (#19380, @brandonhawi)Small bug fixes and documentation updates:
#19497, #19358, #19322, #19383, #19288, #19287, #19230, #19225, @xsh310; #19504, @WeichenXu123; #19499, #19465, #19241, @B-Step62; #19479, #19385, #19297, #19347, #19314, #19286, #19269, @TomeHirata; #18894, @BnnaFish; #19480, #19427, #19351, #19312, #19292, #19303, #19291, #19418, #19395, #19240, #19267, #19102, #19082, #19076, @daniellok-db; #19463, #19370, #19369, #19368, #19367, #19366, #19363, #19354, #19302, #19272, #19266, #19258, #19255, #19242, #19236, #19235, #19203, #19214, #19212, #19210, #19204, #19197, #19196, #19194, #19190, #19182, #19178, #19179, #19163, #19157, #19150, #19137, #19132, #19114, #19115, #19113, #19112, #19111, #19110, #19107, #19091, #19090, #19078, @copilot-swe-agent; #19437, @SomtochiUmeh; #19420, #19329, #19317, #19207, #19086, @kevin-lyn; #19339, #19263, #19438, #19412, #19411, #19355, #19341, #19034, #19029, #19252, @smoorjani; #19416, #19399, #19402, #19353, #19313, #19296, #19294, #19264, #19202, #19206, #19165, #19161, #19158, #19126, #19147, #19099, @harupy; #19357, #19343, #19342, #19335, #19261, #19226, #19227, @BenWilson2; #19344, #19331, #19270, #19239, #19211, @serena-ruan; #19323, @bbqiu; #19373, @alkispoly-db; #19320, #19311, @kriscon-db; #19309, @stefanwayon; #19063, @cyficowley; #19160, @Killian-fal; #19142, #19141, @dbczumar; #19089, @hubertzub-db; #19098, @achen530
MLflow 3.8.0rc0 includes several major features and improvements. More features to come in the final 3.8.0 release!
MLflow 3.8.0rc0 includes several major features and improvements. More features to come in the final 3.8.0 release!
To try out this release candidate:
pip install mlflow==3.8.0rc0
get_judge API enables using DeepEval's evaluation metrics as MLflow scorers, providing access to 20+ evaluation metrics including answer relevancy, faithfulness, and hallucination detection. (#18988, @smoorjani)[Models] Remove deprecated diviner flavor (#18808, @copilot-swe-agent)
MLflow 3.7.0 includes several major features and improvements for GenAI Observability, Evaluation, and Prompt Management.
mlflow.genai.evaluate now supports multi-turn conversations, enabling comprehensive assessment of conversational AI applications with DataFrame and list inputs. (#18971, @AveshCSingh)make_judge API now supports structured outputs, enabling more precise and programmatically consumable evaluation results. (#18529, @TomeHirata)diviner flavor (#18808, @copilot-swe-agent)promptflow flavor (#18805, @copilot-swe-agent)--builtin/-b flag to mlflow scorers list command (#19095, @alkispoly-db)_delete_trace_tag_v3 API (#18813, @Tian-Sky-Lan)/v1/traces endpoint to return protobuf instead of JSON (#18929, @copilot-swe-agent)click!=8.3.0 in MCP extra to fix MCP server failure (#18748, @copilot-swe-agent)uv installation command for external users (#18745, @copilot-swe-agent)InstructionsJudge using scorer description as assessment value (#19121, @alkispoly-db)evaluate_traces MCP tool error: use result_df instead of tables (#18825, @alkispoly-db)anthropic_version field (#17744, @harupy)SqlLoggedModelMetric association with experiment_id (#18382, @mcompen)LoggedModelOutput.to_dictionary() so LoggedModelOutput and runs containing them can be JSON serialized (#19017, @nicklamiller)mlflow gc to remove model artifacts (#17282, @joelrobin18)Sentinel.UNSET handling in MCP server (#18858, @harupy)prompt_template (#19105, @ingo-stallknecht)make_judge API and direct judge invocation (#18897, @xsh310)mlflow gc command behavior for pinned runs and registered models (#18704, @copilot-swe-agent)Small bug fixes and documentation updates:
#19220, #19140, #19141, #18984, #18985, #18822, @dbczumar; #19148, @ingo-stallknecht; #19183, #19201, #19130, #19049, #19030, #18778, #18780, #18556, #18555, @serena-ruan; #19153, #19181, #18784, #18783, #18802, #18881, #18695, #18879, #18782, #18845, #18787, #18786, #18590, @B-Step62; #19208, #19021, #19023, #18723, #18622, @smoorjani; #13314, @alokshenoy; #19138, #19171, #19146, #19067, #19064, #19045, #18968, #18967, #19018, #18966, #18990, #18912, @xsh310; #19168, @mcompen; #19145, #18702, #18642, @BenWilson2; #19126, #19022, #18951, #18887, #18954, #18949, #18934, #18914, #18903, #18877, #18859, #18838, #18828, #18821, #18717, #18710, #18756, #18713, @harupy; #18890, #18862, #18836, #18792, #18818, #18579, @TomeHirata; #19084, #18886, #18911, #18904, #18885, #18837, #18795, #18646, @daniellok-db; #18992, #19025, #19020, #18950, @kevin-lyn; #19069, #19072, #19043, #19027, #19028, #19019, #18995, #18997, #18989, #18991, #18987, #18983, #18980, #18979, #18974, #18972, #18969, #18948, #18940, #18942, #18939, #18938, #18933, #18932, #18931, #18915, #18882, #18865, #18861, #18860, #18846, #18841, #18830, #18824, #18823, #18819, #18789, #18804, #18779, #18775, #18772, #18704, #18606, #18748, #18746, #18745, #18743, #18732, #18737, #18736, #18729, #18718, #18703, #18693, #18686, #18682, #18633, #18675, #18671, #18653, #18652, @copilot-swe-agent; #19001, #18945, @danielseong1; #18815, @kevin-wangg; #19039, #18898, @AveshCSingh; #18742, @Killian-fal; #18923, @HomeLH; #18922, #18920, @UnfixedMold; #18798, @WeichenXu123; #18776, @pcliupc; #18417, @shaperilio
Removed Deprecated Flavors: The diviner and promptflow flavors have been removed from MLflow. Please migrate to supported alternatives. (#18808, #1880…
MLflow 3.7.0rc0 includes several major features and improvements!
mlflow.genai.evaluate, enabling comprehensive assessment of conversational AI applications. (#18971, #19039, @AveshCSingh)diviner and promptflow flavors have been removed from MLflow. Please migrate to supported alternatives. (#18808, #18805, @copilot-swe-agent)MLFLOW_DISABLE_TELEMETRY=true or DO_NOT_TRACK=true. See the usage tracking documentation for details. (#18881, @B-Step62)Stay tuned for the full release, which will be packed with more features and bugfixes.
To try out this release candidate, please run:
pip install mlflow==3.7.0rc0
Deprecate pmdarima, promptflow, diviner flavors (#18597, #18577, @copilot-swe-agent)
MLflow 3.6.0 includes several major features and improvements for AI Observability, Experiment UI, Agent Evaluation and Deployment.
Breaking changes:
_1, _2, ...) from span names (#18531, @serena-ruan)search_traces tool for agentic judge (#18228, @dbrx-euirim)_raise_if_prompt for Unity Catalog tag operations (#18707, @harupy)mlflow.spark.load_model to handle Unity Catalog Volumes paths correctly (#18672, @harupy)delete_traces calls to Databricks MLflow server (#18563, @dbrx-euirim)log_metric to accept mlflow.entities.Dataset (#18585, @harupy)pathlib.Path in validation.py (#16660, @benglewis)Documentation updates:
Small bug fixes and documentation updates:
#18595, @danielseong1; #18622, #18723, #18459, @smoorjani; #18643, @dbczumar; #18731, #18667, #18666, #18728, #18692, #18705, #18690, #18654, #18590, #18429, #18530, #18416, #18401, #18400, #18465, #18453, #18414, #18421, @B-Step62; #18687, #18661, #18665, #18537, #18641, #18631, #18629, #18605, #18426, #18603, #18526, #18587, #18583, #18564, #18536, #18544, #18567, #18565, #18533, #18535, #18501, #18498, #18368, #18357, #18471, #18476, #18356, #18214, #17975, @serena-ruan; #18725, @bbqiu; #18714, #18708, #18679, #18681, #18660, #18659, #18664, #18658, #18689, #18657, #18656, #18627, #18626, #18625, #18424, #18028, @daniellok-db; #18726, @alkispoly-db; #18702, #18513, #18461, #18430, #18336, @BenWilson2; #18579, #18578, #18569, @TomeHirata; #18677, @nicklamiller; #18676, #18663, #18600, #18604, #18602, #18566, #18549, #18538, #18517, #15849, #18492, #18468, #18475, #18469, #18467, #18452, #18449, #18450, #18447, #18442, #18327, #18395, #18418, #18350, #18278, #18242, #18234, #18203, #18175, #18210, @harupy; #18601, #18649, #18616, #18615, #18607, #18598, #18588, #18586, #18584, #18572, #18580, #18571, #18554, #18553, #18552, #18551, #18548, #18546, #18528, #18527, #18525, #18521, #18520, #18515, #18519, #18518, #18506, #18507, #18505, #18502, #18495, #18494, #18472, #18463, #18464, #18462, #18443, #18440, #18399, #18394, #18393, #18392, #18390, #18389, #18380, #18376, #18378, #18377, #18366, #18362, #18361, #18343, #18340, #18318, #18311, #18307, #18269, #18268, #18261, #18260, #18259, #18258, #18257, #18256, #18253, #18254, #18252, #18250, #18243, #18238, #18213, #18206, #18198, #18184, #18179, @copilot-swe-agent; #18575, @dbrx-euirim; #18570, #18116, #18360, #18351, @WeichenXu123; #18488, @raymondzhou-db; #18334, @NJAHNAVI2907
MLflow 3.6.0 includes several major features and improvements for AI Observability, Experiment UI, Agent Evaluation and Deployment.
Breaking changes:
_1, _2, ...) from span names (#18531, @serena-ruan)Features:
mlflow.pytorch pyfunc loader supporting pytorch forecasting model (#18428, @WeichenXu123)search_traces tool for agentic judge (#18228, @dbrx-euirim)Bug fixes:
delete_traces calls to Databricks MLflow server (#18563, @dbrx-euirim)pathlib.Path in validation.py (#16660, @benglewis)Documentation updates:
Small bug fixes and documentation updates:
#18735, #18429, #18530, #18416, #18401, #18400, #18465, #18453, #18414, #18421, @B-Step62; #18641, #18631, #18629, #18605, #18426, #18603, #18526, #18587, #18583, #18564, #18536, #18544, #18567, #18565, #18533, #18535, #18501, #18498, #18368, #18357, #18471, #18476, #18356, #18214, #17975, @serena-ruan; #18600, #18604, #18602, #18566, #18549, #18538, #18517, #15849, #18492, #18468, #18475, #18469, #18467, #18452, #18449, #18450, #18447, #18442, #18327, #18395, #18418, #18350, #18278, #18242, #18234, #18203, #18175, #18210, @harupy; #18625, #18424, #18028, @daniellok-db; #18616, #18615, #18607, #18598, #18588, #18586, #18584, #18572, #18580, #18571, #18554, #18553, #18552, #18551, #18548, #18546, #18528, #18527, #18525, #18521, #18520, #18515, #18519, #18518, #18506, #18507, #18505, #18502, #18495, #18494, #18472, #18463, #18464, #18462, #18443, #18440, #18399, #18394, #18393, #18392, #18390, #18389, #18380, #18376, #18378, #18377, #18366, #18362, #18361, #18343, #18340, #18318, #18311, #18307, #18269, #18268, #18261, #18260, #18259, #18258, #18257, #18256, #18253, #18254, #18252, #18250, #18243, #18238, #18213, #18206, #18198, #18184, #18179, @copilot-swe-agent; #18578, #18569, @TomeHirata; #18575, @dbrx-euirim; #18570, #18116, #18360, #18351, @WeichenXu123; #18513, #18461, #18430, #18336, @BenWilson2; #18459, @smoorjani; #18488, @raymondzhou-db; #18334, @NJAHNAVI2907
[Tracking] Filesystem Backend Deprecation: The filesystem backend is being deprecated in favor of SQLite. See #18534 for details.
MLflow 3.6.0rc0 includes several major features and improvements!
Stay tuned for the full release, which will be packed with more features and bugfixes.
To try out this release candidate, please run: pip install mlflow==3.6.0rc0
MLflow 3.6.0rc0 includes several major features and improvements!
Stay tuned for the full release, which will be packed with more features and bugfixes.
To try out this release candidate, please run:
pip install mlflow==3.6.0rc0
MLflow 3.5.1 is a patch release that includes several bug fixes and improvements.
MLflow 3.5.1 is a patch release that includes several bug fixes and improvements.
Features:
set_databricks_monitoring_sql_warehouse_id API (#18346, @dbrx-euirim)Bug fixes:
Documentation updates:
[Docs] Add deprecation notice for custom prompt judge (#18287, @smoorjani)
MLflow 3.5.0 includes several major features and improvements!
unlink_traces_from_run batch operation (#18316, @harupy)__repr__ method for Judges (#17794, @BenWilson2)get_trace to improve reliability (#18224, @B-Step62)to_predict_fn to handle traces without tags field (#17784, @harupy)delete_trace_tag to prevent 404 errors (#18232, @copilot-swe-agent)job_start API (#18226, @BenWilson2)merge_records() API (#18047, @BenWilson2)allow_missing is set (#17541, @mr-brobot)to_evaluation_dataset method (#17886, @sadelcarpio)mlflow server exiting immediately when optional huey package is missing (#18016, @harupy)max_few_show_examples to max_few_shot_examples (#18246, @srinathmkce)Small bug fixes and documentation updates:
#18349, #18338, #18241, #18319, #18309, #18292, #18280, #18239, #18236, #17786, #18003, #17970, #17898, #17765, #17667, @serena-ruan; #18346, #17882, @dbrx-euirim; #18306, #18208, #18165, #18110, #18109, #18108, #18107, #18105, #18104, #18100, #18099, #18155, #18079, #18082, #18078, #18077, #18083, #18030, #18001, #17999, #17712, #17785, #17756, #17729, #17731, #17733, @daniellok-db; #18339, #18291, #18222, #18210, #18124, #18101, #18054, #18053, #18007, #17922, #17823, #17822, #17805, #17789, #17750, #17752, #17760, #17758, #17688, #17689, #17693, #17675, #17673, #17656, #17674, @harupy; #18331, #18308, #18303, #18146, @smoorjani; #18315, #18279, #18310, #18187, #18225, #18277, #18193, #18223, #18209, #18200, #18178, #17574, #18021, #18006, #17944, @B-Step62; #18290, #17946, #17627, @bbqiu; #18274, @Ninja3047; #18204, #17868, #17866, #17833, #17826, #17835, @TomeHirata; #18273, #18043, #17928, #17931, #17936, #17937, @dbczumar; #18185, #18180, #18174, #18170, #18167, #18164, #18168, #18166, #18162, #18160, #18159, #18157, #18156, #18154, #18148, #18145, #18135, #18143, #18142, #18139, #18132, #18130, #18119, #18117, #18115, #18102, #18075, #18046, #18062, #18042, #18051, #18036, #18027, #18014, #18011, #18009, #18004, #17903, #18000, #18002, #17973, #17993, #17989, #17984, #17968, #17966, #17967, #17962, #17977, #17976, #17972, #17965, #17964, #17963, #17969, #17971, #17939, #17926, #17924, #17915, #17911, #17912, #17904, #17902, #17900, #17897, #17892, #17889, #17888, #17885, #17884, #17878, #17874, #17873, #17871, #17870, #17865, #17860, #17861, #17859, #17857, #17856, #17854, #17853, #17851, #17849, #17850, #17847, #17845, #17846, #17844, #17843, #17842, #17838, #17836, #17834, #17831, #17824, #17828, #17819, #17825, #17817, #17821, #17809, #17807, #17808, #17803, #17800, #17799, #17797, #17793, #17790, #17772, #17771, #17769, #17770, #17753, #17762, #17747, #17749, #17745, #17740, #17734, #17732, #17726, #17723, #17722, #17721, #17719, #17720, #17718, #17716, #17713, #17715, #17710, #17709, #17708, #17707, #17705, #17697, #17701, #17698, #17696, #17695, @copilot-swe-agent; #18151, #18153, #17983, #18040, #17981, #17841, #17818, #17776, #17781, @BenWilson2; #18068, @alkispoly-db; #18133, @kevin-lyn; #17105, #17717, @joelrobin18; #17879, @lkuo; #17996, #17945, #17913, @WeichenXu123
MLflow 3.5.0rc0 includes several major features and improvements
MLflow 3.5.0rc0 includes several major features and improvements
Major new features:
Stay tuned for the full release, which will be packed with more features and bugfixes.
To try out this release candidate, please run:
pip install mlflow==3.5.0rc0
MLflow 3.4.0rc0 includes several major features and improvements
MLflow 3.4.0rc0 includes several major features and improvements
make_judge API enables creation of custom evaluation judges for assessing LLM outputs with domain-specific criteria. (#17647, @BenWilson2, @dbczumar, @alkispoly-db, @smoorjani)Features:
Bug fixes:
Documentation updates:
Small bug fixes and documentation updates:
#17655, #17657, #17597, #17545, #17547, @BenWilson2; #17671, @smoorjani; #17668, #17665, #17662, #17661, #17659, #17658, #17653, #17643, #17642, #17636, #17634, #17631, #17628, #17611, #17607, #17588, #17570, #17575, #17564, #17557, #17556, #17555, #17536, #17531, #17524, #17510, #17511, #17499, #17500, #17494, #17493, #17490, #17488, #17478, #17479, #17425, #17471, #17457, #17440, #17403, #17405, #17404, #17402, #17366, #17346, #17344, #17337, #17316, #17313, #17284, #17276, #17235, #17226, #17229, @copilot-swe-agent; #17664, #17654, #17613, #17637, #17633, #17612, #17630, #17616, #17626, #17617, #17610, #17614, #17602, #17538, #17522, #17512, #17508, #17492, #17462, #17475, #17468, #17455, #17338, #17257, #17231, #17214, #17223, #17218, #17216, @harupy; #17635, #17663, #17426, #16870, #17428, #17427, #17441, #17377, @serena-ruan; #17605, #17306, @daniellok-db; #17624, #17578, #17369, #17391, #17072, #17326, #17115, @dbczumar; #17598, #17408, #17353, @nsthorat; #17601, #17553, @dbrx-euirim; #17586, #17587, #17310, #17180, @TomeHirata; #17516, @bbqiu; #17477, #17474, @WeichenXu123; #17449, @raymondzhou-db; #17470, @jacob-danner; #17378, @arpitjasa-db; #17121, @ctaymor; #17351, #17322, @ispoljari; #17292, @dsuhinin; #17287, #17281, #17230, #17245, #17237, @B-Step62
MLflow 3.4.0rc0 includes several major features and improvements. Stay tuned for the full release, which will be packed with more features and bugfixe
MLflow 3.4.0rc0 includes several major features and improvements. Stay tuned for the full release, which will be packed with more features and bugfixes.
To try out this release candidate, please run: pip install mlflow==3.6.0rc0
Major Features
make_judge API enables creation of custom evaluation judges for assessing LLM outputs with domain-specific criteria. (#17647, @BenWilson2, @dbczumar, @alkispoly-db, @smoorjani)MLflow 3.4.0rc0 includes several major features and improvements
make_judge API enables creation of custom evaluation judges for assessing LLM outputs with domain-specific criteria. (#17647, @BenWilson2, @dbczumar, @alkispoly-db, @smoorjani)Features:
Bug fixes:
Documentation updates:
Small bug fixes and documentation updates:
#17655, #17657, #17597, #17545, #17547, @BenWilson2; #17671, @smoorjani; #17668, #17665, #17662, #17661, #17659, #17658, #17653, #17643, #17642, #17636, #17634, #17631, #17628, #17611, #17607, #17588, #17570, #17575, #17564, #17557, #17556, #17555, #17536, #17531, #17524, #17510, #17511, #17499, #17500, #17494, #17493, #17490, #17488, #17478, #17479, #17425, #17471, #17457, #17440, #17403, #17405, #17404, #17402, #17366, #17346, #17344, #17337, #17316, #17313, #17284, #17276, #17235, #17226, #17229, @copilot-swe-agent; #17664, #17654, #17613, #17637, #17633, #17612, #17630, #17616, #17626, #17617, #17610, #17614, #17602, #17538, #17522, #17512, #17508, #17492, #17462, #17475, #17468, #17455, #17338, #17257, #17231, #17214, #17223, #17218, #17216, @harupy; #17635, #17663, #17426, #16870, #17428, #17427, #17441, #17377, @serena-ruan; #17605, #17306, @daniellok-db; #17624, #17578, #17369, #17391, #17072, #17326, #17115, @dbczumar; #17598, #17408, #17353, @nsthorat; #17601, #17553, @dbrx-euirim; #17586, #17587, #17310, #17180, @TomeHirata; #17516, @bbqiu; #17477, #17474, @WeichenXu123; #17449, @raymondzhou-db; #17470, @jacob-danner; #17378, @arpitjasa-db; #17121, @ctaymor; #17351, #17322, @ispoljari; #17292, @dsuhinin; #17287, #17281, #17230, #17245, #17237, @B-Step62
MLflow 3.3.2 is a patch release that includes several minor improvements and bugfixes
MLflow 3.3.2 is a patch release that includes several minor improvements and bugfixes
Features:
Bug fixes:
Documentation updates:
Small bug fixes and documentation updates:
#17301, #17299, @B-Step62; #17420, #17421, #17398, #17397, #17349, #17361, #17377, #17359, #17358, #17356, #17261, #17263, #17262, @serena-ruan; #17422, #17310, #17357, @TomeHirata; #17406, @sotagg; #17418, @annzhang-db; #17384, #17376, @daniellok-db
MLflow 3.3.1 includes several improvements
MLflow 3.3.1 includes several improvements
Bug fixes:
[Tracking] Fix mlflow.genai.datasets attribute (#17307, @WeichenXu123) [UI] Fix tag display as column in experiment overview (#17296, @joelrobin18) [Tracing] Fix the slowness of dspy tracing (#17290, @TomeHirata) Small bug fixes and documentation updates:
#17295, @gunsodo; #17272, @bbqiu
For a comprehensive list of changes, check out the latest documentation on mlflow.org.
MLflow 3.3.1 includes several major features and improvements
Bug fixes:
mlflow.genai.datasets attribute (#17307, @WeichenXu123)Small bug fixes and documentation updates:
#17295, @gunsodo; #17272, @bbqiu
MLflow 3.3.0 includes several major features and improvements
MLflow 3.3.0 includes several major features and improvements
New features:
memory span type for agentic workflows (#17034, @B-Step62)optimize_prompt including DSPy support (#17052, @TomeHirata)token_count function (#16253, @joelrobin18)Bug fixes:
Documentation updates:
optimize_prompt (#17084, @TomeHirata)Small bug fixes and documentation updates:
#17230, #17264, #17289, #17287, #17265, #17238, #17215, #17224, #17185, #17148, #17193, #17157, #17067, #17033, #17087, #16973, #16875, #16956, #16959, @B-Step62; #17269, @BenWilson2; #17285, #17259, #17260, #17236, #17196, #17169, #17062, #16943, @serena-ruan; #17253, @sotagg; #17212, #17206, #17211, #17207, #17205, #17118, #17177, #17182, #17170, #17153, #17168, #17123, #17136, #17119, #17125, #17088, #17101, #17056, #17077, #17057, #17036, #17018, #17024, #17019, #16883, #16972, #16961, #16968, #16962, #16958, @harupy; #17209, #17202, #17184, #17179, #17174, #17141, #17155, #17145, #17130, #17113, #17110, #17098, #17104, #17100, #17060, #17044, #17032, #17008, #17001, #16994, #16991, #16984, #16976, @copilot-swe-agent; #17069, @hayescode; #17199, #17081, #16928, #16931, @TomeHirata; #17198, @WeichenXu123; #17195, #17192, #17131, #17128, #17124, #17120, #17102, #17093, #16941, @daniellok-db; #17070, #17074, #17073, @dbczumar
MLflow 3.3.0 includes several major features and improvements.
MLflow 3.3.0 includes several major features and improvements.
Stay tuned for the full 3.3.0 release, packed with more features, refinements, and bug fixes. To try out this release candidate:
pip install mlflow==3.3.0rc0
MLflow 3.2.0 includes several major features and improvements
MLflow 3.2.0 includes several major features and improvements
Features:
Bug fixes:
get_model_info to provide logged model info (#16713, @harupy)mlflow.genai.evaluate (#16932, @B-Step62)Documentation updates:
Small bug fixes and documentation updates:
#17003, #17049, #17035, #17026, #16981, #16971, #16953, #16930, #16917, #16738, #16717, #16693, #16694, #16684, #16678, #16656, #16513, #16459, #16277, #16276, #16275, #16170, #16217, @serena-ruan; #16927, #16915, #16913, #16911, #16909, #16889, #16727, #16600, #16543, #16551, #16526, #16533, #16535, #16531, #16472, #16392, #16389, #16385, #16376, #16369, #16367, #16321, #16311, #16307, #16273, #16268, #16265, #16112, #16243, #16231, #16226, #16221, #16196, @copilot-swe-agent; #17050, #17048, #16955, #16894, #16885, #16860, #16841, #16835, #16801, #16701, @daniellok-db; #16898, #16881, #16858, #16735, #16823, #16814, #16647, #16750, #16809, #16794, #16793, #16789, #16780, #16770, #16773, #16771, #16772, #16768, #16752, #16754, #16751, #16748, #16730, #16729, #16346, #16709, #16704, #16703, #16702, #16658, #16662, #16645, #16639, #16640, #16626, #16572, #16566, #16565, #16563, #16561, #16559, #16544, #16539, #16520, #16508, #16505, #16494, #16495, #16491, #16487, #16482, #16473, #16465, #16456, #16458, #16394, #16445, #16433, #16434, #16413, #16417, #16416, #16414, #16415, #16378, #16350, #16323, #15788, #16263, #16256, #16237, #16234, #16219, #16216, #16207, #16199, #16192, #16705, @harupy; #17047, #17017, #17005, #16989, #16952, #16951, #16903, #16900, #16755, #16762, #16757, #15860, #16661, #16630, #16657, #16605, #16602, #16568, #16569, #16553, #16345, #16454, #16489, #16486, #16438, #16266, #16382, #16381, #16303, @B-Step62; #17028, #17027, #17020, @he7d3r; #16969, #16957, #16852, #16829, #16816, #16808, #16775, #16807, #16806, #16624, #16524, #16410, #16403, @TomeHirata; #16987, @wangh118; #16760, #16761, #16736, #16737, #16699, #16718, #16663, #16676, #16574, #16477, #16552, #16527, #16515, #16452, #16210, #16204, #16610, @frontsideair; #16723, #16124, @AveshCSingh; #16744, @BenWilson2; #16683, @dsuhinin; #16877, #16502, @bbqiu; #16619, @AchimGaedkeLynker; #16595, @Aiden-Jeon; #16480, #16479, @shushantrishav; #16398, #16331, #16328, #16329, #16293, @WeichenXu123
🧭 Tracing TypeScript SDK: MLflow Tracing now supports the TypeScript SDK, allowing developers to trace GenAI applications in TypeScript environments.
Features:
Bug fixes:
Documentation updates:
Small bug fixes and documentation updates:
#16930, #16917, #16738, #16717, #16693, #16694, #16684, #16678, #16656, #16513, #16459, #16277, #16276, #16275, #16170, #16217, @serena-ruan; #16927, #16915, #16913, #16911, #16909, #16889, #16727, #16600, #16543, #16551, #16526, #16533, #16535, #16531, #16472, #16392, #16389, #16385, #16376, #16369, #16367, #16321, #16311, #16307, #16273, #16268, #16265, #16112, #16243, #16231, #16226, #16221, #16196, @copilot-swe-agent; #16894, #16885, #16860, #16841, #16835, #16801, #16701, @daniellok-db; #16898, #16881, #16858, #16735, #16823, #16814, #16647, #16750, #16809, #16794, #16793, #16789, #16780, #16770, #16773, #16771, #16772, #16768, #16752, #16754, #16751, #16748, #16730, #16729, #16346, #16709, #16704, #16703, #16702, #16658, #16662, #16645, #16639, #16640, #16626, #16572, #16566, #16565, #16563, #16561, #16559, #16544, #16539, #16520, #16508, #16505, #16494, #16495, #16491, #16487, #16482, #16473, #16465, #16456, #16458, #16394, #16445, #16433, #16434, #16413, #16417, #16416, #16414, #16415, #16378, #16350, #16323, #15788, #16263, #16256, #16237, #16234, #16219, #16216, #16207, #16199, #16192, #16705, @harupy; #16900, #16755, #16762, #16757, #15860, #16661, #16630, #16657, #16605, #16602, #16568, #16569, #16553, #16345, #16454, #16489, #16486, #16438, #16266, #16382, #16381, #16303, @B-Step62; #16852, #16829, #16816, #16808, #16775, #16807, #16806, #16624, #16524, #16410, #16403, @TomeHirata; #16760, #16761, #16736, #16737, #16699, #16718, #16663, #16676, #16574, #16477, #16552, #16527, #16515, #16452, #16210, #16204, #16610, @frontsideair; #16723, #16124, @AveshCSingh; #16744, @BenWilson2; #16683, @dsuhinin; #16502, @bbqiu; #16619, @AchimGaedkeLynker; #16595, @Aiden-Jeon; #16480, #16479, @shushantrishav; #16398, #16331, #16328, #16329, #16293, @WeichenXu123
MLflow 3.1.4 includes several major features and improvements
MLflow 3.1.4 includes several major features and improvements
Small bug fixes and documentation updates:
#16835, #16820, @daniellok-db
MLflow 3.1.3 includes several features and improvements
MLflow 3.1.3 includes several features and improvements
Features:
Bug fixes:
MLFLOW_DEPLOYMENT_PREDICT_TIMEOUT to databricks-sdk (#16783, @bbqiu)Small bug fixes and documentation updates:
#16786, #16692, @daniellok-db; #16594, @ngoduykhanh; #16475, @harupy
> This version has been yanked. MLflow 3.1.3 will be released shortly.
[!WARNING] This version has been yanked. MLflow 3.1.3 will be released shortly.
MLflow 3.1.2 is a patch release that includes several bug fixes.
Bug fixes:
download_artifacts ignoring tracking_uri parameter (#16461, @harupy)Small fixes and documentation updates:
#16568, #16454, #16617, #16605, #16569, #16553, #16625, @B-Step62; #16571, #16552, #16452, #16395, #16446, #16420, #16447, #16554, #16515, @frontsideair; #16558, #16443, #16457, @16442, #16449, @harupy; #16509, #16512, #16524, #16514, #16607, @TomeHirata; #16541, @copilot-swe-agent; #16427, @bbqiu; #16573, @daniellok-db; #16470, #16281, @BenWilson2
MLflow 3.1.1 includes several major features and improvements
MLflow 3.1.1 includes several major features and improvements
Features:
Bug fixes:
Documentation updates:
Small bug fixes and documentation updates:
#16261, @rohitarun-db; #16411, #16352, #16327, #16324, #16279, #16193, #16197, @harupy; #16409, #16348, #16347, #16290, #16286, #16283, #16271, #16223, @TomeHirata; #16326, @mohammadsubhani; #16364, @BenWilson2; #16308, #16218, @serena-ruan; #16262, @raymondzhou-db; #16191, @copilot-swe-agent; #16212, @B-Step62; #16208, @frontsideair; #16205, #16200, #16198, @daniellok-db
[Docs] MLflow 3 breaking changes list (#15716, @WeichenXu123)
<img width="1624" alt="Screenshot 2025-06-12 at 3 20 33" src="https://github.com/user-attachments/assets/66b4b221-a3b8-488f-8109-e17de4d17be2" />
MLflow 3 is now available to everyone, marking the biggest evolution in the best open-source MLOps platform's history and transforming how millions of developers build, deploy, AI applications. While previous versions focused on traditional ML workflows, MLflow 3 fundamentally reimagines the platform for the GenAI era. This isn't just an update, but a complete paradigm shift that brings enterprise-grade GenAI capabilities to the open source community for the first time.
MLflow 3 introduces a refined architecture with the new LoggedModel entity as a first-class citizen, moving beyond the traditional run-centric approach. This enables better organization and comparison of GenAI models. agents, deep learning checkpoints, and model variants across experiments.
Enhanced model tracking provides comprehensive lineage between models, runs, traces, prompts, and evaluation metrics. The new model-centric design allows you to group traces and metrics from different development environments and production, enabling rich comparisons across model versions.
MLflow's evaluation and monitoring capabilities help you systematically measure, improve, and maintain the quality of your GenAI applications throughout their lifecycle. From development through production, use the same quality scorers to ensure your applications deliver accurate, reliable responses while managing cost and latency. Visit documentation for more details.
Real-world GenAI applications need human oversight. MLflow 3 now tracks human annotations and feedback for model predictions, enabling streamlined human-in-the-loop evaluation cycles. This creates a collaborative environment where data scientists, domain experts, and stakeholders can efficiently improve model quality together. (Note: Currently available in Databricks Managed MLflow. Open source release coming in the next few months.)
Transform prompt engineering from art to science. The MLflow Prompt Registry now includes prompt optimization capabilities built on top of the state-of-the-art research, allowing you to automatically improve prompts using evaluation feedback and labeled datasets. This includes versioning, tracking, and systematic prompt engineering workflows.
The MLflow documentation and website has been fully redesigned to support two main user journeys: GenAI development and classic machine learning workflows. The new structure offers dedicated sections for GenAI features (including LLMs, prompt engineering, and tracing), and traditional ML capabilities such as experiment tracking, model registry, deployment, and evaluation.
Get up and running with MLflow 3 in minutes:
pip install 'mlflow>=3.1'
🌐 New Website | 📖 Documentation | 🎉:Release Notes
It is just the beginning. The open source community continues driving innovation toward the world's best open-source MLOps/LLMOps platform. Here's how you can be part of the journey:
The future of AI development is unified, observable, and reliable. MLflow 3.0 brings that future to the open source community today.
Ready to transform your GenAI workflow? Get started now →
mlflow[databricks] (#16097, @dbrx-euirim)MlflowSparkStudy (#15418, @lu-wang-dl)spark_udf support DBConnect + DBR 15.4 / DBR dedicated cluster (#15968, @WeichenXu123)uv (#15875, @harupy)mlflow.genai.optimize_prompt to optimize prompts (#15861, @TomeHirata)ResponsesAgent.predict_stream (#15762, @bbqiu)LogLoggedModelParams (#15717, @artjen)predict_stream in DSPy flavor (#15678, @TomeHirata)search_prompts function to list all the prompts registered (#15445, @joelrobin18)DATABRICKS_CONFIG_PROFILE environment variable. (#15587, @WeichenXu123)smolagents (#15574, @y-okt)allow_missing parameter in load_prompt (#15371, @joelrobin18)mlflow.get_artifact_uri() usage outside active run (#12902, @Shashank1202)Bug fixes:
mlflow gc (#11773, @oleg-z)include_spans=False (#15634, @dbczumar)global_guideline_adherence (#15572, @artjen)Resources from SystemAuthPolicy in CreateModelVersion (#15485, @aravind-segu)ResponsesAgent interface update (#15601, #15741, @bbqiu)Breaking changes:
mlflow.genai.prompts namespace (#16174, @B-Step62)mlflow.evaluate (#15827, @harupy)mlflow.search_trace() to be V3 format (#15643, @B-Step62)Documentation updates:
Small bug fixes and documentation updates:
#16193, #16192, #16171, #16119, #16036, #16130, #16081, #16101, #16047, #16086, #16077, #16045, #16065, #16067, #16063, #16061, #16058, #16050, #16043, #16034, #16033, #15966, #16025, #16015, #16002, #15970, #16001, #15999, #15942, #15960, #15955, #15951, #15939, #15885, #15883, #15890, #15887, #15874, #15869, #15846, #15845, #15826, #15834, #15822, #15830, #15796, #15821, #15818, #15817, #15805, #15804, #15798, #15793, #15797, #15782, #15775, #15772, #15790, #15773, #15776, #15756, #15767, #15766, #15765, #15746, #15747, #15748, #15751, #15743, #15731, #15720, #15722, #15670, #15614, #15715, #15677, #15708, #15673, #15680, #15686, #15671, #15657, #15669, #15664, #15675, #15667, #15666, #15668, #15651, #15649, #15647, #15640, #15638, #15630, #15627, #15624, #15622, #15558, #15610, #15577, #15575, #15545, #15576, #15559, #15563, #15555, #15557, #15548, #15551, #15547, #15542, #15536, #15524, #15531, #15525, #15520, #15521, #15502, #15499, #15442, #15426, #15315, #15392, #15397, #15399, #15394, #15358, #15352, #15349, #15328, #15336, #15335, @harupy; #16196, #16191, #16093, #16114, #16080, #16088, #16053, #15856, #16039, #15987, #16009, #16014, #16007, #15996, #15993, #15991, #15989, #15978, #15839, #15953, #15934, #15929, #15926, #15909, #15900, #15893, #15889, #15881, #15879, #15877, #15865, #15863, #15854, #15852, #15848, @copilot-swe-agent; #16178, #16153, #16155, #15823, #15754, #15794, #15800, #15799, #15615, #15777, #15726, #15752, #15745, #15753, #15738, #15681, #15684, #15682, #15702, #15679, #15623, #15645, #15612, #15533, #15607, #15522, @serena-ruan; #16177, #16167, #16168, #16166, #16152, #16144, #15920, #16134, #16128, #16098, #16059, #16024, #15974, #15917, #15676, #15750, @dbczumar; #16162, #16161, #16137, #16126, #16127, #16099, #16074, #16041, #16040, #16010, #15945, #15697, #15588, #15602, #15581, @rohitarun-db; #16150, #15984, #16125, #16102, #16062, #16060, #15986, #15985, #15983, #15982, #15980, #15763, @smoorjani; #16160, #16149, #16103, #15538, #16055, #16054, #16048, #16012, #16029, #16003, #15940, #15956, #15950, #15906, #15922, #15932, #15930, #15905, #15910, #15902, #15901, #15840, #15896, #15898, #15895, #15850, #15833, #15824, #15819, #15816, #15806, #15803, #15795, #15759, #15791, #15792, #15774, #15769, #15768, #15770, #15755, #15771, #15737, #15690, #15733, #15730, #15687, #15660, #15735, #15688, #15705, #15590, #15663, #15665, #15658, #15594, #15620, #15644, #15648, #15605, #15639, #15642, #15619, #15618, #15611, #15597, #15589, #15580, #15593, #15437, #15584, #15582, #15448, #15351, #15317, #15353, #15320, #15319, @B-Step62; #16151, #16142, #16111, #16106, #16051, #16046, #16044, #15971, #15957, #15810, #15749, #15706, #15683, #15728, #15732, #15707, #15621, #15567, #15566, #15523, #15479, #15404, #15400, #15378, @TomeHirata; #16026, #16072, @AveshCSingh; #15967, @euirim; #15884, #15924, #15395, #15393, #15390, @daniellok-db; #15786, @rahuja23; #15734, @lhrotk; #15809, #15739, #15695, #15654, #15694, #15655, #15653, #15608, #15543, #15573, @dhruyads; #15596, @mrharishkumar; #15742, #15723, #15633, #15606, @ShaylanDias; #15703, #15637, #15613, #15473, @joelrobin18; #15636, #15659, #15616, #15617, @raymondzhou-db; #15674, #15598, #15357, #15586, @WeichenXu123; #15691, @artjen; #15698, @prithvikannan; #15631, @hubertzub-db; #15569, @Anand1923; #15578, @y-okt; #14790, @singh-kristian; #14129, @jamblejoe; #15552, @BenWilson2; #14197, @clarachristiansen; #15505, @Conor0Callaghan; #15509, @tr33k; #15507, @vzamboulingame; #15459, @UnMelow; #13991, @abhishekpawar1060; #12161, @zhouyou9505; #15293, @tornikeo
We're happy to announce MLflow 3.1.0rc0!
We're happy to announce MLflow 3.1.0rc0!
You can upgrade with pip as usual:
pip install mlflow==3.1.0rc0
See https://mlflow.org/docs/3.1.0rc0/mlflow-3/ for what's new in MLflow 3.0.
MLflow 3.0.1 includes several major features and improvements
MLflow 3.0.1 includes several major features and improvements
Features:
Bug fixes:
Small bug fixes and documentation updates:
#16364, @BenWilson2; #16347, @TomeHirata; #16279, #15835, @harupy; #16182, @B-Step62
See https://github.com/mlflow/mlflow/releases/tag/v3.1.0.
See https://github.com/mlflow/mlflow/releases/tag/v3.1.0.
We're happy to announce MLflow 3.0.0rc3!
We're happy to announce MLflow 3.0.0rc3!
You can upgrade with pip as usual:
pip install mlflow==3.0.0rc3
See https://mlflow.org/docs/3.0.0rc3/mlflow-3/ for what's new in MLflow 3.0.
We're happy to announce MLflow 3.0.0rc2!
We're happy to announce MLflow 3.0.0rc2!
You can upgrade with pip as usual:
pip install mlflow==3.0.0rc2
See https://mlflow.org/docs/3.0.0rc2/mlflow-3/ for what's new in MLflow 3.0.
We're happy to announce MLflow 3.0.0rc1!
We're happy to announce MLflow 3.0.0rc1!
You can upgrade with pip as usual:
pip install mlflow==3.0.0rc1
See https://mlflow.org/docs/3.0.0rc1/mlflow-3/ for what's new in MLflow 3.0.
We're happy to announce MLflow 3.0.0rc0!
We're happy to announce MLflow 3.0.0rc0!
You can upgrade with pip as usual:
pip install mlflow==3.0.0rc0
See https://mlflow.org/docs/3.0.0rc0/mlflow-3/ for what's new in MLflow 3.0.
Nothing published for this version
Version 2.22.4 is a patch release to backport several important fixes to MLflow 2.
Version 2.22.4 is a patch release to backport several important fixes to MLflow 2.
Nothing published for this version
Lightweight patch release to backport #15970 to v2.22.2.
Lightweight patch release to backport #15970 to v2.22.2.
MLflow 2.22.1 includes several major features and improvements
MLflow 2.22.1 includes several major features and improvements
Features:
Bug fixes:
Documentation updates:
Small bug fixes and documentation updates:
#15523, #15728, @TomeHirata; #13997, #16025, #15647, #16030, @harupy; #15786, @rahuja23; #15703, @joelrobin18; #15612, @serena-ruan; #16031, @daniellok-db; #15841, @frontsideair; #15807, @B-Step62
MLflow 2.22.0 brings important bug fixes and improves the UI and tracking capabilities.
MLflow 2.22.0 brings important bug fixes and improves the UI and tracking capabilities.
Features:
get_last_active_trace_id, which affects model serving/monitoring logic (#15233, @B-Step62)Bug Fixes:
langchain_tracer + added unit tests (#14971, @joelrobin18)": " were being truncated (#14896, @harupy)Small bug fixes and documentation updates:
#15396, #15379, #15292, #15305, #15078, #15251, #15267, #15208, #15104, #15045, #15084, #15055, #15056, #15048, #14946, #14956, #14903, #14854, #14830, @serena-ruan; #15417, #15256, #15186, #15007, @TomeHirata; #15119, @bbqiu; #15413, #15314, #15311, #15303, #15301, #15288, #15275, #15269, #15272, #15268, #15262, #15266, #15264, #15261, #15252, #15249, #15244, #15236, #15235, #15237, #15140, #14982, #14898, #14893, #14861, #14870, #14853, #14849, #14813, #14822, @harupy; #15333, #15298, #15300, #15156, #15019, #14957, @B-Step62; #15313, #15297, #14880, @daniellok-db; #15066, #15074, #14913, @joelrobin18; #15232, @kbolashev; #15242, @dbczumar; #15210, #15178, @WeichenXu123; #15187, #15177, @hubertzub-db; #15059, #15070, #15050, #15012, #14959, #14918, #15005, #14965, #14858, #14930, #14927, #14786, #14883, #14863, #14852, #14788, @Gumichocopengin8; #15134, #15129, #15120, #15117, #15002, #14997, #14996, #14998, #14975, #14874, @mlflow-automation; #14920, #14919, @jaceklaskowski
Nothing published for this version
MLflow 2.21.3 includes a few bug fixes and feature updates.
MLflow 2.21.3 includes a few bug fixes and feature updates.
Features:
return_type argument to mlflow.search_traces() API (#15085, @B-Step62)Bug fixes:
Documentation updates:
Small bug fixes and documentation updates:
#15205, @mlflow-app[bot]; #15184, #15157, #15137, @TomeHirata; #15118, @bbqiu; #15172, @harupy
MLflow 2.21.2 is a patch release that introduces minor features and bug fixes.
MLflow 2.21.2 is a patch release that introduces minor features and bug fixes.
MLflow 2.21.1 is a patch release that introduces minor features and addresses some minor bugs.
MLflow 2.21.1 is a patch release that introduces minor features and addresses some minor bugs.
Features:
--install-java option (#14868, @rgangopadhya)Bug fixes:
OTEL_EXPORTER_OTLP_PROTOCOL definition (#15008, @gabrielfu)Documentation updates:
Small bug fixes and documentation updates:
#15009, #14995, #15039, #15040, @TomeHirata; #15010, #15053, @B-Step62; #15014, #15025, #15030, #15050, #15070, @Gumichocopengin8; #15035, #15064, @joelrobin18; #15058, @serena-ruan; #14945, @turbotimon
[Models] Handle LangGraph breaking change (#14794, @B-Step62)
We are excited to announce the release of MLflow 2.21.0! This release includes a number of significant features, enhancements, and bug fixes.
Features:
Bug fixes:
ExperimentViewRunsControlsActionsSelectTags doesn't set loading state to false when set-tag request fails. (#14907, @harupy)": " get truncated (#14896, @harupy)mlflow.doctor to fall back to mlflow-skinny when mlflow is not found (#14782, @harupy)Documentation updates:
Small bug fixes and documentation updates:
#14994, #14992, #14990, #14979, #14964, #14969, #14944, #14948, #14957, #14958, #14942, #14940, #14935, #14929, #14805, #14876, #14833, #14748, #14744, #14666, #14668, #14664, #14667, #14580, #14475, #14439, #14397, #14363, #14361, #14377, #14378, #14337, #14324, #14339, #14259, @B-Step62; #14981, #14943, #14914, #14930, #14924, #14927, #14786, #14910, #14859, #14891, #14883, #14863, #14852, #14788, @Gumichocopengin8; #14946, #14978, #14956, #14906, #14903, #14854, #14860, #14857, #14824, #14830, #14767, #14772, #14770, #14766, #14651, #14629, #14636, #14572, #14498, #14328, #14265, @serena-ruan; #14989, #14895, #14880, #14878, #14866, #14821, #14817, #14815, #14765, #14803, #14773, #14783, #14784, #14776, #14759, #14541, #14553, #14540, #14499, #14495, #14481, #14479, #14456, #14022, #14411, #14407, #14408, #14315, #14346, #14325, #14322, #14326, #14310, #14309, #14320, #14308, @daniellok-db; #14986, #14904, #14898, #14893, #14861, #14870, #14853, #14849, #14813, #14822, #14818, #14802, #14804, #14814, #14779, #14796, #14735, #14731, #14728, #14734, #14727, #14726, #14721, #14719, #14716, #14692, #14683, #14687, #14684, #14674, #14673, #14662, #14652, #14650, #14648, #14647, #14646, #14639, #14637, #14635, #14634, #14633, #14630, #14628, #14624, #14623, #14621, #14619, #14615, #14613, #14603, #14601, #14600, #14597, #14570, #14564, #14554, #14551, #14550, #14515, #14529, #14528, #14525, #14516, #14514, #14486, #14476, #14472, #14477, #14364, #14431, #14414, #14398, #14412, #14399, #14359, #14369, #14381, #14349, #14350, #14347, #14348, #14342, #14329, #14250, #14318, #14323, #14306, #14280, #14279, #14272, #14270, #14263, #14222, @harupy; #14985, #14850, #14800, #14799, #14671, #14665, #14594, #14506, #14457, #14395, #14371, #14360, #14327, @TomeHirata; #14755, #14567, #14367, @bbqiu; #14892, @brilee; #14941, #14932, @hubertzub-db; #14913, @joelrobin18; #14756, @jiewpeng; #14701, @jaceklaskowski; #14568, #14450, @BenWilson2; #14535, @njbrake; #14507, @arunprd; #14489, @RuchitAgrawal; #14467, @seal07; #14460, @ManzoorAhmedShaikh; #14374, @wasup-yash; #14333, @singh-kristian; #14362, #14353, #14296, #13789, @dsuhinin; #14358, @apoxnen; #14335, @Fresnel-Fabian; #14178, @emmanuel-ferdman
The following features are marked for deprecation:
MLflow 2.21.0rc0 is a pre-release for testing out major features planned in the stable release. To install, run the following command:
pip install mlflow==2.21.0rc0
Please try it out and report any issues on the issue tracker!
The following features are marked for deprecation:
Nothing published for this version
MLflow 2.20.3 is a patch release includes several major features and improvements
MLflow 2.20.3 is a patch release includes several major features and improvements
Features:
Bug fixes:
astream_event API (#14598, @B-Step62)Small bug fixes and documentation updates:
#14640, #14574, #14593, @serena-ruan; #14338, #14693, #14664, #14663, #14377, @B-Step62; #14680, @JulesLandrySimard; #14388, #14685, @harupy; #14704, @brilee; #14698, #14658, @bbqiu; #14660, #14659, #14632, #14616, #14594, @TomeHirata; #14535, @njbrake
MLflow 2.20.2 is a patch release includes several bug fixes and features
MLflow 2.20.2 is a patch release includes several bug fixes and features
Features:
Bug fixes:
Documentation updates:
Small bug fixes and documentation updates:
#14410, #14569, #14440, @harupy; #14510, #14544, #14491, #14488, @bbqiu; #14518, @serena-ruan; #14517, #14500, #14461, #14478, @TomeHirata; #14512, @shaikmoeed; #14496, #14473, #14475, @B-Step62; #14467, @seal07; #14022, #14453, #14539, @daniellok-db; #14450, @BenWilson2; #14449, @SaiMadhavanG
MLflow 2.20.1 is a patch release includes several bug fixes and features:
MLflow 2.20.1 is a patch release includes several bug fixes and features:
Features:
Bug fixes:
Other small updates:
#14337, #14382, @B-Step62; #14356, @daniellok-db, #14354, @artjen, #14360, @TomuHirata,
We are excited to announce the release of MLflow 2.20.0! This release includes a number of significant features, enhancements, and bug fixes.
We are excited to announce the release of MLflow 2.20.0! This release includes a number of significant features, enhancements, and bug fixes.
💡Type Hint-Based Model Signature: Define your model's signature in the most Pythonic way. MLflow now supports defining a model signature based on the type hints in your PythonModel's predict function, and validating input data payloads against it. (#14182, #14168, #14130, #14100, #14099, @serena-ruan)
🧠 Bedrock / Groq Tracing Support: MLflow Tracing now offers a one-line auto-tracing experience for Amazon Bedrock and Groq LLMs. Track LLM invocation within your model by simply adding mlflow.bedrock.tracing or mlflow.groq.tracing call to the code. (#14018, @B-Step62, #14006, @anumita0203)
🗒️ Inline Trace Rendering in Jupyter Notebook: MLflow now supports rendering a trace UI within the notebook where you are running models. This eliminates the need to frequently switch between the notebook and browser, creating a seamless local model debugging experience. Check out this blog post for a quick demo! (#13955, @daniellok-db)
⚡️Faster Model Validation with uv Package Manager: MLflow has adopted uv, a new Rust-based, super-fast Python package manager. This release adds support for the new package manager in the mlflow.models.predict API, enabling faster model environment validation. Stay tuned for more updates! (#13824, @serena-ruan)
🖥️ New Chat Panel in Trace UI: THe MLflow Trace UI now shows a unified chat panel for LLM invocations. The update allows you to view chat messages and function calls in a rich and consistent UI across LLM providers, as well as inspect the raw input and output payloads. (#14211, @TomuHirata)
Other Features:
ChatAgent base class for defining custom python agent (#13797, @bbqiu)context parameter optional for calling PythonModel instance (#14059, @serena-ruan)ChatModel (#14068, @stevenchen-db)Bug fixes:
log_image (#14281, @TomeHirata)loaded_model variable (#14109, @yang-chengg)DatabricksSDKModelsArtifactRepository.list_artifacts is called on a file (#14027, @shichengzhou-db)Documentation updates:
Small bug fixes and documentation updates:
#14294, #14252, #14233, #14205, #14217, #14172, #14188, #14167, #14166, #14163, #14162, #14161, #13971, @TomeHirata; #14299, #14280, #14279, #14278, #14272, #14270, #14268, #14269, #14263, #14258, #14222, #14248, #14128, #14112, #14111, #14093, #14096, #14095, #14090, #14089, #14085, #14078, #14074, #14070, #14053, #14060, #14035, #14014, #14002, #14000, #13997, #13996, #13995, @harupy; #14298, #14286, #14249, #14276, #14259, #14242, #14254, #14232, #14207, #14206, #14185, #14196, #14193, #14173, #14164, #14159, #14165, #14152, #14151, #14126, #14069, #13987, @B-Step62; #14295, #14265, #14271, #14262, #14235, #14239, #14234, #14228, #14227, #14229, #14218, #14216, #14213, #14208, #14204, #14198, #14187, #14181, #14177, #14176, #14156, #14169, #14099, #14086, #13983, @serena-ruan; #14155, #14067, #14140, #14132, #14072, @daniellok-db; #14178, @emmanuel-ferdman; #14247, @dbczumar; #13789, #14108, @dsuhinin; #14212, @aravind-segu; #14223, #14191, #14084, @dsmilkov; #13804, @kriscon-db; #14158, @Lodewic; #14148, #14147, #14115, #14079, #14116, @WeichenXu123; #14135, @brilee; #14133, @manos02; #14121, @LeahKorol; #14025, @nojaf; #13948, @benglewis; #13942, @justsomerandomdude264; #14003, @Ajay-Satish-01; #13982, @prithvikannan; #13638, @MaxwellSalmon
Your coding agent can read these notes before it upgrades. Set up the MCP server →