Automatic inference for composable models of heterogeneous data
Mixle is a Python framework for building probabilistic systems over data
that is mixed, structured, temporal, neural, or produced by language
models. Its center is a composable distribution and estimator contract:
describe the model shape, fit it through a common inference surface, and
carry the fitted object into scoring, sampling, calibration, audit, and
deployment workflows.
The long-term direction is broader than a distribution catalog. Mixle is
being developed as a unified modeling layer for heterogeneous records,
mixtures, HMMs, Transformer leaves, task distillation, uncertainty-aware
LLM systems, local reasoning workflows, design of experiments,
structured decisions, and production evidence.
Start with the quickstart
Read the maturity guide
Find an API
What Mixle Provides
-------------------
Mixle is designed for applications where a single estimator class is not enough.
One workflow may include a structured probability model, a latent state model,
a calibrated decision rule, and a neural or LLM component. Mixle keeps those
pieces connected through explicit model structure and capability-checked
operations.
At the stable center are distribution families, estimators, samplers, encoders,
combinators, mixtures, HMMs, and the ``optimize`` inference entry point. Around
that center are newer workflow layers for automatic model recommendation,
probabilistic-programming expressions, neural leaves, task replacement, LLM
uncertainty, design loops, model evolution, and production metadata. Maturity
is called out where it matters so users can choose the right validation
standard.
Core Principles
---------------
Composition over special cases
Build larger models from distributional components instead of writing a new
training loop for every data shape.
Inference from structure
The estimator or prototype defines the route: direct estimation, EM,
conjugate updates, gradient fitting, Bayesian objectives, PPL lowering, or
calibrated task workflows.
Operational uncertainty
Posterior queries, conformal prediction, semantic entropy, density gates,
abstention policies, and escalation rules are modeled as part of system
behavior rather than post-processing.
Inspectable automation
Automatic estimator selection, model recommendation, and LLM-designed
specifications expose assumptions, validation checks, confidence gaps, and
fallback behavior.
Common Workflows
----------------
Fit a heterogeneous probability model
Start with :doc:`quickstart`, then read :doc:`concepts`,
:doc:`distributions`, :doc:`stats-structured`, and :doc:`hmms-latent`.
Infer a first model from raw data
Read :doc:`automatic-inference` and
:doc:`automatic-modeling-internals`. These pages cover
``get_estimator(data)``, prototype-driven fitting, model recommendation,
validation, and fallback behavior.
Work with neural or language-model components
Read :doc:`neural-llm`, :doc:`representation`,
:doc:`task-distillation`, :doc:`task-serving`, and
:doc:`agentic-task-distillation`.
Add uncertainty to LLM or reasoning systems
Read :doc:`uncertainty` and :doc:`reasoning-systems` for semantic entropy,
claim reliability, calibrated abstention, graph-producing LLMs, and
cross-modal evidence.
Scale, serve, or audit fitted models
Read :doc:`engines`, :doc:`compute-layer`,
:doc:`utilities-and-parallelism`, :doc:`data`, :doc:`production`, and
:doc:`lifecycle`.
Build a local reasoning workflow
Read :doc:`reasoning-ecosystem` for substrate storage, skills, reasoner
actions, pool jobs, telemetry, and the optional ``Scientist`` workflow.
Explore scientific design and structured decisions
Read :doc:`doe`, :doc:`analysis`, :doc:`evolution`,
:doc:`relations`, :doc:`operations`, and :doc:`enumeration`.
Minimal Example
---------------
The public fitting surface accepts an estimator, a prototype distribution, or
an estimator inferred from data:
.. code-block:: python
from mixle.inference import optimize
from mixle.stats import GaussianDistribution, MixtureDistribution
from mixle.utils.automatic import get_estimator
reals = [-1.2, -0.9, -1.1, 0.8, 1.2, 1.1]
rows = [("free", 4), ("paid", 19), ("free", 5), ("paid", 23)]
proto = MixtureDistribution(
[GaussianDistribution(-1.0, 1.0), GaussianDistribution(1.0, 1.0)],
[0.5, 0.5],
)
fitted_from_shape = optimize(reals, proto, prev_estimate=proto, out=None)
inferred_estimator = get_estimator(rows)
fitted_from_data = optimize(rows, inferred_estimator, out=None)
Passing ``proto`` as the model argument gives ``optimize`` the family shape.
Passing the same object as ``prev_estimate`` also uses its parameter values as
the starting point, which is usually what you want for a mixture example. See
:doc:`automatic-inference` for the full route.
Project Direction
-----------------
Mixle is moving toward a single modeling interface for hybrid probabilistic,
neural, task, and reasoning systems. The stable distribution library remains
the foundation. The forward-looking work is the connective tissue: automatic
model design, uncertainty-aware LLM behavior, neural leaves, representation
learning, active design, self-improvement loops, structured decisions, and
production evidence carried through the same inspectable model lifecycle.
The standard is practical: if a system can be described as a composition of
evidence, latent structure, learned components, and calibrated decisions, Mixle
should make it possible to fit, inspect, compare, and deploy that system without
breaking the abstraction apart.
Manual Map
----------
The foundational pages are :doc:`installation`, :doc:`quickstart`,
:doc:`concepts`, :doc:`maturity`, and :doc:`package-map`. The tutorial index in
:doc:`tutorials/index` provides task-sized walkthroughs. The generated
reference under :doc:`api/modules` is exhaustive; :doc:`api-overview` is the
human map for finding the right import.
.. toctree::
:caption: Start Here
:hidden:
:maxdepth: 2
installation
maturity
whats-new-0-6-2
quickstart
concepts
package-map
lifecycle
tutorials/index
.. toctree::
:caption: Core Workflows
:hidden:
:maxdepth: 2
neural-llm
automatic-inference
models
representation
task-distillation
task-serving
agentic-task-distillation
uncertainty
reasoning-systems
reasoning-ecosystem
hmms-latent
processes
automatic-modeling-internals
cookbook
.. toctree::
:caption: Reference Guides
:hidden:
:maxdepth: 2
api-overview
capabilities-contracts
compute-layer
distributions
stats-univariate
stats-structured
stats-latent-bayes
inference
inference-toolkit
ppl
operations
relations
engines
enumeration
data
doe
analysis
evolution
production
utilities-and-parallelism
experimental-program
examples
troubleshooting
glossary
extending
development
.. toctree::
:caption: API Reference
:hidden:
:maxdepth: 2
api/modules
.. toctree::
:caption: Architecture Notes
:hidden:
:maxdepth: 2
design-notes