diff --git a/pymc/sampling/jax.py b/pymc/sampling/jax.py
index 6c01192825..2ab2f20280 100644
--- a/pymc/sampling/jax.py
+++ b/pymc/sampling/jax.py
@@ -66,6 +66,8 @@
 __all__ = (
     "get_jaxified_graph",
     "get_jaxified_logp",
+    "sample_blackjax_adjusted_mclmc",
+    "sample_blackjax_mclmc",
     "sample_blackjax_nuts",
     "sample_numpyro_nuts",
 )
@@ -259,8 +261,14 @@ def _blackjax_inference_loop(
         algorithm = blackjax.nuts
     elif algorithm_name == "hmc":
         algorithm = blackjax.hmc
+    elif algorithm_name == "mclmc":
+        algorithm = blackjax.mclmc
+    elif algorithm_name == "adjusted_mclmc":
+        algorithm = blackjax.adjusted_mclmc_dynamic
     else:
-        raise ValueError("Only supporting 'nuts' or 'hmc' as algorithm to draw samples.")
+        raise ValueError(
+            "Only supporting 'nuts', 'hmc', 'mclmc', or 'adjusted_mclmc' as algorithm to draw samples."
+        )
 
     adapt = blackjax.window_adaptation(
         algorithm=algorithm,
@@ -726,3 +734,44 @@ def sample_jax_nuts(
 
 sample_numpyro_nuts = partial(sample_jax_nuts, nuts_sampler="numpyro")
 sample_blackjax_nuts = partial(sample_jax_nuts, nuts_sampler="blackjax")
+
+
+# Custom partial functions for the MCLMC samplers
+def sample_blackjax_mclmc(*args, **kwargs):
+    """
+    Draw samples from the posterior using the MCLMC (Microcanonical Langevin Monte Carlo) method.
+
+    From the ``blackjax`` library.
+
+    Parameters are the same as for sample_jax_nuts.
+
+    MCLMC is based on https://arxiv.org/abs/2212.08549. It numerically integrates a specialized SDE
+    to produce samples from the target distribution.
+
+    This implementation uses the unadjusted MCLMC algorithm, which may have some asymptotic bias
+    but is typically faster than adjusted MCLMC or NUTS for many high-dimensional problems.
+    """
+    kwargs.setdefault("nuts_sampler_kwargs", {}).setdefault("algorithm", "mclmc")
+    # MCLMC has different default target_accept
+    if "target_accept" not in kwargs:
+        kwargs["target_accept"] = 0.9
+    return sample_jax_nuts(*args, nuts_sampler="blackjax", **kwargs)
+
+
+def sample_blackjax_adjusted_mclmc(*args, **kwargs):
+    """
+    Draw samples from the posterior using the adjusted MCLMC (Microcanonical Langevin Monte Carlo) method.
+
+    From the ``blackjax`` library.
+
+    Parameters are the same as for sample_jax_nuts.
+
+    Adjusted MCLMC adds a Metropolis-Hastings correction step to the unadjusted MCLMC algorithm,
+    ensuring asymptotic unbiasedness at the cost of some computational efficiency.
+    It is recommended when exact asymptotic convergence to the posterior is required.
+    """
+    kwargs.setdefault("nuts_sampler_kwargs", {}).setdefault("algorithm", "adjusted_mclmc")
+    # Adjusted MCLMC also has different default target_accept
+    if "target_accept" not in kwargs:
+        kwargs["target_accept"] = 0.9
+    return sample_jax_nuts(*args, nuts_sampler="blackjax", **kwargs)
diff --git a/pymc/sampling/mcmc.py b/pymc/sampling/mcmc.py
index f2dfa6e9c2..d8a96acb7b 100644
--- a/pymc/sampling/mcmc.py
+++ b/pymc/sampling/mcmc.py
@@ -292,6 +292,105 @@ def all_continuous(vars):
         return True
 
 
+def _sample_mclmc(
+    sampler: Literal["blackjax"],
+    draws: int,
+    tune: int,
+    chains: int,
+    target_accept: float,
+    random_seed: RandomState | None,
+    initvals: StartDict | Sequence[StartDict | None] | None,
+    model: Model,
+    var_names: Sequence[str] | None,
+    progressbar: bool,
+    idata_kwargs: dict | None,
+    compute_convergence_checks: bool,
+    mclmc_sampler_kwargs: dict | None,
+    adjusted: bool = False,
+    **kwargs,
+):
+    """
+    Sample using the MCLMC sampler from BlackJax.
+
+    Parameters
+    ----------
+    sampler : {"blackjax"}
+        Name of the external sampler to use.
+    draws : int
+        Number of samples to draw.
+    tune : int
+        Number of tuning steps.
+    chains : int
+        Number of chains to run in parallel.
+    target_accept : float
+        Target acceptance rate.
+    random_seed : RandomState, optional
+        Random seed.
+    initvals : dict or list of dict, optional
+        Initial values for variables.
+    model : Model
+        PyMC model.
+    var_names : list of str, optional
+        List of variable names to sample.
+    progressbar : bool
+        Whether to display a progress bar.
+    idata_kwargs : dict, optional
+        Keyword arguments for InferenceData conversion.
+    compute_convergence_checks : bool
+        Whether to compute convergence checks.
+    mclmc_sampler_kwargs : dict, optional
+        Keyword arguments for the MCLMC sampler.
+    adjusted : bool, default=False
+        Whether to use the adjusted MCLMC algorithm.
+
+    Returns
+    -------
+    InferenceData
+        ArviZ InferenceData object with sampling results.
+    """
+    if mclmc_sampler_kwargs is None:
+        mclmc_sampler_kwargs = {}
+
+    if sampler == "blackjax":
+        import pymc.sampling.jax as pymc_jax
+
+        if adjusted:
+            idata = pymc_jax.sample_blackjax_adjusted_mclmc(
+                draws=draws,
+                tune=tune,
+                chains=chains,
+                target_accept=target_accept,
+                random_seed=random_seed,
+                initvals=initvals,
+                model=model,
+                var_names=var_names,
+                progressbar=progressbar,
+                idata_kwargs=idata_kwargs,
+                compute_convergence_checks=compute_convergence_checks,
+                **mclmc_sampler_kwargs,
+            )
+        else:
+            idata = pymc_jax.sample_blackjax_mclmc(
+                draws=draws,
+                tune=tune,
+                chains=chains,
+                target_accept=target_accept,
+                random_seed=random_seed,
+                initvals=initvals,
+                model=model,
+                var_names=var_names,
+                progressbar=progressbar,
+                idata_kwargs=idata_kwargs,
+                compute_convergence_checks=compute_convergence_checks,
+                **mclmc_sampler_kwargs,
+            )
+        return idata
+    else:
+        raise ValueError(
+            f"Sampler {sampler} not found. Currently only 'blackjax' is supported for MCLMC."
+        )
+
+
 def _sample_external_nuts(
     sampler: Literal["nutpie", "numpyro", "blackjax"],
     draws: int,
@@ -429,6 +528,7 @@ def sample(
     step=None,
     var_names: Sequence[str] | None = None,
     nuts_sampler: Literal["pymc", "nutpie", "numpyro", "blackjax"] = "pymc",
+    mclmc_sampler: Literal["blackjax"] | None = None,
     initvals: StartDict | Sequence[StartDict | None] | None = None,
     init: str = "auto",
     jitter_max_retries: int = 10,
@@ -440,6 +540,7 @@ def sample(
     return_inferencedata: Literal[True] = True,
     idata_kwargs: dict[str, Any] | None = None,
     nuts_sampler_kwargs: dict[str, Any] | None = None,
+    mclmc_sampler_kwargs: dict[str, Any] | None = None,
     callback=None,
     mp_ctx=None,
     blas_cores: int | None | Literal["auto"] = "auto",
@@ -461,6 +562,7 @@ def sample(
     step=None,
     var_names: Sequence[str] | None = None,
     nuts_sampler: Literal["pymc", "nutpie", "numpyro", "blackjax"] = "pymc",
+    mclmc_sampler: Literal["blackjax"] | None = None,
     initvals: StartDict | Sequence[StartDict | None] | None = None,
     init: str = "auto",
     jitter_max_retries: int = 10,
@@ -472,6 +574,7 @@ def sample(
     return_inferencedata: Literal[False],
     idata_kwargs: dict[str, Any] | None = None,
     nuts_sampler_kwargs: dict[str, Any] | None = None,
+    mclmc_sampler_kwargs: dict[str, Any] | None = None,
     callback=None,
     mp_ctx=None,
     model: Model | None = None,
@@ -493,6 +596,7 @@ def sample(
     step=None,
     var_names: Sequence[str] | None = None,
     nuts_sampler: Literal["pymc", "nutpie", "numpyro", "blackjax"] = "pymc",
+    mclmc_sampler: Literal["blackjax"] | None = None,
     initvals: StartDict | Sequence[StartDict | None] | None = None,
     init: str = "auto",
     jitter_max_retries: int = 10,
@@ -504,6 +608,7 @@ def sample(
     return_inferencedata: bool = True,
     idata_kwargs: dict[str, Any] | None = None,
     nuts_sampler_kwargs: dict[str, Any] | None = None,
+    mclmc_sampler_kwargs: dict[str, Any] | None = None,
     callback=None,
     mp_ctx=None,
     blas_cores: int | None | Literal["auto"] = "auto",
@@ -823,6 +928,33 @@ def joined_blas_limiter():
                 **kwargs,
             )
 
+    if mclmc_sampler is not None:
+        if not exclusive_nuts:
+            raise ValueError(
+                "Model can not be sampled with MCLMC alone. It either has discrete variables or a non-differentiable log-probability."
+            )
+
+        adjusted = mclmc_sampler_kwargs.pop("adjusted", False) if mclmc_sampler_kwargs else False
+
+        with joined_blas_limiter():
+            return _sample_mclmc(
+                sampler=mclmc_sampler,
+                draws=draws,
+                tune=tune,
+                chains=chains,
+                target_accept=kwargs.pop("nuts", {}).get("target_accept", 0.9),
+                random_seed=random_seed,
+                initvals=initvals,
+                model=model,
+                var_names=var_names,
+                progressbar=progress_bool,
+                idata_kwargs=idata_kwargs,
+                compute_convergence_checks=compute_convergence_checks,
+                mclmc_sampler_kwargs=mclmc_sampler_kwargs,
+                adjusted=adjusted,
+                **kwargs,
+            )
+
     if exclusive_nuts and not provided_steps:
         # Special path for NUTS initialization
         if "nuts" in kwargs:
diff --git a/tests/sampling/test_mcmc_external.py b/tests/sampling/test_mcmc_external.py
index 2d32277061..b5a44c1d6e 100644
--- a/tests/sampling/test_mcmc_external.py
+++ b/tests/sampling/test_mcmc_external.py
@@ -86,3 +86,34 @@ def test_step_args():
         )
 
     npt.assert_almost_equal(idata.sample_stats.acceptance_rate.mean(), 0.5, decimal=1)
+
+
+def test_mclmc_sampling():
+    """Test that blackjax MCLMC sampling works."""
+    pytest.importorskip("blackjax")
+    with Model():
+        a = Normal("a", 0, 1)
+        trace = sample(
+            draws=50,
+            tune=50,
+            random_seed=345,
+            mclmc_sampler="blackjax",
+            progressbar=False,
+        )
+    assert "a" in trace.posterior.data_vars
+
+
+def test_adjusted_mclmc_sampling():
+    """Test that blackjax adjusted MCLMC sampling works."""
+    pytest.importorskip("blackjax")
+    with Model():
+        a = Normal("a", 0, 1)
+        trace = sample(
+            draws=50,
+            tune=50,
+            random_seed=345,
+            mclmc_sampler="blackjax",
+            mclmc_sampler_kwargs={"adjusted": True},
+            progressbar=False,
+        )
+    assert "a" in trace.posterior.data_vars