diff --git a/docs/source/methods.rst b/docs/source/methods.rst
index 560eaeae..95ad5962 100644
--- a/docs/source/methods.rst
+++ b/docs/source/methods.rst
@@ -29,12 +29,16 @@ Posterior approximation methods
     taken by averaging checkpoints over the stochastic optimization trajectory. The covariance is also estimated
     empirically along the trajectory, and it is made of a diagonal component and a low-rank non-diagonal one.
 
+- **Hamiltonian Monte Carlo (HMC)** `[Neal, 2010] <https://arxiv.org/pdf/1206.1901.pdf>`_
+    HMC approximates the posterior as a steady-state distribution of a Monte Carlo Markov chain with Hamiltonian dynamics.
+    After the initial "burn-in" phase, each step of the chain generates a sample from the posterior. HMC is typically applied
+    in the full-batch scenario.
+
 - **Stochastic Gradient Hamiltonian Monte Carlo (SGHMC)** `[Chen et al., 2014] <http://proceedings.mlr.press/v32/cheni14.pdf>`_
-    SGHMC approximates the posterior as a steady-state distribution of a Monte Carlo Markov chain with Hamiltonian dynamics.
-    After the initial "burn-in" phase, each step of the chain generates samples from the posterior.
+    SGHMC implements a variant of HMC algorithm that expects noisy gradient estimate computed on mini-batches of data.
 
 - **Cyclical Stochastic Gradient Langevin Dynamics (Cyclical SGLD)** `[Zhang et al., 2020] <https://openreview.net/pdf?id=rkeS1RVtPS>`_
-    Cyclical SGLD adapts the cyclical cosine step size schedule, and alternates between *exploration* and *sampling* stages to better
+    Cyclical SGLD adopts the cyclical cosine step size schedule, and alternates between *exploration* and *sampling* stages to better
     explore the multimodal posteriors for deep neural networks.
 
 Parametric calibration methods
diff --git a/docs/source/references/prob_model/posterior/sgmcmc.rst b/docs/source/references/prob_model/posterior/sgmcmc.rst
index 2ab3ce31..c4391486 100644
--- a/docs/source/references/prob_model/posterior/sgmcmc.rst
+++ b/docs/source/references/prob_model/posterior/sgmcmc.rst
@@ -4,12 +4,37 @@ SG-MCMC procedures approximate the posterior as a steady-state distribution of
 a Monte Carlo Markov chain, that utilizes noisy estimates of the gradient
 computed on minibatches of data.
 
+Hamiltonian Monte Carlo (HMC)
+=============================
+
+HMC `[Neal, 2010] <https://arxiv.org/pdf/1206.1901.pdf>`_ is a MCMC sampling
+algorithm that simulates a Hamiltonian dynamical system to rapidly explores
+the posterior.
+
+.. autoclass:: fortuna.prob_model.posterior.sgmcmc.hmc.hmc_approximator.HMCPosteriorApproximator
+
+.. autoclass:: fortuna.prob_model.posterior.sgmcmc.hmc.hmc_posterior.HMCPosterior
+    :show-inheritance:
+    :no-inherited-members:
+    :exclude-members: state
+    :members: fit, sample, load_state, save_state
+
+.. autoclass:: fortuna.prob_model.posterior.sgmcmc.hmc.hmc_state.HMCState
+    :show-inheritance:
+    :no-inherited-members:
+    :inherited-members: init, init_from_dict
+    :members: convert_from_map_state
+    :exclude-members: params, mutable, calib_params, calib_mutable, replace, apply_gradients, encoded_name, create
+    :no-undoc-members:
+    :no-special-members:
+
+
 Stochastic Gradient Hamiltonian Monte Carlo (SGHMC)
 ===================================================
 
 SGHMC `[Chen T. et al., 2014] <http://proceedings.mlr.press/v32/cheni14.pdf>`_
 is a popular MCMC algorithm that uses stochastic gradient estimates to scale
-to large datasets.
+HMC to large datasets.
 
 .. autoclass:: fortuna.prob_model.posterior.sgmcmc.sghmc.sghmc_approximator.SGHMCPosteriorApproximator