diff --git a/README.md b/README.md
index 047ac3767..ffcd915b2 100644
--- a/README.md
+++ b/README.md
@@ -34,13 +34,13 @@ Check out our interactive [demo](https://aphp.github.io/edsnlp/demo/) !
 You can install EDS-NLP via `pip`. We recommend pinning the library version in your projects, or use a strict package manager like [Poetry](https://python-poetry.org/).
 
 ```shell
-pip install edsnlp==0.15.0
+pip install edsnlp==0.16.0
 ```
 
 or if you want to use the trainable components (using pytorch)
 
 ```shell
-pip install "edsnlp[ml]==0.15.0"
+pip install "edsnlp[ml]==0.16.0"
 ```
 
 ### A first pipeline
diff --git a/changelog.md b/changelog.md
index 9e752f924..41aad3cee 100644
--- a/changelog.md
+++ b/changelog.md
@@ -1,12 +1,12 @@
 # Changelog
 
-## Unreleased
+## v0.16.0 (2025-0.3-26)
 
 ### Added
-- Hyperparameter Tuning for EDS-NLP: Introduced a new script `edsnlp.tune` for hyperparameter tuning using Optuna. This feature allows users to efficiently optimize model parameters with options for single-phase or two-phase tuning strategies. Includes support for parameter importance analysis, visualization, pruning, and automatic handling of GPU time budgets.
-- Provided a [detailed tutorial](./docs/tutorials/tuning.md) on hyperparameter tuning, covering usage scenarios and configuration options.
+- Hyperparameter Tuning for EDS-NLP: introduced a new script `edsnlp.tune` for hyperparameter tuning using Optuna. This feature allows users to efficiently optimize model parameters with options for single-phase or two-phase tuning strategies. Includes support for parameter importance analysis, visualization, pruning, and automatic handling of GPU time budgets.
+- Provided a [detailed tutorial](https://aphp.github.io/edsnlp/v0.16.0/tutorials/tuning/) on hyperparameter tuning, covering usage scenarios and configuration options.
 - `ScheduledOptimizer` (e.g., `@core: "optimizer"`) now supports importing optimizers using their qualified name (e.g., `optim: "torch.optim.Adam"`).
-- `edsnlp/ner_crf.` now compute confidence score on spans.
+- `eds.ner_crf` now computes confidence score on spans.
 
 ### Changed
 
@@ -25,7 +25,7 @@
   - Raise an error if the batch size in `stream.shuffle(batch_size=...)` is not compatible with the stream
 - `eds.split` now keeps doc and span attributes in the sub-documents.
 
-# v0.15.0 (2024-12-13)
+## v0.15.0 (2024-12-13)
 
 ### Added
 
diff --git a/docs/index.md b/docs/index.md
index 3397951e2..7e8d7602b 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -15,13 +15,13 @@ Check out our interactive [demo](https://aphp.github.io/edsnlp/demo/) !
 You can install EDS-NLP via `pip`. We recommend pinning the library version in your projects, or use a strict package manager like [Poetry](https://python-poetry.org/).
 
 ```{: data-md-color-scheme="slate" }
-pip install edsnlp==0.15.0
+pip install edsnlp==0.16.0
 ```
 
 or if you want to use the trainable components (using pytorch)
 
 ```{: data-md-color-scheme="slate" }
-pip install "edsnlp[ml]==0.15.0"
+pip install "edsnlp[ml]==0.16.0"
 ```
 
 ### A first pipeline
diff --git a/docs/tutorials/training.md b/docs/tutorials/training.md
index 3ec95f1f9..31e105ae6 100644
--- a/docs/tutorials/training.md
+++ b/docs/tutorials/training.md
@@ -37,7 +37,7 @@ readme = "README.md"
 requires-python = ">3.7.1,<4.0"
 
 dependencies = [
-    "edsnlp[ml]>=0.15.0",
+    "edsnlp[ml]>=0.16.0",
     "sentencepiece>=0.1.96"
 ]
 
diff --git a/docs/tutorials/tuning.md b/docs/tutorials/tuning.md
index d0189d839..f5cf12c2a 100644
--- a/docs/tutorials/tuning.md
+++ b/docs/tutorials/tuning.md
@@ -35,7 +35,7 @@ readme = "README.md"
 requires-python = ">3.7.1,<4.0"
 
 dependencies = [
-    "edsnlp[ml]>=0.15.0",
+    "edsnlp[ml]>=0.16.0",
     "sentencepiece>=0.1.96",
     "optuna>=4.0.0",
     "plotly>=5.18.0",
diff --git a/edsnlp/__init__.py b/edsnlp/__init__.py
index a2391cc3c..373f9b69c 100644
--- a/edsnlp/__init__.py
+++ b/edsnlp/__init__.py
@@ -15,7 +15,7 @@
 import edsnlp.pipes
 from . import reducers
 
-__version__ = "0.15.0"
+__version__ = "0.16.0"
 
 BASE_DIR = Path(__file__).parent
 
diff --git a/pyproject.toml b/pyproject.toml
index 217c38527..f0d016bd5 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -76,7 +76,8 @@ ml = [
     "rich-logger>=0.3.1",
     "torch>=1.13.0",
     "foldedtensor>=0.3.4",
-    "safetensors>=0.3.0",
+    "safetensors>=0.3.0; python_version>='3.8'",
+    "safetensors>=0.3.0,<0.5.0; python_version<'3.8'",
     "transformers>=4.0.0,<5.0.0",
     "accelerate>=0.20.3,<1.0.0",
 ]