diff --git a/README.md b/README.md
index d3570c89f..f3bcb398d 100644
--- a/README.md
+++ b/README.md
@@ -155,7 +155,7 @@ Hardware specific features 🔥
 </summary>
 <br />
 
-It is no secret that deep learning is mosly relying on matrix multiplication as its core operation,
+It is no secret that deep learning is mostly relying on matrix multiplication as its core operation,
 since this is how fully-connected neural networks are modeled.
 
 More and more, hardware manufacturers optimize their chips specifically for matrix multiplication
diff --git a/backend-comparison/README.md b/backend-comparison/README.md
index 43f5ca549..7db17dfe2 100644
--- a/backend-comparison/README.md
+++ b/backend-comparison/README.md
@@ -78,7 +78,7 @@ Running benchmarks...
 
 By default `burnbench` uses a compact output with a progress bar which hides the
 compilation logs and benchmarks results as they are executed. If a benchmark
-failed to run, the `--verbose` flag can be use to investigate the error.
+failed to run, the `--verbose` flag can be used to investigate the error.
 
 #### Authentication and benchmarks sharing
 
diff --git a/xtask/src/utils/mdbook.rs b/xtask/src/utils/mdbook.rs
index 8c4e8162d..fed0922d2 100644
--- a/xtask/src/utils/mdbook.rs
+++ b/xtask/src/utils/mdbook.rs
@@ -8,7 +8,7 @@ use crate::utils::process::handle_child_process;
 
 use super::Params;
 
-/// Run an mdbook command with the passed directory as the current directory
+/// Run a mdbook command with the passed directory as the current directory
 pub(crate) fn run_mdbook_with_path<P: AsRef<Path>>(
     command: &str,
     params: Params,