diff --git a/environments/code_execution_server/lcb_modal_endpoint.py b/environments/code_execution_server/lcb_modal_endpoint.py
index 6bdfcdf6..ca627532 100644
--- a/environments/code_execution_server/lcb_modal_endpoint.py
+++ b/environments/code_execution_server/lcb_modal_endpoint.py
@@ -452,9 +452,9 @@ def grade_stdio(
             if stripped_prediction_line == stripped_gt_out_line:
                 continue
 
-            # CASE 2: element-wise comparision
+            # CASE 2: element-wise comparison
             # if there are floating elements
-            # use `decimal` library for good floating point comparision
+            # use `decimal` library for good floating point comparison
             # otherwise gotcha: np.isclose(50000000000000000, 50000000000000001) = True
             # note that we should always be able to convert to decimals
 
diff --git a/environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/engine/run.py b/environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/engine/run.py
index 7535e314..c33970ea 100644
--- a/environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/engine/run.py
+++ b/environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/engine/run.py
@@ -292,7 +292,7 @@ class Run:
             self._generate_configs()
         else:
             raise TypeError(
-                f"Unsupportet type for 'resume', got {type(self.engine.resume)=}."
+                f"Unsupported type for 'resume', got {type(self.engine.resume)=}."
             )
 
     def _calc_max_steps(self) -> int:
diff --git a/environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/evaluation/plot.py b/environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/evaluation/plot.py
index 76fffe55..5515beca 100644
--- a/environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/evaluation/plot.py
+++ b/environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/evaluation/plot.py
@@ -375,7 +375,7 @@ def create_figure(dataframe_list: list[pd.DataFrame], config: AttributeDict):
     if num_cols == 1 and n_rows_max > 1:
         figsize = (2**3 * scale, 2 * 3 * n_rows_max * scale)
     elif num_cols == 2:
-        # TODO: after removing cbar from left subifgure, it is squished
+        # TODO: after removing cbar from left subfigure, it is squished
         #       there is an argument to share the legend, we should use that
         figsize = (12 * scale, 5.4 * n_rows_max * scale)
     elif num_cols > 2:
diff --git a/environments/community/ufc_prediction_env/ufc_image_env.py b/environments/community/ufc_prediction_env/ufc_image_env.py
index 7837b57f..867c0a7b 100644
--- a/environments/community/ufc_prediction_env/ufc_image_env.py
+++ b/environments/community/ufc_prediction_env/ufc_image_env.py
@@ -154,7 +154,7 @@ class UFCImageEnv(BaseEnv):
                 "Let's break down this matchup that's got everyone talking!\n\n"
                 "In the red corner, we have:(YOUR FIRST IMAGE):\n"
                 "And in the blue corner: (YOUR SECOND IMAGE):\n\n"
-                "Now, act as your favorite fight comentator, I want you to:\n"
+                "Now, act as your favorite fight commentator, I want you to:\n"
                 "create a fight commentary of whats happening in the fight live\n"
                 "Give us your best fight commentary! Make it exciting, make it dramatic, "
                 "make it sound like you're calling the fight live! "
diff --git a/environments/community/ufc_prediction_env/ufc_predictor_ui.py b/environments/community/ufc_prediction_env/ufc_predictor_ui.py
index 773d72a6..60178b7b 100644
--- a/environments/community/ufc_prediction_env/ufc_predictor_ui.py
+++ b/environments/community/ufc_prediction_env/ufc_predictor_ui.py
@@ -53,7 +53,7 @@ def predict():
             "Let's break down this matchup that's got everyone talking!\n\n"
             "In the red corner, we have:(YOUR FIRST IMAGE):\n"
             "And in the blue corner: (YOUR SECOND IMAGE):\n\n"
-            "Now, as your favorite fight comentator, I want you to:\n"
+            "Now, as your favorite fight commentator, I want you to:\n"
             "create a fight commentary of whats happening in the fight live\n"
             "Give us your best fight commentary! Make it exciting, make it dramatic, "
             "make it sound like you're calling the fight live! "
diff --git a/environments/community/word_hunt/README.md b/environments/community/word_hunt/README.md
index c4f53789..379f56ab 100644
--- a/environments/community/word_hunt/README.md
+++ b/environments/community/word_hunt/README.md
@@ -52,7 +52,7 @@ Found words:
   - 4-letter words: 400 points
   - 5-letter words: 800 points
   - 6+ letter words: 1400 + (400 × (length - 6)) points
-  - These scoring rules are based on the version of Word Hunt in GamePidgeon, the iOS app
+  - These scoring rules are based on the version of Word Hunt in GamePigeon, the iOS app
 
   The default maximum output tokens a response will produce is 100 tokens, which simulates the time pressure and encourages strategic resource allocation.
 
diff --git a/environments/eval_environments/ifeval_instructions/instructions.py b/environments/eval_environments/ifeval_instructions/instructions.py
index 86fa58f6..fe616e60 100644
--- a/environments/eval_environments/ifeval_instructions/instructions.py
+++ b/environments/eval_environments/ifeval_instructions/instructions.py
@@ -1155,7 +1155,7 @@ class RephraseParagraph(Instruction):
 
         Args:
           original_paragraph: A string presenting the original paragraph. The
-            rephrases response should have betweeb low-high words in common.
+            rephrases response should have between low-high words in common.
           low: An integer presenting the lower bound of similar words.
           high: An integer representing the upper bound of similar words.