diff --git a/.github/workflows/causal_lm_cpp.yml b/.github/workflows/causal_lm_cpp.yml
index 80089a4e8..5a96741b5 100644
--- a/.github/workflows/causal_lm_cpp.yml
+++ b/.github/workflows/causal_lm_cpp.yml
@@ -191,6 +191,8 @@ jobs:
 
   cpp-greedy_causal_lm-windows:
     runs-on: windows-latest
+    env:
+      PYTHONIOENCODING: "utf8"
     defaults:
       run:
         shell: cmd
@@ -626,6 +628,8 @@ jobs:
 
   cpp-continuous-batching-windows:
     runs-on: windows-latest
+    env:
+      PYTHONIOENCODING: "utf8"
     defaults:
       run:
         shell: cmd
diff --git a/.github/workflows/genai_package.yml b/.github/workflows/genai_package.yml
index 06e589dfb..9e439eb11 100644
--- a/.github/workflows/genai_package.yml
+++ b/.github/workflows/genai_package.yml
@@ -80,6 +80,7 @@ jobs:
     runs-on: windows-latest
     env:
       CMAKE_BUILD_PARALLEL_LEVEL: null
+      PYTHONIOENCODING: "utf8"
     defaults:
       run:
         shell: cmd
diff --git a/.github/workflows/genai_python_lib.yml b/.github/workflows/genai_python_lib.yml
index 29ceda216..38552174b 100644
--- a/.github/workflows/genai_python_lib.yml
+++ b/.github/workflows/genai_python_lib.yml
@@ -62,6 +62,7 @@ jobs:
     runs-on: windows-latest
     env:
       CMAKE_BUILD_PARALLEL_LEVEL: null
+      PYTHONIOENCODING: "utf8"
     defaults:
       run:
         shell: cmd
diff --git a/samples/cpp/beam_search_causal_lm/README.md b/samples/cpp/beam_search_causal_lm/README.md
index a10428891..201a9b5ce 100644
--- a/samples/cpp/beam_search_causal_lm/README.md
+++ b/samples/cpp/beam_search_causal_lm/README.md
@@ -18,6 +18,7 @@ optimum-cli export openvino --trust-remote-code --model TinyLlama/TinyLlama-1.1B
 `beam_search_causal_lm TinyLlama-1.1B-Chat-v1.0 "Why is the Sun yellow?"`
 
 To enable Unicode characters for Windows cmd open `Region` settings from `Control panel`. `Administrative`->`Change system locale`->`Beta: Use Unicode UTF-8 for worldwide language support`->`OK`. Reboot.
+Also, you can enable UTF-8 mode by setting environment variable `PYTHONIOENCODING="utf8"`.
 
 Discrete GPUs (dGPUs) usually provide better performance compared to CPUs. It is recommended to run larger models on a dGPU with 32GB+ RAM. For example, the model meta-llama/Llama-2-13b-chat-hf can benefit from being run on a dGPU. Modify the source code to change the device for inference to the GPU.