diff --git a/litellm/llms/codestral/completion/transformation.py b/litellm/llms/codestral/completion/transformation.py
index e3c3fd1231c..a62be8e0a67 100644
--- a/litellm/llms/codestral/completion/transformation.py
+++ b/litellm/llms/codestral/completion/transformation.py
@@ -45,6 +45,7 @@ class CodestralTextCompletionConfig(OpenAITextCompletionConfig):
             "top_p",
             "max_tokens",
             "max_completion_tokens",
+            "min_tokens",
             "stream",
             "seed",
             "stop",
diff --git a/tests/test_litellm/llms/codestral/completion/test_transformation.py b/tests/test_litellm/llms/codestral/completion/test_transformation.py
new file mode 100644
index 00000000000..fc7af4ecffc
--- /dev/null
+++ b/tests/test_litellm/llms/codestral/completion/test_transformation.py
@@ -0,0 +1,31 @@
+import litellm
+from litellm.llms.codestral.completion.transformation import (
+    CodestralTextCompletionConfig,
+)
+
+
+def test_codestral_supported_openai_params_includes_min_tokens():
+    config = CodestralTextCompletionConfig()
+    supported_params = config.get_supported_openai_params(model="codestral-latest")
+    assert "min_tokens" in supported_params
+    assert "max_tokens" in supported_params
+    assert "suffix" in supported_params
+
+
+def test_codestral_map_openai_params_min_tokens():
+    config = CodestralTextCompletionConfig()
+    mapped = config.map_openai_params(
+        non_default_params={"min_tokens": 10},
+        optional_params={},
+        model="codestral-latest",
+        drop_params=False,
+    )
+    assert mapped == {"min_tokens": 10}
+
+
+def test_get_supported_openai_params_for_text_completion_codestral():
+    supported = litellm.get_supported_openai_params(
+        model="codestral/codestral-latest",
+        custom_llm_provider="text-completion-codestral",
+    )
+    assert "min_tokens" in supported