diff --git a/instrumentation/opentelemetry-instrumentation-google-genai/.changelog/930.fixed b/instrumentation/opentelemetry-instrumentation-google-genai/.changelog/930.fixed
new file mode 100644
index 000000000..df6e9e925
--- /dev/null
+++ b/instrumentation/opentelemetry-instrumentation-google-genai/.changelog/930.fixed
@@ -0,0 +1 @@
+Report text parts with `thought=True` as `ReasoningPart` instead of `TextPart` in `generate_content` and `generate_content_stream` messages.
diff --git a/instrumentation/opentelemetry-instrumentation-google-genai/src/opentelemetry/instrumentation/google_genai/message.py b/instrumentation/opentelemetry-instrumentation-google-genai/src/opentelemetry/instrumentation/google_genai/message.py
index febfef29b..0a4476632 100644
--- a/instrumentation/opentelemetry-instrumentation-google-genai/src/opentelemetry/instrumentation/google_genai/message.py
+++ b/instrumentation/opentelemetry-instrumentation-google-genai/src/opentelemetry/instrumentation/google_genai/message.py
@@ -14,6 +14,7 @@
     InputMessage,
     MessagePart,
     OutputMessage,
+    ReasoningPart,
     Role,
     ServerToolCallPart,
     ServerToolCallResponsePart,
@@ -91,6 +92,8 @@ def tool_call_id(name: str | None) -> str:
         return f"{idx}"
 
     if (text := part.text) is not None:
+        if part.thought:
+            return ReasoningPart(content=text)
         return TextPart(content=text)
 
     if inline_data := part.inline_data:
diff --git a/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/nonstreaming_base.py b/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/nonstreaming_base.py
index 0f5dfd69b..0e775404d 100644
--- a/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/nonstreaming_base.py
+++ b/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/nonstreaming_base.py
@@ -13,6 +13,7 @@
     GoogleMaps,
     MediaModality,
     ModalityTokenCount,
+    Part,
     ToolDict,
 )
 from pydantic import BaseModel, Field
@@ -751,6 +752,24 @@ def test_span_attributes_with_content_capture(self):
                 '[{"name":"_mock_callable_tool","description":"Description of some tool.","parameters":null,"type":"function"},{"name":"mock_tool","description":"Description of mock tool.","parameters":null,"type":"function"},{"name":"google_maps","type":"google_maps"}]',
             )
 
+    @patch.dict(
+        "os.environ",
+        {"OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT": "SPAN_ONLY"},
+    )
+    def test_reports_thought_parts_as_reasoning(self):
+        self.configure_valid_response(
+            parts=[
+                Part(text="Let me think.", thought=True),
+                Part(text="It is sunny."),
+            ]
+        )
+        self.generate_content(model="gemini-2.0-flash", contents="Hi")
+        span = self.otel.get_span_named("generate_content gemini-2.0-flash")
+        self.assertEqual(
+            span.attributes[gen_ai_attributes.GEN_AI_OUTPUT_MESSAGES],
+            '[{"role":"assistant","parts":[{"content":"Let me think.","type":"reasoning"},{"content":"It is sunny.","type":"text"}],"finish_reason":"","name":null}]',
+        )
+
     def test_log_has_extra_genai_attributes(self):
         self.configure_valid_response(text="Yep, it works!")
         tok = context_api.attach(
diff --git a/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/streaming_base.py b/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/streaming_base.py
index e83aefed1..48f55f0ea 100644
--- a/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/streaming_base.py
+++ b/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/streaming_base.py
@@ -9,6 +9,7 @@
     GenerateContentResponseUsageMetadata,
     MediaModality,
     ModalityTokenCount,
+    Part,
 )
 
 from opentelemetry import context as context_api
@@ -163,6 +164,21 @@ def test_merges_streamed_chunks_into_one_output_message(self):
             '[{"role":"assistant","parts":[{"content":"Hello, world","type":"text"}],"finish_reason":"","name":null}]',
         )
 
+    @patch.dict(
+        "os.environ",
+        {"OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT": "SPAN_ONLY"},
+    )
+    def test_reports_streamed_thought_parts_as_reasoning(self):
+        self.configure_valid_response(part=Part(text="Let me ", thought=True))
+        self.configure_valid_response(part=Part(text="think.", thought=True))
+        self.configure_valid_response(text="It is sunny.")
+        self.generate_content(model="gemini-2.0-flash", contents="Hi")
+        span = self.otel.get_span_named("generate_content gemini-2.0-flash")
+        self.assertEqual(
+            span.attributes[gen_ai_attributes.GEN_AI_OUTPUT_MESSAGES],
+            '[{"role":"assistant","parts":[{"content":"Let me think.","type":"reasoning"},{"content":"It is sunny.","type":"text"}],"finish_reason":"","name":null}]',
+        )
+
     def test_includes_token_counts_in_span_not_aggregated_from_responses(self):
         # Tokens should not be aggregated in streaming. Cumulative counts are returned on each response.
         self.configure_valid_response(
diff --git a/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/test_message.py b/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/test_message.py
index 8aed71709..3afd007d7 100644
--- a/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/test_message.py
+++ b/instrumentation/opentelemetry-instrumentation-google-genai/tests/generate_content/test_message.py
@@ -11,6 +11,7 @@
 )
 from opentelemetry.util.genai.types import (
     GenericPart,
+    ReasoningPart,
     ServerToolCallPart,
     ServerToolCallResponsePart,
     TextPart,
@@ -48,11 +49,28 @@ def test_to_system_instructions_non_text_parts(self):
             ],
         )
 
+    def test_to_system_instructions_thought_part(self):
+        content = genai_types.Content(
+            parts=[genai_types.Part(text="Let me think.", thought=True)]
+        )
+        instructions = to_system_instructions(content=content)
+        self.assertEqual(instructions, [GenericPart(type="reasoning")])
+
     def test_to_system_instructions_empty_or_unknown_part(self):
         content = genai_types.Content(parts=[genai_types.Part()])
         instructions = to_system_instructions(content=content)
         self.assertEqual(instructions, [])
 
+    def test_to_part_maps_thought_text_to_reasoning(self):
+        self.assertEqual(
+            _to_part(genai_types.Part(text="Let me think.", thought=True), 0),
+            ReasoningPart(content="Let me think."),
+        )
+        self.assertEqual(
+            _to_part(genai_types.Part(text="Sunny.", thought=False), 0),
+            TextPart(content="Sunny."),
+        )
+
     def test_to_part_maps_code_execution(self):
         call_id = (
             {"id": "code-1"}