fix(google): add usage data (#573)

## Description
<!-- What does this PR do? -->
Found that gemini/vertexai didn't have usage data being added to their
completionchunks.

## PR Type

<!-- Delete the types that don't apply --!>

🐛 Bug Fix

## Relevant issues

<!-- e.g. "Fixes #123" -->

## Checklist
- [x] I have added unit tests that prove my fix/feature works
- [x] New and existing tests pass locally
- [x] Documentation was updated where necessary
- [x] I have read and followed the [contribution
guidelines](https://github.com/mozilla-ai/any-llm/blob/main/CONTRIBUTING.md)```

Author: njbrake

src/any_llm/providers/gemini/utils.py

@@ -9,6 +9,7 @@
     ChatCompletionChunk,
     ChoiceDelta,
     ChunkChoice,
+    CompletionUsage,
     CreateEmbeddingResponse,
     Embedding,
     Reasoning,
@@ -260,12 +261,21 @@ def _create_openai_chunk_from_google_chunk(
         finish_reason="stop" if getattr(candidate.finish_reason, "value", None) == "STOP" else None,
     )
 
+    usage = None
+    if response.usage_metadata:
+        usage = CompletionUsage(
+            prompt_tokens=response.usage_metadata.prompt_token_count or 0,
+            completion_tokens=response.usage_metadata.candidates_token_count or 0,
+            total_tokens=response.usage_metadata.total_token_count or 0,
+        )
+
     return ChatCompletionChunk(
         id=f"chatcmpl-{time()}",
         choices=[choice],
         created=int(time()),
         model=str(response.model_version),
         object="chat.completion.chunk",
+        usage=usage,
     )
 
 

tests/unit/providers/test_gemini_provider.py

@@ -396,3 +396,54 @@ async def test_gemini_with_built_in_tools() -> None:
         assert generation_config.tools is not None
         assert len(generation_config.tools) == 1
         assert generation_config.tools[0] == google_search
+
+
+@pytest.mark.asyncio
+async def test_streaming_completion_includes_usage_data() -> None:
+    """Test that streaming chunks include usage metadata when available."""
+    from any_llm.providers.gemini.utils import _create_openai_chunk_from_google_chunk
+
+    mock_response = Mock()
+    mock_response.candidates = [Mock()]
+    mock_response.candidates[0].content = Mock()
+    mock_response.candidates[0].content.parts = [Mock()]
+    mock_response.candidates[0].content.parts[0].text = "Hello"
+    mock_response.candidates[0].content.parts[0].thought = None
+    mock_response.candidates[0].finish_reason = Mock()
+    mock_response.candidates[0].finish_reason.value = "STOP"
+    mock_response.model_version = "gemini-2.5-flash"
+
+    mock_response.usage_metadata = Mock()
+    mock_response.usage_metadata.prompt_token_count = 10
+    mock_response.usage_metadata.candidates_token_count = 5
+    mock_response.usage_metadata.total_token_count = 15
+
+    chunk = _create_openai_chunk_from_google_chunk(mock_response)
+
+    assert chunk.usage is not None, "Usage data should be included in streaming chunks"
+    assert chunk.usage.prompt_tokens == 10
+    assert chunk.usage.completion_tokens == 5
+    assert chunk.usage.total_tokens == 15
+
+
+@pytest.mark.asyncio
+async def test_streaming_completion_without_usage_metadata() -> None:
+    """Test that streaming chunks handle missing usage metadata gracefully."""
+    from any_llm.providers.gemini.utils import _create_openai_chunk_from_google_chunk
+
+    mock_response = Mock()
+    mock_response.candidates = [Mock()]
+    mock_response.candidates[0].content = Mock()
+    mock_response.candidates[0].content.parts = [Mock()]
+    mock_response.candidates[0].content.parts[0].text = "Hello"
+    mock_response.candidates[0].content.parts[0].thought = None
+    mock_response.candidates[0].finish_reason = Mock()
+    mock_response.candidates[0].finish_reason.value = None
+    mock_response.model_version = "gemini-2.5-flash"
+
+    mock_response.usage_metadata = None
+
+    chunk = _create_openai_chunk_from_google_chunk(mock_response)
+
+    assert chunk.usage is None, "Usage should be None when metadata is not available"
+    assert chunk.choices[0].delta.content == "Hello"