video-understanding

# Video Understanding with Gemini

Analyze video content using Google Gemini 3 Flash API. Extract summaries, transcripts, timestamps, and answer questions about video content from local files or YouTube URLs.

## Prerequisites

```bash
# Install SDK
uv pip install google-genai

# Set API key
export GEMINI_API_KEY="your-api-key"
```

## Input Methods

### 1. Files API Upload (Recommended for local files)

For files >100MB or videos >1 minute, use Files API for reliable upload.

```python
from google import genai

client = genai.Client(api_key=os.getenv("GEMINI_API_KEY"))

# Upload video file
video_file = client.files.upload(file="/path/to/video.mp4")

# Wait for processing
import time
while video_file.state.name == "PROCESSING":
    time.sleep(5)
    video_file = client.files.get(name=video_file.name)

if video_file.state.name == "FAILED":
    raise ValueError("Video processing failed")

# Analyze
response = client.models.generate_content(
    model="gemini-3-flash-preview",
    contents=[video_file, "Summarize this video"]
)
print(response.text)
```

### 2. Inline Data (For small files <20MB)

```python
import base64

with open("/path/to/short_video.mp4", "rb") as f:
    video_data = base64.standard_b64encode(f.read()).decode("utf-8")

response = client.models.generate_content(
    model="gemini-3-flash-preview",
    contents=[
        {"inline_data": {"mime_type": "video/mp4", "data": video_data}},
        "What is happening in this video?"
    ]
)
```

### 3. YouTube URL (Public videos only)

```python
from google.genai import types

response = client.models.generate_content(
    model="gemini-3-flash-preview",
    contents=types.Content(
        parts=[
            types.Part(
                file_data=types.FileData(
                    file_uri="https://www.youtube.com/watch?v=VIDEO_ID"
                )
            ),
            types.Part(text="Summarize this video with key timestamps")
        ]
    )
)
```

**YouTube Limits:**
- Public videos only (no priv