Accelerating the search for antimicrobial peptides with Google Antigravity
deepmind.google·2h ago
Cloud & Infrastructure
Google Cloud Blog·1d ago·5 min read
Predictable AI spend in BigQuery: Introducing cost management for generative AI functions
TL;DR
Google Cloud has introduced token-based cost management for its generative AI functions in BigQuery, allowing users to better control expenses when utilizing large language models in SQL queries. This feature is crucial for managing costs effectively when processing large datasets.
✦ Why It Matters
Start using token-based cost management in BigQuery to control your generative AI expenses today.
Key Takeaways
- Implement daily quotas to manage AI token consumption.
- Use AI.COUNT_TOKENS to estimate input token usage.
- Adjust quotas dynamically based on project needs.
- Monitor historical token consumption to detect spending spikes.
How It Works
The token-based cost management system allows administrators to set daily quotas for AI token usage, which are enforced at both project and user levels. This prevents unexpected costs by stopping queries that exceed the set limits, ensuring that teams can innovate without overspending.