Google PaLM API
I recently got access to Google Maker Suit. It enabled me to try Google’s LLM capabilities.
However, it is only open to the USA currently. I have been using US VPN to access. The API is also working on machines in the USA or may need a proxy.
Fortunately, after getting API keys. I can just use Google Colab to try.
Pros:
- UI
- Google’s UI is better than its competitor
- You can:
- Save to Drive
- Export to ColabEasy few-shot training using example prompts and answers
- Data Prompt option to use structured data
- Easily adjust safety settings
- API
- PaLM 2
- Firebase extensions: Easily implement in a current Firebase project
- Input token limit 8096
- 30 free API calls per minute (3 in ChatGPT)
- PaLM Colab Magic: Enables to work with tabular data easily. Google Sheets support
Cons:
- Output token limit 1024
- I think currently is worse than GPT 4 even 3.5 in many tasks
- They say the knowledge cutoff date is mid-2021. However, it knows something about current news but not much.
PaLM
PaLM Settings
Enjoy Reading This Article?
Here are some more articles you might like to read next:
- Notes from YC x Paris 2025
- Hosting Boris Cherny, Creator of Claude Code
- MenaML 2025 / Sponsored by Google DeepMind and QCRI
- Devin AI vs. Manus AI | My Take on Two Popular Autonomous Agents
- Last Invention: AGI | An Introductory Non-Technical Article
- Accepted to Notion AI Private Beta
- Tech Talks 4: Encoding Structured Data For LLMs, Neslihan Bulut
- Accepted to Devin AI Preview!
- Tech Talks 3: Yusuf Dalva, Bilkent CS Alumni, Virginia Tech Phd Student
- Introducing Bilkent GPT (Alpha)