Meet the PaLM API

Access the advanced capabilities of Google’s large language models like PaLM 2 with the PaLM API. Use it to build generative AI applications for use cases like content generation, dialog agents, summarization, classification, and more.

Get started:

  1. Join the waitlist to get your API key via MakerSuite and start prototyping quickly

    Try the PaLM API

    $ curl \
        -H 'Content-Type: application/json' \
        -d '{"prompt": {"text": "Write a story about a magic backpack."}}' \
  2. Explore the prompt gallery for inspiration and ideas on what the PaLM API can do

  3. Already a Google Cloud customer? You can also access the PaLM API in Vertex AI

How the PaLM API performs

PaLM 2, the model behind the PaLM API, excels at complex tasks

Thanks to the PaLM 2 model architecture, the PaLM API offers fast and efficient performance. With speeds up to 75+ tokens per second and a context window of 8,000 tokens, it can handle complex prompts and instructions quickly.

PaLM API partners

Baseplate Chroma GradientJ Hubble LangChain Magick Stack AI Vellum Vercel Weaviate

Integrated into Google tools you already use

Colab Firebase