Skip to main content

No project description provided

Project description

Vellum Python Library

pypi license badge fern shield

The Vellum Python SDK provides access to the Vellum API from python.

API Docs

You can find Vellum's complete API docs at docs.vellum.ai.

Installation

pip install --upgrade vellum-ai

Usage

Below is how you would invoke a deployed Prompt from the Vellum API. For a complete list of all APIs that Vellum supports, check out our API Reference.

from vellum import (
    StringInputRequest,
)
from vellum.client import Vellum

client = Vellum(
    api_key="YOUR_API_KEY",
)

def execute() -> str:
    result = client.execute_prompt(
        prompt_deployment_name="<example-deployment-name>>",
        release_tag="LATEST",
        inputs=[
            StringInputRequest(
                name="input_a",
                type="STRING",
                value="Hello, world!",
            )
        ],
    )
    
    if result.state == "REJECTED":
        raise Exception(result.error.message)

    return result.outputs[0].value

if __name__ == "__main__":
    print(execute())

[!TIP] You can set a system environment variable VELLUM_API_KEY to avoid writing your api key within your code. To do so, add export VELLUM_API_KEY=<your-api-token> to your ~/.zshrc or ~/.bashrc, open a new terminal, and then any code calling vellum.Vellum() will read this key.

Async Client

This SDK has an async version. Here's how to use it:

import asyncio

import vellum
from vellum.client import AsyncVellum

client = AsyncVellum(api_key="YOUR_API_KEY")

async def execute() -> str:
    result = await client.execute_prompt(
        prompt_deployment_name="<example-deployment-name>>",
        release_tag="LATEST",
        inputs=[
            vellum.StringInputRequest(
                name="input_a",
                value="Hello, world!",
            )
        ],
    )

    if result.state == "REJECTED":
        raise Exception(result.error.message)
    
    return result.outputs[0].value

if __name__ == "__main__":
    print(asyncio.run(execute()))

Contributing

While we value open-source contributions to this SDK, most of this library is generated programmatically.

Please feel free to make contributions to any of the directories or files below:

examples/*
src/vellum/lib/*
tests/*
README.md

Any additions made to files beyond those directories and files above would have to be moved over to our generation code (found in the separate vellum-client-generator repo), otherwise they would be overwritten upon the next generated release. Feel free to open a PR as a proof of concept, but know that we will not be able to merge it as-is. We suggest opening an issue first to discuss with us!

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

vellum_ai-0.8.11.tar.gz (139.9 kB view details)

Uploaded Source

Built Distribution

vellum_ai-0.8.11-py3-none-any.whl (373.7 kB view details)

Uploaded Python 3

File details

Details for the file vellum_ai-0.8.11.tar.gz.

File metadata

  • Download URL: vellum_ai-0.8.11.tar.gz
  • Upload date:
  • Size: 139.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.5.1 CPython/3.8.18 Linux/5.15.0-1071-azure

File hashes

Hashes for vellum_ai-0.8.11.tar.gz
Algorithm Hash digest
SHA256 b9378fb9b7a25a049fa42d201562df991a504aada6dca140de087925970e0bb9
MD5 9d5df56412fb109501508e3c0a2a6623
BLAKE2b-256 d027cb2a7c8ac0aaf951fee346819f874e1fe1de6f0656e6345dc8a0ac55c212

See more details on using hashes here.

Provenance

File details

Details for the file vellum_ai-0.8.11-py3-none-any.whl.

File metadata

  • Download URL: vellum_ai-0.8.11-py3-none-any.whl
  • Upload date:
  • Size: 373.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.5.1 CPython/3.8.18 Linux/5.15.0-1071-azure

File hashes

Hashes for vellum_ai-0.8.11-py3-none-any.whl
Algorithm Hash digest
SHA256 466d36cfb1945c599c2d5bfd626078d57dae36ec5299e0bdaad491b32773051a
MD5 b0adc2faa917fe48b57713ae3c5976b2
BLAKE2b-256 8bcd63759021bd159c1691d6ded2364bf8e8e045e008ceb66cad21141979f73c

See more details on using hashes here.

Provenance

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page