griptape

griptape is a modular Python framework for LLM workflows, tools, memory, and data that enables developers to:

🤖 Build AI agents, sequential LLM pipelines and sprawling DAG workflows for complex use cases.
⛓️ Augment LLMs with chain of thought capabilities.
🧰️ Integrate other services and functionality into LLMs as tools (e.g., calculators, web scrapers, spreadsheet editors, and API connectors); run tools in any environment (local, containerized, cloud, etc.); use tools directly in griptape or convert them into ramps abstractions, such as ChatGPT Plugins, LangChain tools, or Fixie.ai agents.
💾 Add memory to AI pipelines for context preservation and summarization.

Documentation

Please refer to Griptape Docs for:

Getting started guides.
Core concepts and design overviews.
Examples.
Contribution guidelines.

Quick Start

First, install griptape and griptape-tools:

pip install griptape griptape-tools -U

Second, configure an OpenAI client by getting an API key and adding it to your environment as OPENAI_API_KEY. griptape uses OpenAI Completions API to execute LLM prompts and to work with LlamaIndex data structures.

With griptape, you can create structures, such as Agents, Pipelines, and Workflows, that are composed of different types of tasks. You can also define structures as JSON objects and load them into griptape dynamically. Let's define a simple two-task pipeline that uses tools:

from decouple import config
from griptape.core import ToolLoader
from griptape.drivers import OpenAiPromptDriver, MemoryStorageDriver
from griptape.executors import LocalExecutor
from griptape.memory import Memory
from griptape.ramps import StorageRamp
from griptape.structures import Pipeline
from griptape.tasks import ToolkitTask, PromptTask
from griptape.tools import WebScraper

storage = StorageRamp(
    driver=MemoryStorageDriver()
)

scraper = WebScraper(
    ramps={
        "get_content": [storage]
    }
)

pipeline = Pipeline(
    memory=Memory(),
    tool_loader=ToolLoader(
        tools=[scraper],
        executor=LocalExecutor()
    )
)

pipeline.add_tasks(
    ToolkitTask(
        tool_names=[scraper.name]
    ),
    PromptTask(
        "Say the following in spanish: {{ input }}"
    )
)

result = pipeline.run("Give me a summary of https://en.wikipedia.org/wiki/Large_language_model")

print(result.output.value)

Boom! Our first LLM pipeline with two sequential tasks generated the following exchange:

Q: Give me a summary of https://en.wikipedia.org/wiki/Large_language_model
[chain of thought output... will vary depending on the model driver you're using]
A: Los modelos de lenguaje de gran tamaño son herramientas utilizadas para tareas de 
procesamiento del lenguaje natural, como detectar falsedades, completar oraciones y comprender 
el lenguaje. Algunos modelos notables incluyen BERT, GPT-2, GPT-3, GPT-Neo y GLaM. The Pile es 
un conjunto de datos extenso utilizado para el modelado del lenguaje. Estos modelos han sido 
desarrollados e investigados en trabajos como TruthfulQA, HellaSwag y BERT: Pre-entrenamiento 
de transformadores bidireccionales profundos para la comprensión del lenguaje.

Versioning

griptape is in early development and its APIs and documentation are subject to change. Until we stabilize the API and release version 1.0.0, we will use minor versions (i.e., x.Y.z) to introduce features and breaking features, and patch versions (i.e., x.y.Z) for bug fixes.

Contributing

Contributions in the form of bug reports, feature ideas, or pull requests are super welcome! Take a look at the current issues and if you'd like to help please submit a pull request with some tests.

License

griptape is available under the Apache 2.0 License.

Metadata

Release files for griptape 0.7.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for griptape 0.7.3
File	Size	Uploaded
griptape-0.7.3.tar.gz	37.9 kB	May 3, 2023	Details

Built distribution (wheel)

Table of built distributions (wheels) for griptape 0.7.3
File	Interpreter	ABI	Platform	Reset
griptape-0.7.3-py3-none-any.whl 68.1 kB May 3, 2023	Python 3	none	any	Details

Total release size: 106.0 kB

Release files / griptape-0.7.3.tar.gz

Download URL	griptape-0.7.3.tar.gz
Size	37.9 kB
Tags	Source
SHA-256 checksum How to use checksums	`713d0f5b2958aca5a3cf919334e9b4329c9a8d65490546dfceb42191b391daa2`
BLAKE2b-256 checksum How to use checksums	`747765f0218016115d4bf29f7e75803dbfc023ecd60aa221acb95fc8047cc346`
Upload date	May 3, 2023
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`poetry/1.4.1 CPython/3.9.16 Darwin/21.6.0`

Release files / griptape-0.7.3-py3-none-any.whl

Download URL	griptape-0.7.3-py3-none-any.whl
Size	68.1 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`8697bbba1c35331218c6e26dea4f35faf4cda3f54badb7a85e5595d2dd7bc4b4`
BLAKE2b-256 checksum How to use checksums	`6eae2b714b55c7d744550081094f256def31be6ef57af85565b6f338165a145a`
Upload date	May 3, 2023
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`poetry/1.4.1 CPython/3.9.16 Darwin/21.6.0`