Environment:Spcl Graph of thoughts Python 3 8 Runtime

Knowledge Sources	Graph of Thoughts pyproject.toml
Domains	Infrastructure, LLM_Reasoning
Last Updated	2026-02-14 03:30 GMT

Overview

Python 3.8+ environment with core scientific and LLM dependencies for the Graph of Thoughts framework.

Description

This environment provides the base Python runtime and all required pip packages to run the Graph of Thoughts framework. It is OS-independent and does not require GPU hardware for the core framework or the OpenAI-based examples. The dependency stack includes OpenAI client libraries, scientific computing packages (NumPy, SciPy, Pandas), visualization (Matplotlib), symbolic math (SymPy), and deep learning libraries (PyTorch, Transformers, Accelerate, BitsAndBytes). The deep learning libraries are required even for OpenAI-only usage because they are declared as hard dependencies in pyproject.toml.

Usage

Use this environment for any usage of the Graph of Thoughts framework, including running examples, extending the framework with custom Prompters/Parsers, or using the OpenAI ChatGPT language model. This is the base prerequisite for all implementations in the repository.

System Requirements

Category	Requirement	Notes
OS	Any (OS Independent)	Declared in pyproject.toml classifiers
Python	>= 3.8	Declared in pyproject.toml `requires-python`
Disk	~2GB	For pip packages including PyTorch

Dependencies

System Packages

`python` >= 3.8
`pip` (latest recommended)
`git` (for editable install from source)

Python Packages

`backoff` >= 2.2.1, < 3.0.0
`openai` >= 1.0.0, < 2.0.0
`matplotlib` >= 3.7.1, < 4.0.0
`numpy` >= 1.24.3, < 2.0.0
`pandas` >= 2.0.3, < 3.0.0
`sympy` >= 1.12, < 2.0
`torch` >= 2.0.1, < 3.0.0
`transformers` >= 4.31.0, < 5.0.0
`accelerate` >= 0.21.0, < 1.0.0
`bitsandbytes` >= 0.41.0, < 1.0.0
`scipy` >= 1.10.1, < 2.0.0

Credentials

No credentials are required for the base runtime. See Environment:Spcl_Graph_of_thoughts_OpenAI_API_Access for API keys needed by the ChatGPT language model.

Quick Install

# Install from PyPI
pip install graph_of_thoughts

# Or install from source (editable mode)
git clone https://github.com/spcl/graph-of-thoughts.git
cd graph-of-thoughts
pip install -e .

Code Evidence

Python version requirement from `pyproject.toml:17`:

requires-python = ">=3.8"

Full dependency list from `pyproject.toml:22-34`:

dependencies = [
  "backoff>=2.2.1,<3.0.0",
  "openai>=1.0.0,<2.0.0",
  "matplotlib>=3.7.1,<4.0.0",
  "numpy>=1.24.3,<2.0.0",
  "pandas>=2.0.3,<3.0.0",
  "sympy>=1.12,<2.0",
  "torch>=2.0.1,<3.0.0",
  "transformers>=4.31.0,<5.0.0",
  "accelerate>=0.21.0,<1.0.0",
  "bitsandbytes>=0.41.0,<1.0.0",
  "scipy>=1.10.1,<2.0.0",
]

Build system from `pyproject.toml:1-2`:

[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"

Common Errors

Error Message	Cause	Solution
`ModuleNotFoundError: No module named 'graph_of_thoughts'`	Package not installed	Run `pip install graph_of_thoughts` or `pip install -e .` from repo root
`ImportError: cannot import name 'OpenAI' from 'openai'`	openai package version < 1.0.0	Upgrade with `pip install openai>=1.0.0`
Version conflicts during pip install	Conflicting package versions in existing environment	Create a fresh virtual environment: `python -m venv got_env && source got_env/bin/activate`

Compatibility Notes

OS Independent: The framework runs on Linux, macOS, and Windows. The pyproject.toml explicitly declares `"Operating System :: OS Independent"`.
Python Version: While Python 3.8 is the minimum, some dependencies (e.g., newer versions of NumPy, Pandas) may drop Python 3.8 support. Python 3.10+ is recommended.
PyTorch CPU: PyTorch is listed as a hard dependency but only GPU features are needed for the Llama2HF backend. CPU-only PyTorch suffices for OpenAI-based usage.

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment