help in installation

[mdrafic456]

Hi guys,

I am trying to install this tools, however the "make config" command is not working getting an error " make: *** No rule to make target 'config'. Stop."

tried with below command provided by the ChatGPT
"uv run pentestgpt-legacy"

Output:
warning: The tool.uv.dev-dependencies field (used in pyproject.toml) is deprecated and will be removed in a future release; use dependency-groups.dev instead
Welcome to PentestGPT (modernized legacy), an interactive penetration testing assistant.
The settings are:

• reasoning model: Google Gemini · gemini-3.1-pro
• generation model: Google Gemini · gemini-3.1-pro
• parsing model: Google Gemini · gemini-3.5-flash
• log directory: logs
  ⠧ Initializing sessions...2026-07-23 06:51:21.213 | ERROR | pentestgpt_legacy.utils.pentest_gpt:initialize:162 - 404 NOT_FOUND. {'error': {'code': 404, 'message': 'models/gemini-3.1-pro is not found for API version v1beta, or is not supported for generateContent. Call ModelService.ListModels to see the list of available models and their supported methods.', 'status': 'NOT_FOUND'}}
  PentestGPT execution failed: 404 NOT_FOUND. {'error': {'code': 404, 'message': 'models/gemini-3.1-pro is not found for API version v1beta, or is not supported for generateContent. Call ModelService.ListModels to see the list of available models and their supported methods.', 'status': 'NOT_FOUND'}}

Can any one help me fix this issue? please

[tgbardelli]

I'm having the same issue after cloning the repo, running on MacOS 26.5.2.

user@host pentestgpt % make config
make: *** No rule to make target `config'. Stop.

[therealmrzoidberg-tech]

Ivejust started doing so testing with kali linux,haven used linux for ages so im no expert so im just telling...
Try this,run from PentestGPT folder
Run this first
export GEMINI_API_KEY="your key"

regarding the url part im not shure because i have tested so much and done so many changes so im not sure about what the correct way.

export GEMENI_BASE_URL="http://l127.0.0:1:1434/v1"

uv run pentestgpt-legacy --provider Gemeni:gemini-3.1-pro \
--reasoning-model "Gemeni:gemini-3.1-pro"
--generation-model "emeni:gemini-3.1-pro"
--parsing-model "gemini-3.5-flash"