OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Gflow CLI

Gflow CLI is a Python command-line tool and MCP server that operates Google Flow to generate Veo videos and Imagen images. It supports text-to-video, image-to-video and text-to-image workflows, including scripted and batched submissions. The tool drives Flow through a persistent Playwright Chromium browser profile, using the user's authenticated Flow session rather than a direct model API.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Gflow CLI — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you already have Google Flow access and want to automate image or video generation from a terminal, script, batch process or MCP-connected coding agent.

Who it is for

  • Developers who want to run Google Flow generation from terminal scripts or batch jobs.
  • Teams building content workflows that incorporate Veo video or Imagen image generation.
  • Users of compatible coding agents who want to expose Flow generation actions through the MCP server.

Documented uses

  • Submit text prompts for Veo text-to-video generation.
  • Turn an image into video with Veo.
  • Generate images from text prompts with Imagen.
  • Automate repeated prompts in scripts and batch workflows.
  • Make Google Flow generation actions available to compatible coding agents through MCP.
Before you choose

Requirements & limits

The quick start requires a Google account with Flow access and authenticates through a browser. Veo video generations use the account's Flow credit allowance; image generation is described as free. The project is alpha and unaffiliated with Google.

  • It drives Flow through a persistent Playwright Chromium browser profile, so it depends on browser-based access rather than a direct model API.
  • The project labels itself alpha and reverse-engineered; Flow interface or endpoint changes may affect operation.
  • The README states automation is subject to Google's terms and does not promise an account outcome.
  • The repository specifies Python and an MIT license.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and authenticate Gflow CLI

Set up the CLI and authenticate it with a Google account that has Google Flow access.

Before you start

  • A Google account with Flow access.
  • uv or pip for installing the Python CLI.
  • Playwright Chromium; the quick start installs it through uv.
  • Chrome for the documented real-Chrome authentication flow.

Steps

  1. Install the package with the command below or the command below.

    Documented command
    uv tool install gflow-cli
    Documented command
    pip install gflow-cli

  2. Install Chromium with the command below.

    Documented command
    uv tool run --from gflow-cli playwright install chromium

  3. Run gflow auth login --browser chrome and complete authentication in the opened browser.

Expected outcome

The CLI has a browser-based authenticated session for subsequent Google Flow operations.

Guide limitations

  • The project is alpha, reverse-engineered and unaffiliated with Google.
  • Automation is subject to Google's terms; Flow interface or endpoint changes may disrupt operation.
  • A Flow account with access is required, and Veo video generations use that account's credit allowance.
Documentation 1 ↗
Source-based practical guide

Generate an image or video from the terminal

Submit one text-to-image or text-to-video generation through an authenticated Gflow CLI session.

Before you start

  • Gflow CLI installed and authenticated as described in its quick start.
  • A Google account with Flow access; video generation uses Flow credits.

Steps

  1. For an image, run gflow image t2i "a hot air balloon over Tokyo at sunrise".

  2. For a video, run gflow video t2v "Slow cinematic push-in on a sunlit forest clearing" --aspect 16:9.

  3. Find generated outputs under $GFLOW_CLI_OUTPUT_DIR.

Expected outcome

The requested generation is submitted to Google Flow, with outputs directed to the configured output directory.

Guide limitations

  • Veo video generation bills against the account's Flow credit allowance; image generation is described as free.
  • The tool drives Flow through a headed browser, and changes to Flow's interface or endpoints may affect operation.
  • The README notes the first call takes 30 to 90 seconds while Chromium warms up.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
ffroliva/gflow-cli · Official homepage title · retained in its source language
How we review sourcesSuggest a correction