Multi-modal
Available now

Gemini 4 Argon

by Google DeepMind

Google's reasoning model, now in Morphic Copilot.
It plans shots, then calls the models.

Gemini 4 Argon

Key features

Technical specifications

Reasoning

A thinking model that plans and directs; it makes no media itself

5 types

Text, images, video, audio, and PDF

Text

Plans, scripts, prompts, and tool calls

1M tokens

Input and output share one 1,048,576-token budget

Sep 2026

Announced by Google on September 30, 2026

Copilot

Pick it as the model behind any new Copilot chat

Basic and up

In the Copilot model picker on every paid plan

MCP

Reach Morphic from an MCP app such as Gemini CLI

Use cases

A woman in a dark coat faces a man holding an envelope on a rain-slicked city rooftop at night

AI microdramas

Turn a premise into a 10-episode outline with hooks and cliffhangers, then generate the opening shots with a fixed cast.

A 3D-rendered armored game character shown from the front, side, and three-quarter view on a grey background

3D-style characters

Define materials, silhouette and lighting, then get a front, side and three-quarter reference sheet for one character.

An astronaut in a white suit steps through a circular hatch into a dark, abandoned space station corridor

Cinematic short films

Hand it a script and get the shot list, camera moves and first shots, with the lead looking the same from shot to shot.

An unlabeled glass perfume bottle on black stone, lit by a single warm beam through mist

Product ads

Write three ad angles for one product, from mysterious to energetic, and generate the visuals for each from a single brief.

Three anime-style teenagers on a rooftop look out over a futuristic city at a pink sunset

Anime series

Outline the story arc and cast, build character references, and plan the opening sequence of an anime-style series.

A dancer in a silver jacket spins on a wet neon-lit street at night, with light streaks behind her

Music videos

Map scene changes to the track, generate the clips and music, and get a cut order for the edit.

Prompt examples

Microdrama series

Plan a 10-episode thriller with cliffhangers, then shoot the Episode 1 opener

Edit prompt

Character sheet

Design a 3D-style armored game hero: front, side and three-quarter views

Edit prompt

Sci-fi short

Plan a two-minute space station sci-fi short and make the first three shots

Edit prompt

Ad variants

Write three 15-second perfume ads, mysterious, minimal, energetic, with visuals

Edit prompt

Anime opening

Create three anime-style leads in a future city and plan a 20-second opening

Edit prompt

Music video

Plan a 30-second neon music video, generate the clips, and suggest the cut

Edit prompt

Gemini 4 Argon overview

Gemini 4 Argon is Google's reasoning model for work that takes many steps. It writes no pixels. What it adds to a creative project is planning: it turns a premise into a script, the script into scenes, and the scenes into prompts, references and a cut order. On Morphic that plan runs straight away, because Copilot gives Argon the image, video and audio models to call.

How to use Gemini 4 Argon in Morphic Copilot

  1. Open a project and start a new Copilot chat.
  2. Choose Gemini 4 Argon in the model picker at the bottom right of the prompt bar.
  3. Attach what you already have: a script, a brief, reference images or a clip.
  4. Describe the whole project, then let it plan. Switch the chat to Plan only if you want the plan before anything renders.
  5. Approve the plan and generate. Results land in the project, ready for review and the edit.

Gemini 4 Argon over Morphic MCP

Working in Gemini CLI or another MCP app that runs Argon? Add the Morphic MCP server (https://mcp.morphic.com/mcp), approve the connection, and it can generate into your projects from there. The Gemini 4 Argon guide walks through the setup and six project workflows step by step.

Gemini 4 Argon limits to plan around

Argon plans and reviews; the media models render. Consistency still comes from references, so give each recurring character a character sheet and check episodes before they ship. A still styled as a 3D render is not an exportable 3D model. And treat its video review as a second pair of eyes, with the final approval yours.

Simple pricing

Get started for free today, with the option to upgrade or cancel anytime.

Basic

$19/ month
billed as $0 per year

2400 monthly credits

1 user only

All models

Workflows

Standard

$24/ month
billed as $0 per year

3625 monthly credits

1 user only

All models

Workflows

Pro

$45/ month
billed as $0 per year

6350 shared monthly credits

1 user

+ up to 4 more at extra cost

All models

Workflows

Max

$170/ month
billed as $0 per year

24650 shared monthly credits

1 user

+ up to 9 more at extra cost

All models

Workflows

Enterprise

For higher limits

Custom

pricing and billing terms

High-volume credits
Custom seat limits
All models
Workflows
Pricing Gradient

Free

For playing around

$0

forever free

Up to 20 credits
1 user only
Limited models
Workflows
Compare plan details

FAQs

What is Gemini 4 Argon?
Gemini 4 Argon is Google's top-tier reasoning model of the Gemini 4 generation, announced on September 30, 2026. It reads text, images, video, audio and PDFs, writes text, and is built for multi-step work: planning, coding, analysis and tool use. It has a 1M-token context window shared between input and output.
Can I use Gemini 4 Argon on Morphic?
Yes. Open a new Copilot chat, choose Gemini 4 Argon in the model picker at the bottom right of the prompt bar, and describe the project. It is available on every paid plan, from Basic up. You can also reach Morphic from an MCP app such as Gemini CLI.
Can Gemini 4 Argon generate images or videos?
Not by itself: its output is text. On Morphic that does not matter, because Copilot hands each generation to Morphic's image, video and audio models. Argon decides what to make and writes the prompts; the media models render it, and the results land in your project.
How do I connect Gemini 4 Argon to Morphic through MCP?
Use an MCP app that runs Gemini 4 Argon, such as Gemini CLI. Add the Morphic MCP server (https://mcp.morphic.com/mcp), sign in to Morphic and approve the connection, then ask which Morphic projects and tools it can see. Setup steps for each app are on the MCP settings page in Morphic.
Can I make AI microdramas with Gemini 4 Argon?
Yes. Argon writes the episode outline, hooks, cliffhangers and shot list, and Copilot generates the shots with a character sheet for each recurring role, which keeps the cast consistent. Review each episode before you publish, since no plan guarantees every generated shot matches.
Can Gemini 4 Argon create 3D models?
It plans 3D-style characters and scenes, and Copilot generates images that look like 3D renders, such as turnaround reference sheets. Those are images, not rigged or exportable meshes. For production geometry, plan the character with Argon and build the model in a dedicated 3D tool.
Does the Gemini app connect to Morphic MCP?
Check your app first. The supported route is an MCP app such as Gemini CLI, which has its own setup steps on the MCP settings page. A consumer chat app may not let you add your own MCP servers. Inside Morphic, Copilot needs no setup at all.
How is Gemini 4 Argon different from Gemini Omni?
Argon is a reasoning model that writes text and directs work. Gemini Omni is a video model that renders clips with sound. On Morphic they work together: Argon can plan a scene in Copilot and send the shot to Gemini Omni, Veo, Kling or another video model.

More about Gemini 4 Argon

Guides, prompts, and comparisons for Gemini 4 Argon.

How to use Gemini 4 Argon with Morphic: Copilot and MCP

Use Gemini 4 Argon with Morphic through Copilot or MCP to plan and produce microdramas, 3D-style characters, short films, ads, anime and music videos.

Other models

Explore the rest of the Morphic model catalog.

See all models →

FLUX Video Upscale

Black Forest Labs

Black Forest Labs' video upscaler. Up to 4K, with fine detail rebuilt and the sound kept.

FLUX 3 Max

fal.ai

fal's post-trained FLUX 3. Tuned to follow the script, built for speed.

Krea 3

Krea AI

Krea's fast image editing model. Combine up to 4 images in one edit.

OmniHuman 2.0

ByteDance

ByteDance's audio-driven human video model. One photo, one track, five minutes in one take.