One key, every model
Every image and video model.One API.
Monogram puts FLUX, Veo, Sora, Kling, Seedream and every other leading picture and motion model behind a single endpoint, with ready-made skills for real work and a studio your whole team will actually enjoy.
- Models
- 32
- Skills
- 20
- Endpoint
- 1
- FLUX.2 [pro]
- FLUX.2 [klein]
- FLUX.1 Kontext [max]
- Gemini 3 Pro Image
- Gemini 3.1 Flash Image
- Imagen 4 Ultra
- GPT Image 1.5
- GPT Image 1 Mini
- Seedream 4.5
- Ideogram 3.0
- Recraft V3
- Stable Image Ultra
- Qwen-Image
- Luma Photon
- Topaz Image Upscale
- Bria RMBG 2.0
- Veo 3.1
- Veo 3.1 Fast
- Sora 2 Pro
- Sora 2
- Kling 2.5 Turbo
- Seedance 1.5 Pro
- Seedance 1.0 Pro Fast
- Hailuo 02
- Runway Gen-4 Turbo
- Runway Aleph
- Luma Ray 2
- Wan 2.5
- LTX-2
- Pika 2.2
- Topaz Video Upscale
- Lipsync 2 Pro
01The API
Change one string. Change the model.
Every model answers the same request. Pick a different one on the right and watch the call: only the model name moves, so trying a new model takes a minute, not a sprint.
- 01
One schema
Prompts, reference images, durations and outputs use the same shape for every provider.
- 02
One key
No accounts to open with each lab. Rotate a single key and every integration follows.
- 03
One bill
A single balance and invoice, with cost broken down by model, project and key.
Pick a model
curl https://api.monogram.dev/v1/generations \
-H "Authorization: Bearer $MONOGRAM_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "bfl/flux-2-pro",
"input": {
"prompt": "A crystal perfume bottle on black marble, gold rim light, 1920s poster",
"aspect_ratio": "4:5"
}
}'{
"id": "gen_01JDK8W2QF",
"model": "bfl/flux-2-pro",
"status": "succeeded",
"output": [
{
"url": "https://cdn.monogram.dev/gen_01JDK8W2QF/0.png",
"width": 1024,
"height": 1280
}
]
}ImageVideo, delivered as a job with a webhook
02The roster
A full house, in stills and in motion.
From typography specialists to cinematic video with sound, every model is one string away. New releases join the roster as they launch, under the same API.
Image models
See all03Skills
The work, already wired.
A skill is a ready-made recipe that chains the right models for a job, from a studio product shot to a finished storyboard. Call it by name from your code, run it in the studio, or hand it to your AI agent.
Product Shot
Turn a phone photo of a product into a polished studio shot.
- Cut out the product
- Build the studio scene
- Match light and shadow
- Upscale for print
Consistent Character
Keep the same face, outfit and style across every image.
- Learn the character
- Generate each scene
- Check the likeness
Animate a Still
Bring a photo to life as a short, looping clip.
- Write the motion
- Animate the scene
- Blend the loop
Storyboard to Film
Turn a shot list into matching clips, cut together with sound.
- Design key frames
- Animate each shot
- Assemble the cut
Ad Creative Pack
Headline-ready ad visuals in every format, with legible text.
- Draft headline variants
- Render with type
- Extend to every format
Model Shootout
Run one prompt across several models and compare them side by side.
- Fan out the prompt
- Generate in parallel
- Score and rank
04The studio
A studio as good as the models inside it.
Prompt, compare and run skills in a calm, fast workspace. Everything you make is one click from an API call, so designers and developers always work from the same page.
A cut-crystal perfume bottle with a gold collar, soft window light, long leaf shadows
Interface preview with sample imagery drawn in ASCII. Hover or tap any frame to see the photo underneath.
Side by side
Run one prompt on several models and pick the winner at a glance.
Skill runner
Fill in a form, watch each step land, keep every intermediate file.
Shared library
Every generation searchable by prompt, model and project, for the whole team.
Copy as code
Any result becomes an API call in one click, so design and code never drift.
05Built for production
The plumbing, handled.
Routing, retries, budgets and records are part of the API, not a project for your team. Ship the feature and leave the operations to us.
Automatic fallbacks
If a provider slows down or fails, the request moves to the next model you approved and the response tells you which one answered.
Jobs and webhooks
Long video renders run as jobs with progress events, signed webhooks and results that wait for you.
Spend controls
Budgets per key, per project and per model, with alerts well before a limit is reached.
Usage you can read
Every generation is logged with model, latency and cost, and exports cleanly to your own tools.
Agent ready
An MCP server exposes every model and skill to AI assistants and agents, inside the budgets you set.
Your work stays yours
Outputs are kept for 30 days or not at all, your choice, and nothing you send is used for training.
06Pricing
Pay for pictures, not plumbing.
One balance covers every model. Start on pay as you go, add seats when the team joins, talk to us when volume grows.
Pay as you go
$0
per month
Top up a balance and pay each model's rate per generation. Ideal for building and testing.
- Every model and every skill
- API, SDKs and MCP server
- Studio for one person
- Community support
Studio
Most teams$24
per seat per month
For teams who create together, with shared libraries and controls for every project.
- Everything in Pay as you go
- Shared libraries and brand kits
- Budgets per project and key
- Priority queue for video
- Email support
Enterprise
Custom
annual agreement
Volume rates, private routing and a named contact for teams running at scale.
- Volume pricing on every model
- SSO, roles and audit logs
- Regional data processing
- Uptime commitment
- Dedicated support channel
Prices in US dollars, before tax. Generation costs follow each model rate and are drawn from your balance.
07Questions
Asked and answered.
Anything else on your mind? Mention it when you request access and a person will reply.
Which models can I use?
Every model in the roster, across image generation, video generation and utility work such as upscaling, background removal and lip sync. New models join as they launch, and older versions stay available while you migrate.
Do I need an account with each provider?
No. Monogram holds the provider relationships, so you get one key, one balance and one invoice for everything you call.
What exactly is a skill?
A skill is a named recipe that chains models and tools for one job, such as a studio product shot or a storyboard turned into film. You send the inputs, Monogram runs every step and returns the result, with each intermediate file available if you want it.
Can AI agents use Monogram?
Yes. Every model and skill is available through our MCP server, so assistants and agents can create and edit images and video for you, within the budgets you set on their key.
What happens if a model is down or slow?
You choose fallbacks per request or per project. If your first choice is unavailable, the request moves to the next model you approved, and the response says which model answered.
Who owns what I generate?
You do, under the terms of the model that made it. We never train on your inputs or outputs, and you decide how long results are stored.
One key.Every model.
Tell us what you are building and we will send your key, so your first picture or film is minutes away.