Ry Magno

Head of Product Design @contra 🇵🇭

[Ry Magno @rymagno]

submissions are looking unbe-weave-able! @figma x @contra @figmaweave

[Contra @contra]

Creatives are putting @figmaweave to work for the $100K @figma Makeathon.

[Figma @figma]

Design matters now more than ever.

The Config Makeathon starts now.

[Rodrigo Davies @rodrigodavies]

Are you in SF for #Config next week? Join us for a fun discussion of creative intelligence in the age of AI - with @adele__li @joanrod_ai @viccpoes @contraben

[ben @contraben]

who's at Config next week ?

Contra Labs is hosting our annual "Config Kick Off Party" on Monday the 22nd for

>creatives
>researchers
>designers

to talk about the latest AI powered workflows, the role of human taste in creative work + all things creative intelligence.

+ a live panel with product leaders from Open AI, Qiver, Krea, and of course @figma

See you there! (link in comments)

[ben @contraben]

Big congrats to the @ideogram_ai team on v4.

We tested it blind against @GeminiApp 3.1, Grok @imagine, and @bfl_ai FLUX.2 [max] with

>10 professional designers
> across 240 images

Quick thread on it's strengths and how to prompt it 👇

[DANN© @DannPetty]

$1M earned on Contra.

Yes, it’s possible.

[Figma @figma]

The first ever Config Makeathon is coming. June 4. $100K in prizes. Powered by @contra

Bonus: pre-register by June 3 for access to Figma’s design agent beta.

[Ry Magno @rymagno]

@figma @contra letsgoooo 🔥

[Contra Labs @contralabs_ai]

https://t.co/kpILKMjCLr

[Matt @freelancer_matt]

Create a payment link in 30 seconds on @contra

- Available everywhere
- One-time or subscriptions

[ben @contraben]

For @GeminiApp, your prompt is the difference between Client V1 and Production Ready.

We observed 10 designers going through a real world campaign workflow:

>Hero stills
>Social cuts
>Secondary assets

total: 29 deliverables.

Only 24% were "Production Ready"

And they all wrote prompts the same way.

Below are the prompts and outputs 👇⬇️

[Contra @contra]

The Stitch Challenge is LIVE 🚀

Use @stitchbygoogle to build an interface that feels alive. Show us your process, your iteration, and how Stitch fits into your workflow.

$10K in prizes. Unlimited possibilities.

[ben @contraben]

Can @AdobeFirefly's "Edit" feature compete with Photoshop?

Select a region, describe the fix, keep the rest of the frame intact. That's the pitch.

Here’s what we observed across 4 sessions and 8 targeted edits with real working creatives:

>1 edit landed cleanly 🥇
>5 landed partial 🥈
>2 missed entirely ❌

Firefly understood the request almost every time, it just couldn't hold the rest of the image still while executing it.

Avoiding drift is hard.

In the partials, the target moved but something else broke.

>A product lost prominence / focus
>A shadow stayed broken
>A new artifact appeared on the wall while the format crept closer to the brief

Social was the hardest category. 0 clean wins across 4 attempts.

With social content, crop, pose, and product placement ARE the deliverable, so any drift fails the job.

TLDR: Firefly can edit like Photoshop, but avoiding drifts is the real challenge.

I personally use this feature a lot, how have you solved for this ?

[Stitch by Google @stitchbygoogle]

The Stitch Challenge is coming!

$10K in total prizes. May 27th.

Fueled by the good folks @contra

[ben @contraben]

In our @anthropicai Claude Design study,

5 designers approved a design system before they typed their first prompt.

>Brand palette
>type system
>components

the whole thing all set up.

Only 1 of them named any of it in their opening prompt.

That designer was the only one to finish production-ready.

The other 4 assumed Claude would carry the system over. It didn't.

TLDR: Claude doesn't reliably carry the design system you just approved. If you don't name it in the prompt, it doesn’t exist.

It's never been a better time to be a designer, but you must learn the art of the prompt.

[ben @contraben]

NEW RESEARCH 🔬:

>5 designers ran Claude Design on a real client brief.

Claude starts strong, but then every designer eventually reached for Figma.

Claude Design is a fast first draft, not a precision tool, designers still go to Figma for that.

Claude reads visual cues, the design system, and brief together.

It then fixes logos and wordmarks. Ships hover states and scroll behaviors. (love this)

but then the layout fails.

>60% of designers flagged spacing at edit 1.
>Every designer flagged it by edit 5.

Direct edits got ignored or undone.

TLDR: Generate the skeleton in @claudeai. Set the details in @figma. (Not dead lol)

Excellent work by the Anthropic team @nateparrott @Flomerboy @callmejohnnie

[ben @contraben]

Krea 2 Large is two different models.

>27% win rate when the style is described in words.
>51% when the style is a reference image.

A 24-point swing based on whether the brief is text or image.

On reference-image prompts, @krea_ai jumps to second place

>beating Gemini 3 Pro 🤯
>Seedream 5.0 Lite 🔥
>within 19 points of ChatGPT Images 2.0 👀

Model rankings without prompt-type context are misleading

Very impressive from the Krea team given how early it is @viccpoes @asciidiego

[ben @contraben]

"Text is solved", but "Typography" isn't.

>Every macro typography theme came in positive.
>Every micro theme came in negative.

@openai's GPT Images 2.0 nails the system, but then it breaks on execution.

>hierarchy
>composition,
>brand fit
>font choice

all net positive.

>size
>weight
>legibility
>broken characters

all net negative.

Three rounds of editing fixed contrast.

Every structural problem stayed where it started.

Letters rendered as the wrong letters and never got corrected.

Reality -> use GPT for the system. Take it to @figma for the execution.

Killer work from the GPT Image team @adele__li @BoyuanChen0 @kenjihata @kiwhansong0

[Contra Labs @contralabs_ai]

https://t.co/g13zZghdYN

[ben @contraben]

We benchmarked @openai’s ChatGPT Images 2.0 against the top image models. It won everything.

Then we asked the harder question: can brand designers really ship with it?

Here's where the model breaks.