Log inSign up
Log inSign up
Contra Labs
290 posts
Contra Labs profile banner
@contralabs

Contra Labs

@contralabs
Creative intelligence.
New York, NY
contralabs.com
Joined November 2025
1,437 Following
3,849 Followers
RepliesRepliesRepostsRepostsMediaMediaArticlesArticles

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
  • Pinned
    @contralabs
    Contra Labs
    @contralabs
    Oct 8
    3 years of data. 1 finding that ties it together: AI raised the bar for creativity. Creative Intelligence 2026 is live.
    @contraben
    ben
    Contra
    @contraben
    Oct 8
    AI is making creativity more valuable than ever before and.. we have 3 years of data to prove it. Introducing the Creative Intelligence 2026 report from @contralabs. Full report: c87.to/creative-intel…
    00:00
    10
  • @contralabs
    Contra Labs
    @contralabs
    Sep 14
    GPT-6 Astra, the latest frontier-level model from @OpenAI, is the only model in our latest Human Creativity Benchmark study to build a complete, working user flow. It lands 2nd overall on our leaderboard, behind Claude Fable 5.1 and ahead of GPT-5.6 Sol in 4th. We ran it against
    7
  • @contralabs
    Contra Labs
    @contralabs
    Sep 11
    ChatGPT Images 2.5 Flare lands 2nd overall on our Human Creativity Benchmark leaderboard, almost on par with Muse Image and ahead of ChatGPT Images 2.0 and Nano Banana Pro. It is the latest frontier-level image model from @OpenAI. We ran it against Muse Image, Krea 2 Medium,
    4
  • @contralabs
    Contra Labs
    @contralabs
    Aug 27
    Seedance 2.5 tops our motion video leaderboard. It won 72.6% of comparisons against Ray V3.2, Gemini Omni Flash, Grok Imagine, and Seedance 2.0. And it gets stronger the more specific the prompt.
    00:00
    6
  • @contralabs
    Contra Labs
    @contralabs
    Aug 26
    Article cover image
    Article
    How model performance shifted across four website design evals
    Across four website design evals, we saw that model performance shifted as the prompts and eval setup changed. Claude Fable 5 and Claude Opus 5 both improved against Kimi K3, while GPT 5.6 Sol...
    1
Edit with