Twitxy Saved

Diambra

Verified account This account is verified. Learn more

@diambra_ai

Gaming environments for multi-agent competition and multi-agent cooperation research
AIvsAI: app.diambra.ai/fight
T&C's: old.dev.diambra.ai/terms
diambra.ai Joined May 2022
105 Following 8.1K Followers 1.6K Posts

Latest posts 5

Diambra Verified account @diambra_ai
Excited to see what Cambridge University students can come up with!
Bill Leoutsakos Verified account @Bi11Leou Nov 14, 2025
I am helping organise a hackathon on the RL environments space with DIAMBRA (diambra.ai) which is building RL training environments for AI agents.

This is a 3 week long hackathon starting on the 17th of November, where you have to train your own agents in the diambra platform (compute provided) to win in the street fighter game - your AI will fight other participants AI in the finals.

The prize pool is $3k! And a follow up tournament in 1 month.

Some big sponsors such as 0G Labs and Mistral AI. And team is discussing KRAFTON Inc., Unity (Game engine), Supercell and Anyscale to come in too.

Sign up here: diambra.ai/cambridge , then join the what's app group.

If you are a cracked dev who enjoys building on the cutting edge of AI don't miss it! Feel free to DM me for any questions or write them in the comments.
View quoted post on X
Diambra Verified account @diambra_ai
What DIAMBRA has been doing for over 4 years is finally getting picked up by the big VCs like Andreessen Horowitz and large AI labs such as Anthropic

At Diambra, we built 9 RL environments in gaming and benchmarking engine, able to spin thousands more for emulated games - and fully OpenAI Gym compatible -> Meaning easy expansion to robotics too (e.g. MuJoCo).

Article:
“All the big AI labs are building RL environments in-house,” said Jennifer Li, general partner at Andreessen Horowitz, in an interview with TechCrunch. “But as you can imagine, creating these datasets is very complex, so AI labs are also looking at third-party vendors that can create high-quality environments and evaluations. Everyone is looking at this space.”

According to The Information, leaders at Anthropic have discussed spending more than $1 billion on RL environments over the next year.

The hope for investors and founders is that one of these startups emerge as the “Scale AI for environments,” referring to the $29 billion data-labeling powerhouse that powered the chatbot era.

techcrunch.com/2025/09/21/sil…
techcrunch.com Silicon Valley bets big on 'environments' to train AI agents | TechCrunch A wave of startups are creating RL environments to help AI labs train agents. It might be Silicon Valley’s next craze in the making.
Diambra Verified account @diambra_ai
@FakePsyho , a Polish coder who developed OpenAI's FIVE algorithm for game Dota 2, wins the AtCoder WorldTour Finals - the most prestigious invite only coding competition annually.

This year beating OpenAI, securing humanity's excellent over AI (for now!)

We hope this inspires coders in our platform developing AI algorithms using games - Psyho is a clear example of the synergies between Gaming and AI for AI research/advancement.

(video by TVP World)