Callosum

The Intelligent Systems Company. Co-evolving chips & intelligence.

[Callosum @CallosumAI]

Today we announce our $100M seed round to redefine how humanity computes in AI's next chapter.

The future of compute and AI is heterogeneous. Callosum is building it:
https://t.co/t5KnOE0Sos

[Callosum @CallosumAI]

Our CTO @achterbrain joined @nvidia on stage at the start-up day of GTC Berlin, alongside Clara Cornu, @talfanevans, @nouralhassan and Matteo Sarcinella.

He discussed how Callosum runs heterogeneous workloads efficiently across every generation and type of NVIDIA hardware, and what it takes to scale new technology companies fast in Europe.

Thanks for having us @nvidia

[Callosum @CallosumAI]

Watch the full conversation here: https://t.co/bpOg3IMqoT

[Callosum @CallosumAI]

Our CTO, Jascha @achterbrain, joined The Compute 100 to discuss neuroscience, the multi-chip world and what we're building at Callosum.

Thanks for having us @gabyllorenzi @BSchech @thecompute100

[The Compute 100 @thecompute100]

Today’s guest is @achterbrain , co-founder and CTO of @CallosumAI .

Jascha’s bet is that the future of compute is heterogeneous: different chips working together on different parts of a problem.

Beyond the multichip future, we discuss:

- What neuroscience taught Jascha about designing computing systems

- How chipmakers discover what their hardware is best at

- Where thermodynamic, biological, and quantum computing could fit

- Hiring people who welcome having their ideas challenged

- Building an AI infrastructure company in London

Timestamps:
0:00 Introduction
2:55 What heterogeneous computing actually means
11:02 From PhD research to Callosum
18:36 Turning heterogeneous compute into a product
23:24 Sourcing compute and working with chipmakers
28:53 Matching workloads to the right hardware
34:24 What heterogeneous computing can do today
38:29 The next waves of chips
48:02 Building the team
51:33 Why build in London?
56:11 Helping enterprises upgrade their compute stacks
1:01:04 Fundraising and building momentum
1:03:40 Where specialization wins, and where it doesn’t

Listen to the show, and follow along for new episodes every Tuesday.

[Brian Schechter @BSchech]

@CallosumAI raised a $100M seed on the idea that the future of compute is many different chips, each handling the part of the problem it’s best at. Great conversation with @achterbrain about how they’re building it. New episode of The Compute 100.

[Society for Technological Advancement @sotalikesfuture]

SoTA Scientific Parallelism Hackathon | 21st-22nd November | London

Which scientific questions could we investigate better if computation were faster or could run at a greater scale?

The Society for Technological Advancement (@sotalikesfuture), supported by Callosum (@CallosumAI), Inherent Labs (@inherent_labs), the Advanced Research + Invention Agency (@ARIA_research) and Fractile (@fractile_ai), is holding a hackathon to tackle open scientific questions that could benefit from faster computation or experiments at a greater scale.

Contestants will use parallel computing on challenges across biology, chemistry, physics, materials science, astronomy, and Earth sciences ��� pushing GPUs to their limits and rethinking architectures for this new paradigm.

[Diogo Almeida @CompleteSkeptic]

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?

I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev

• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions

AFAICT the shortest path to AI-based economic revolution

[Cerebras @cerebras]

“If all AI is built on one platform, we'll only have a certain envelope of future outcomes. Whereas if the future of AI is built on a broad population of platforms by a broad population of users and organizations and nations, we're much more likely to experience the benefit of this technology rapidly and avoid the risks.”

Andy Hock and @DanAkarca of @CallosumAI on the vision behind the new Cerebras-Callosum partnership.

[Danyal @DanAkarca]

“We're in the early stages now of this Cambrian explosion. What are the boundary cases where we want certain tasks on certain models on certain silicon? We’re making that programmable. This is where all the frontier performance is going to come from.”

Proud to launch our new partnership between @CallosumAI and @cerebras. Thanks to Andy for a great conversation!

[Cerebras @cerebras]

"You now have new surface area to expose the benefits of the silicon in new ways.

We had some really amazing results when doing this over a huge amount of different silicon. Cerebras was unbelievably good.

We could actually not only improve the speed by a significant amount in these long-context reasoning problems, we also found that we could also make them very cost-effective because we're being efficient with the tokens."

— @DanAkarca, co-founder of @CallosumAI, in conversation with our Andy Hock on the new Cerebras-Callosum partnership.

[@dMatrix @dMatrix_AI]

Heterogeneous compute is becoming the standard for AI infrastructure.

@CallosumAI just announced their global partner network. d-Matrix is proud to be part of the US ecosystem alongside @Intel, @AMD, and others.

No single chip does every job well. The right workload deserves the right silicon.

Corsair was built for exactly this: purpose-built for inference, with the memory bandwidth and efficiency to make AI systems practical instead of unsustainable.

Read more: https://t.co/nqZCmgeR2g

#AIInference #HeterogeneousCompute

[Danyal @DanAkarca]

Very exciting times for @CallosumAI + @Cerebras 🧠

[Cerebras @cerebras]

Faster agent loops. New systems that were not practical before. Cerebras inference is coming to @CallosumAI APIs. 🟧

[UK Business and Trade - North America @biztradegovukNA]

.@UKSovereignAI has made its first-ever investment, backing British AI infrastructure company @CallosumAI in a $100M seed round with US investor @DCVC and others 🤝

Proof of the UK's £1.1B (~$1.4B) AI infrastructure investment, and what's on offer for North American businesses.

[SambaNova @SambaNovaAI]

@CallosumAI

[SambaNova @SambaNovaAI]

Congrats to the @CallosumAI team on the $100M seed round! 🎉

We’re excited to be partnering around a shared vision for heterogeneous AI compute, using the right chip for the right workload.

As inference becomes more demanding, we're especially excited to power the fast, premium decode layer with SambaNova. 🦾

[Jason Pontin @jason_pontin]

And, in a day of @Callosum news, @cerebras announces its partnership with the company here:

https://t.co/Ek4nWLJe8K

[Charles Kremer @krech565]

Really excited to share some of what I have been working on!

[Danyal @DanAkarca]

New levels of cost, speed & capability attained from programming heterogeneity in critical workloads, with our partners @HelmGuard & others

[Callosum @CallosumAI]

4/ Read the full case study here, including what more of our partners are saying: https://t.co/Dln7rqQZCE.

[Callosum @CallosumAI]

We’re excited to release Tailored Inference, unlocking new frontiers in speed and cost-efficiency by optimising inference to the right models, algorithms and chips.

By leveraging the inherent structure of inference workloads, it allows us to decompose them to discover new solutions with improved cost-efficiency for our partners.

[Callosum @CallosumAI]

In these critical workloads, we found solutions built around closed-source frontier models are typically slow and/or expensive, prohibiting fast detection of threats at scale.

By discovering new fast and cheap solutions, it allows both rapid and frequent detection, enabling a new level of capability.