kevin (@kevinjosethomas) — posts, photos & videos

@kevinjosethomas reposted

@jannik_stra

prime agent tapping into all of our offerings - prime sandboxes, prime inference - to rewrite itself. next it can start a training run on glm 5.3 using prime's model factory (stay tuned)

Prime Intellect@PrimeIntellect · 15h

Over two weeks, Prime Agent orchestrated a swarm of over 2,000 agents to rewrite itself in Rust.

With 10,000+ sandboxes, 200B+ GLM-5.3 tokens, and 16,000 agent-to-agent messages, this was our largest test yet of Prime Agent’s multi-agent capabilities and our infrastructure powering these agent swarms.

Prime Agent now reaches usable input ~13× faster and uses 83% less startup memory than before.

0:15

Open quoted post →

4 5 0 59 2.8K 19

@kevinjosethomas reposted

josh

@josh_soupy

14h

3 weeks ago i got this message 🤯

kevin@kevinjosethomas · 15h

This has been insanely fun to work on for the past couple weeks!

This rewrite tested the limits of all our products: thousands of subagents via Prime Agent, billions of tokens via Prime Inference, and hundreds of sandboxes via Prime Sandboxes!

It's incredibly rewarding to see our full stack come together, and Prime Agent is now faster than ever :)

Open quoted post →

1 1 0 9 878 1

@kevinjosethomas reposted

Lan

@ad0rnai

15h

We rewrote Prime Agent in Rust with 5x faster cold start, 8x less memory, 2x streaming throughput, and full behavioral parity.

Prime Agent is a harness designed from the ground up for long-running agents. The RLM architecture allows many agents to coordinate across a session that outlives any single context window, and is what enabled 2,000+ agents to rewrite the harness they were running on.

Prime Intellect@PrimeIntellect · 15h

0:15

Open quoted post →

9 2 0 62 3.7K 23

@kevinjosethomas reposted

Prime Intellect

@PrimeIntellect

15h

Across the six harnesses we benchmarked, Prime Agent Rust had the lowest time to usable input, startup memory, and installed size.

The rewrite also split the codebase into nine crates with enforced dependency boundaries. Changes to shared protocol types are now checked across the client, daemon, and session workers.

3 4 2 83 7.8K 22

@kevinjosethomas reposted

Seth Karten

@sethkarten

15h

Agent swarms turn software work into dependency graphs and state machines.

We limit-tested this by having Prime Agent automate its own Rust rewrite.

Now enjoy one of the fastest CLI agents :)

Prime Intellect@PrimeIntellect · 15h

0:15

Open quoted post →

8 9 0 115 6K 38

@kevinjosethomas reposted

Florian Brand

@xeophon

15h

one of my widely held beliefs is that open models are way more capable than ppl (and benches) realize

this project is one of many that reinforced my beliefs. an incredible team + a few tokens are all you need : )

Prime Intellect@PrimeIntellect · 15h

0:15

Open quoted post →

16 7 0 157 6.5K 35

@kevinjosethomas reposted

samsja

@samsja19

15h

prime agent rewrote itself in rust , all powered by glm5.3 and our own inference, spawning thousands of subagent that all communicate with each other

🦋🦋

Prime Intellect@PrimeIntellect · 15h

0:15

Open quoted post →

15 12 0 227 12.3K 66

kevin

@kevinjosethomas

15h

Prime Intellect@PrimeIntellect · 15h

0:15

Open quoted post →

11 10 2 92 7.2K 15

kevin

@kevinjosethomas

17h

amazing blog by @hallerite! and contextualizes a lot of our motivations behind Prime Agent and RLM :)

Prime Intellect@PrimeIntellect · 17h

Inference scaling works, but every agent eventually faces the same problem: its context window fills up.

In this essay, we argue the only scalable fix leads straight to swarms.

0:10

Open quoted post →

1 1 0 26 1.3K 7

@kevinjosethomas reposted

Prime Intellect

@PrimeIntellect

Oct 8

Open superintelligence accelerates open scientific research.

The Alzheimer’s Translation Challenge will share a 150M cell atlas on the Alzehimer's Disease workbench and Prima Mente's modeling platform.

Prime Intellect is lending infrastructure support to researchers to build models based on this dataset. The best hypotheses will be tested in Prima Mente's wet lab.

0:10

Prima Mente@PrimaMente · Oct 8

Today we're launching the Alzheimer's Translation Challenge, a global AI competition to discover new therapeutic hypotheses for Alzheimer's disease.

At its core is a new 150M cell atlas of neurons, astrocytes, and microglia across different genetic backgrounds under combinatorial perturbations, with multi-modal readouts.

The data will be made available through the AD workbench, Hugging Face and Prima Mente's modeling platform.

The challenge is led by @PrimaMente and @AlzData in collaboration with @nvidia, @huggingface, @nebiusai, Talisman Therapeutics, @UltimaGenomics, @cellanome, @PrimeIntellect, @GoodfireAI, and @boltz_bio.

In 1906, Alois Alzheimer described a new disease of the brain. After 120 years of progress, we still don't fully understand what drives this disease. We want to change that.

Participants will use the dataset to build AI models, assessed through a series of evals from general benchmarks to more complex tasks.

Top teams will be shortlisted, and their experimental hypotheses tested in Prima Mente's wet lab. Teams win when their hypotheses demonstrate the predicted molecular and functional effect.

0:41

Open quoted post →

11 44 11 432 25.6K 164

@kevinjosethomas reposted

Stu Cao

@caoshuxun

Oct 4

Two months ago, I left Annapurna Labs, ending a five-year chapter at Amazon. Amazon treated me exceptionally well and gave me more opportunities to grow than I can count. I’m grateful for the people I worked with and everything I learned there.

I followed my heart to Prime Intellect. Its mission to build an open superintelligence stack deeply resonated with me. I believe every person and every business should be empowered to build their own intelligence. Your AI agent should be yours, not a third-party stranger.

Walking into Prime’s SF office, I felt ten years younger. These past two months have been the most exciting of my career.
Three things stand out:

1. The people.
Prime’s team doesn’t all come from famous AI labs, but I’ve rarely seen this level of curiosity and determination.
When we made our high interactivity GLM-5.3 serving endpoint available internally, everyone switched their agents to it. They never went back.
The speed improved productivity. Just as importantly, the team kept sharing feedback, helping us refine performance and tool-calling reliability. Soon, RL rollouts, synthetic data generation, and long-horizon agent research were running on the same infrastructure.
As of this week, our internal inference usage is 600–700 billion tokens per day. That experience gave us the confidence to open the service to the public.

2. The full stack.
Prime brings together hosted training with prime-rl, verifiers, sandboxes, an RL environment hub, the Prime Agent harness, and now Prime Inference.
You can train, deploy, and run agents in one place, without stitching together infrastructure through a massive engineering effort. And most of these components are fully open source.
Our inference stack builds on that same foundation, with deep collaboration across NVIDIA Dynamo, Inferact, and the broader vLLM community.

3. The compute.
For anyone working on LLM infrastructure, access to compute at this scale is a privilege. Building and operating inference across large scale of GB200 racks and B300s is not something I take for granted. Vera is already in our hands, and Rubin is on the horizon. NVIDIA’s latest hardware gives us the foundation to push inference performance further.

Alongside the public launch, we published a technical post on how we built Prime Inference: primeintellect.ai/blog/prime-inf…
Inference has moved incredibly fast over the past year, and it can be hard to separate useful advances from the noise. We use GLM-5.3 optimization to show the work end to end: how we serve agent workloads with low latency while also supporting high concurrency, dependable availability, and reliable tool calls. Recommend for a read if you are interested in learning about P/D, topology choice, NVFP4 KV compression kernel fusion, tool call reliability.

Fast inference matters. Serving real agents reliably at scale is what counts.

23 14 0 162 10.7K 72

kevin

@kevinjosethomas

Oct 2

absolute goats @chimcis @caoshuxun! enabling and encouraging my 10B+ daily token spend 😵‍💫

Prime Intellect@PrimeIntellect · Oct 2

Introducing Prime Inference:

We've served trillions of tokens for RL and dedicated customer deployments

To own your intelligence, you need to own your inference

Unpacking our inference stack

0:45

Open quoted post →

4 1 0 67 3.5K 9

kevin

@kevinjosethomas

Oct 1

got pipped

2 0 0 28 1K 2

kevin

@kevinjosethomas

Sep 24

prime agent v0.9.6 is out:

◆ Support for GPT-6 Sol, Opus 5.5, and Grok 4.7
◆ /mcp plugin catalog with one-click connections to Linear, Notion, Posthog, Stripe, and 60+ more services
◆ Huge perf and reliability pass 🫡

Lots more coming soon :)

0:06

24 13 3 195 12.6K 54

kevin

@kevinjosethomas

Sep 24

We’re hiring a full-time engineer to help build Prime Agent :)

Own features end to end across agent harnesses, cloud agents, and self-improving loops -- integrated with our sandboxes, evals, and hosted training stack.

Full-time, in person in SF: primeintellect.ai/careers/8597d1…

27 12 3 293 16K 153

kevin

@kevinjosethomas

Sep 22

morning: gym with chi
workday: office with chi
evening: home with chi
weekend: hang with chi

chi@chimcis · Sep 22

Last week, I joined @PrimeIntellect as an intern to help optimize their inference stack :)

Very grateful to get the chance to work with such a talented team of researchers and engineers

Plenty to do and much to learn, super excited for what’s ahead!

Open quoted post →

8 1 0 93 5.3K 8

kevin

@kevinjosethomas

Sep 16

v0.9.5 brings a refreshed TUI, lots of perf improvements, and more resilient providers. Prime Agent also ships a standalone binary now 🫡

Seth Karten@sethkarten · Sep 16

prime agent v0.9.5

we fixed a lot of bugs and, of course, we had prime agent feature its favorite updates

it picked our perf work. then it created the video itself.