Prime Intellect (@PrimeIntellect) / X

Prime Intellect

2,864 posts

Prime Intellect

@PrimeIntellect

Open Superintelligence Stack

Joined June 2020

Affiliates Replies Media

Pinned
Prime Intellect
@PrimeIntellect
Jul 8
Announcing our $130M Series A to build the Open Superintelligence Stack Led by Radical Ventures, with NVIDIA, Intel Capital, Dell Capital, and existing investors Train, deploy, and continuously improve your own models using our stack. Own your intelligence.
00:00
Prime Intellect reposted
kenneth
@kennethnym
Jul 13
today is my first day @PrimeIntellect i'm so grateful to be able to work alongside such incredible team, i have so much to learn !! :)
Prime Intellect reposted
AI Engineer
@aiDotEngineer
Jul 13
Congrats to PI on the unicorn round and $100M ARR! we were proud to have @willccbb introduce verifiers at the first AIE NYC a year ago and now... it is v1! Will joins a rare list of three-time AIE speakers, and his talk on the full PI stack is linked below!
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect reposted
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect reposted
Florian Brand
@xeophon
Jul 13
verifiers v1 is finally out, something we worked on for months to nail it properly Lets talk about some of the non-obvious things that these changes unlock:
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect reposted
Mika Senghaas
@mikasenghaas
Jul 13
super proud to get this out and in the hands of the people !! a quick thread on some personal highlights
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect reposted
Anirudh Ravichandran
@anravich94
Jul 13
We envision verifiers V1 becoming the standard format for environments. It's built to be highly composable, support training and evaluation on rollouts and yet be highly flexible. Put plainly, these abstractions make it really easy to design your environments, and evaluate and
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect reposted
vLLM
@vllm_project
Jul 13
🎉 Congrats to @PrimeIntellect on Verifiers v1! Its training rollouts run on vLLM for exact token IDs and logprobs, no tokenization drift, keeping rollouts and training in sync. vLLM powers a growing set of open RL infra, prime-rl and others, and it's an area we're going deep
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect reposted
λux
@novasarc01
Jul 13
some cool things i liked about verifiers v1: - my favorite shift from v0 to v1 is the move from an environment-centric abstraction to a rollout-centric one. v1 has this clean abstraction: taskset × harness × runtime → trace tasksets owning both data and scoring is much
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect reposted
will brown
@willccbb
Jul 12
you can now train: - a 100B-parameter reasoning model - for 40-turn SWE agent tasks - in your own coding harness - for 1000 RL steps - on just 6 H200 nodes - in under 2 days infra co-design is magical
Prime Intellect
@PrimeIntellect
Jul 12
Replying to @PrimeIntellect
verifiers v1 plugs straight into prime-rl for training. We have been using v1 internally for all our production runs. In this run, we train GLM-4.5-Air on ScaleSWE tasks with under-4-minute steps and 35-turn rollouts, completing 1K steps in 2 days on just 6 H200 nodes.
Prime Intellect reposted
will brown
@willccbb
Jul 12
this has been a labor of love for months with @mikasenghaas and @xeophon driving incredible progress, and `v1` is now finally ready for prime time we set out to fully modernize `verifiers` for the agent harness era, and unlocked some insane efficiency gains along the way
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect
@PrimeIntellect
Jul 12
Today, we are releasing verifiers v1 — an overhaul of our environment stack for the modern era of agentic RL and evals. We decompose environments into a taskset, a harness, and a runtime. Run complex agentic tasks like coding and computer use at scale, in any harness.
Prime Intellect
@PrimeIntellect
Jul 12
Replying to @PrimeIntellect
verifiers v1 plugs straight into prime-rl for training. We have been using v1 internally for all our production runs. In this run, we train GLM-4.5-Air on ScaleSWE tasks with under-4-minute steps and 35-turn rollouts, completing 1K steps in 2 days on just 6 H200 nodes.
Prime Intellect
@PrimeIntellect
Jul 12
Read the full verifiers v1 deep dive:
verifiers v1: Decomposing Tasksets and Harnesses for Agentic RL & Evaluations
From primeintellect.ai
Prime Intellect reposted
Will Knight
@willknight
Jul 10
For my newsletter this week, I tried fine-tuning my own models using tools from @PrimeIntellect and @adaption_ai. Most of the big labs are using AI to build powerful models, but the same approach could also give people greater control over their own AI.
I Built a Self-Improving AI, and So Can You
From wired.com
Prime Intellect reposted
Kyle Kranen
@KranenKyle
Jul 10
It has been an absolute pleasure to work with the @PrimeIntellect team to accelerate their inference and PrimeRL with Dynamo! Check out the blog detailing our collaboration!