Verifiers
A library of modular components for creating RL environments and training LLM agents.
AGENTS
Agentic Coding
Train and deploy coding agents that plan, write, and verify software autonomously — powered by RL environments and verifiable rewards.
FIG. 01 / SETUP
48 h
From zero to a self-improving loop
FIG. 02 / ENVIRONMENTS
2,500+
RL environments ready to compose
FIG. 03 / OVERHEAD
0
Reward engineering required
Environment Hub
Access and contribute to 2,500+ open-source RL environments and a community of researchers and developers.
Request trainingCoding agents are only as good as the feedback they learn from. With verifiable rewards — tests passing, builds compiling, diffs applying cleanly — agents can be trained end-to-end against real engineering tasks instead of proxy benchmarks.
Use the Environments Hub to compose repositories, toolchains, and test harnesses into reproducible RL environments, then train with Prime-RL across on-demand or reserved clusters.
Every rollout runs in isolated sandboxes, so agents can execute code, browse, and call tools safely at scale. What passes in training is what ships in production.
Platform
Everything this solution needs. None of the overhead.
1
Environments Hub
Compose repositories, toolchains, and test harnesses into reproducible RL environments — versioned and shareable.
2
Prime-RL Training
Train across on-demand or reserved clusters, from H200 to B300 — without writing any infrastructure glue.
3
Sandboxed Rollouts
Every rollout runs isolated, so agents can execute code, browse, and call tools safely at scale.
4
Verifiable Rewards
Score outcomes against real signals — tests passing, builds compiling, diffs applying cleanly.
PI is the closed-loop AI stack where models observe their own failures, generate new training signal, and improve — without you writing a single reward function.
Deploying with
At industry standards
FIG. 01 / EFFICIENCY
3.2×
Avg. improvement per cycle
FIG. 02 / VELOCITY
48 h
First improvement deployed
FIG. 03 / SCALE
100B+
Parameters on-stack
FIG. 04 / LABOR
0
Reward functions written
Self-generating eval suites
Auto-instrumented agents
RL across 2,500+ environments
Serverless model inference
Real-time training observability
Production data flywheel
[ What teams say ]
"We used to spend two weeks curating evals. PI generates better evals from failures in two hours. We just don't think about reward engineering anymore."
2 wks → 2 hEval curation time
Compute. Find reliable compute operated globally from a single GPU to largest clusters.
On demand
Instant access to 1-256 GPUs. Use your GPUs across clouds in a single platform.
- 1.1
SLURM, K8s Orchestration
Orchestrate dynamic workloads with enterprise-grade scheduling and container automation.
- 1.2
Infiniband Networking
Scale distributed training with high-bandwidth interconnects across nodes.
- 1.3
Grafana Monitoring Dashboards
Visualize metrics in real time with customizable dashboards for full system observability.
Liquid Reserved Clusters
Request large-scale clusters from 50+ providers. Sell-back idle GPUs to our spot market.
- 2.1
Get quotes from 50+ datacenters within 24 hours
One request, parallel bids for options, from H100, H200, to B200, B300, GB300 NVL72
- 2.2
Re-sell idle GPUs back to our spot market
Resell idle nodes on our spot market with no manual ops. Reclaim capacity instantly when you need it.
- 2.3
Direct assistance from our research and infra engineering team
Dedicated solutions engineer from cluster bring-up through steady-state
[ Ready? ]
Your model should get smarter while you sleep.
Engineers, at full stride. Point PI at your checkpoint — first improvement in about 48 hours.