TRAIN · P95

AI experiments.
Under control.

P95 is Numerata's agentic training platform: manage every training job from anywhere, from a single run to hundreds running in parallel.

Visit site →

INSTALL

Ships as an agent skill. Add it and it instruments your experiments automatically.

AGENT SKILL: let your agent instrument runs and report back

npx skills add numerataz/instrument-experiment

OR CALL IT YOURSELF: install the Python SDK directly

pip install p95
# or
uv add p95

BUILT FOR THE FULL WORKFLOW

From a single run to hundreds of parallel jobs, P95 scales with you.

HYPERPARAMETER SWEEPS

Define a search space and let P95 schedule trials automatically. Grid, random, or Bayesian search. Results stream in as each run completes.

CLOUD JOBS

Submit training jobs to the cloud directly from your CLI or SDK. Monitor progress, stream logs, and compare results against previous runs, all in one place.

TEAM COLLABORATION

Share projects and runs with your team, and compare experiments side by side.

INSTRUMENT ONCE, SCALE ANYWHERE

Wrap your training loop with a Run context. The same code runs locally, in a sweep, or as a cloud job.

TRAIN.PY

from p95 import Run

with Run(project="resnet", name="lr-0.001") as run:
    run.log_config({"lr": 0.001, "optimizer": "adam"})

    for epoch in range(num_epochs):
        loss, acc = train(model, loader)
        run.log_metrics({"loss": loss, "acc": acc}, step=epoch)

Configs, metrics, and tags
in one place.

Runs automatically
grouped under sweeps.

Works with PyTorch, JAX,
TensorFlow, or plain Python.

BROWSE RUNS WITHOUT LEAVING YOUR SHELL

A clean, keyboard-driven interface. Navigate projects, inspect run configs and metrics, and watch live charts update, all without leaving your terminal.

GET STARTED

Ready to run your first sweep? Create an account and start tracking experiments in minutes.

Explore demo →