Hosted RL

Train RL policies on cloud GPUs

Pick an environment — gymnasium classic control, Atari, MuJoCo, LLM RLHF, or robotics sim — pick a baseline algorithm repo, pick a GPU, and stream training metrics live. When the reward curve flattens, deploy the trained policy as a scale-to-zero cog or serverless endpoint.

Launch a training run OSS starter repos

CLI

app api POST /api/rl/jobs '{
  "envId": "cartpole-v1",
  "algoId": "ppo",
  "hardware": "gpu-rtx4090"
}'
app api GET  /api/rl/jobs/<job-id>          # poll metrics
app api POST /api/rl/jobs/<job-id>/deploy \
  '{"target": "serverless"}'                # policy endpoint

Per-second GPU billing from the shared credit balance; jobs scale to zero the moment training finishes.

1 · environment

Pick an environment

2 · algorithm repo

Pick a baseline

3 · gpu tier

Pick a GPU

Per-second billing with the platform margin included — the same catalog as Cog Studio and pricing.

live training metrics

Watch the reward curve climb

Launch a run above to stream reward, loss, entropy, and throughput here — polled live from /api/rl/jobs/{id}. Signed-out visitors get an instant simulated preview.

open source

Starter repos

Every baseline on this page is an MIT-licensed repo you can clone, hack, and bring back — each ships a clean single-file training loop, a cog.yaml, an app.nz policy server, and a serverless deploy manifest per docs/BUILDING_COGS.md.

MIT · PPO + DQN

appnz-rl-ppo-starter

Single-file PPO on gymnasium (classic control, Atari, MuJoCo) with cog.yaml, an app.nz policy server, and a serverless deploy manifest.

github.com/lee101/appnz-rl-ppo-starter
MIT · GRPO

appnz-rl-rlhf-lite

Minimal GRPO-style RLHF loop on a tiny LLM with programmable rewards, packaged for cog + serverless deploy on app.nz.

github.com/lee101/appnz-rl-rlhf-lite

Bring your own repo too: any container that trains and then serves the cog HTTP contract can run here. Build it with hosted builds, register it on /cogs, and the training metrics API is one POST away.

From reward curve to production endpoint

Train on per-second billed GPUs, deploy the policy as a scale-to-zero endpoint, and call it from anything with an HTTP client.

Launch a run Cog Studio Install the CLI