Wollnut Labs
For Developers

Build with Wollnut Labs

Programmatic access to enterprise GPUs. Spin up H100s, H200s, and B200s with a single API call.

API ReferenceGet API Key

Get Started in 3 Steps

From zero to GPU in under a minute.

1

Create an API Key

Sign up and generate an API key from your dashboard settings. Keys use the wn_ak_ prefix.

Dashboard Settings
2

Choose a GPU

Browse available GPU plans — H100, H200, B200 — and pick the right one for your workload.

View Plans
3

Launch & Code

Create an instance via API, SSH in, and start training. Billing is per-minute, stop anytime.

API Docs

Quickstart

Copy-paste examples to get started fast.

bash
# List your instances
curl -s -H "Authorization: Bearer wn_ak_your_key" \
  https://your-domain.com/api/v1/instances | jq

# Create a new GPU instance
curl -s -X POST \
  -H "Authorization: Bearer wn_ak_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "training-run-01",
    "planId": "GPU.H100.1x",
    "templateSlug": "pytorch-cuda",
    "sshKeyId": "your_ssh_key_id"
  }' \
  https://your-domain.com/api/v1/instances | jq

# Check your credit balance
curl -s -H "Authorization: Bearer wn_ak_your_key" \
  https://your-domain.com/api/v1/billing/balance | jq

# Stop an instance
curl -s -X POST \
  -H "Authorization: Bearer wn_ak_your_key" \
  https://your-domain.com/api/v1/instances/INSTANCE_ID/stop | jq

Authentication

All authenticated endpoints require a Bearer token.

API Key Format

All Wollnut Labs API keys use the wn_ak_ prefix followed by 64 hex characters. Keys are hashed with SHA-256 before storage — we never store plaintext keys.

http
# Include in every request
Authorization: Bearer wn_ak_a1b2c3d4e5f6...

API Key Scopes

Restrict what each key can access by assigning scopes:

allFull access to all endpoints
instancesCreate, read, start, stop, delete instances
billingRead balance, usage, transactions
ssh-keysManage SSH keys

Rate Limits

Requests are rate-limited per user to ensure fair usage.

Endpoint TypeRate Limit
Instance Create10 requests / minute
Instance Actions30 requests / minute
Billing Operations10 requests / minute
Read Operations120 requests / minute
Public Endpoints60 requests / minute

When rate-limited, the API returns 429 Too Many Requests. Retry after a short delay.

Platform Features

Built for production ML and AI workloads.

Per-Minute Billing

Only pay for what you use. Instances are billed by the minute with automatic stop when credits run out.

Secure by Default

API keys are SHA-256 hashed, SSH key authentication, and all API calls validated with Zod schemas.

Real-Time Status

Background jobs continuously sync instance status from E2E Networks. Get instant updates on provisioning.

Full REST API

Manage everything programmatically — instances, billing, SSH keys, templates, and more.

Start building today

Sign up for free and get $5 in credits to explore the API.

Get Started FreeFull API Reference