Getting started

The fastest path is to hand this page to your agent. If you want to drive the API yourself, use the manual path below: five steps from code to results.

Path 0: Tell your agent

Paste this into Claude Code, Cursor, or any capable coding agent:

Fetch https://computalot.com/skill.md and follow it to set up Computalot
(on-demand GPU/CPU compute). Then use it to <describe your workload>
and report the results.

The skill guides the agent through authentication, balance checks, funding, job submission, and results. You review what comes back. Pricing explains the cost before anything runs.

The manual path: run your own code

Create a sandboxed project, upload your code, and submit typed jobs.

Prerequisites

export BASE_URL="https://computalot.com"
export TOKEN="flk_..."   # API key or wallet session token

If you do not have a token yet, see Authentication for the access model and the waitlist path. Before your first submission, read GET /api/v1/account/balance. Then you know what the account can spend. Billing covers holds, the ledger, and funding quotes.

Step 1: Create a project

curl -sS "$BASE_URL/api/v1/projects" \
  -X POST \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "my-project",
    "remote_dir": "/root/my-project"
  }'

Step 2: Write your code

Your project needs a Dockerfile (it defines the runtime), a computalot.project.json manifest, and your code. See Projects for the full setup guide.

# job.py
import json, os
 
payload = json.load(open(os.environ["COMPUTALOT_TASK_PAYLOAD"]))
print(f"Processing: {payload}")
 
result = {"status": "ok", "input": payload}
json.dump(result, open(os.environ["COMPUTALOT_TASK_RESULT"], "w"))

Step 3: Upload your project

tar czf code.tar.gz Dockerfile computalot.project.json job.py
 
curl -sS "$BASE_URL/api/v1/projects/my-project/push" \
  -X POST \
  -H "Authorization: Bearer $TOKEN" \
  --data-binary @code.tar.gz

After a successful push, the latest revision is published immediately. You can submit jobs immediately. The first job can take longer while Computalot prepares runtime capacity on demand.

Optional: to prepare currently available workers before a burst, call init:

curl -sS "$BASE_URL/api/v1/projects/my-project/init" \
  -X POST \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{}'

If project init returns 402 Payment Required with a shortfall quote, fund the account. Then retry the same POST /api/v1/projects/my-project/init request.

Step 4: Submit a job

curl -sS "$BASE_URL/api/v1/jobs" \
  -X POST \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "type": "structured_runner",
    "runner_command": ["python3", "job.py"],
    "payload": {"test": true},
    "project": "my-project",
    "timeout_s": 120
  }'

If job submit returns 402 Payment Required with a shortfall quote, fund the account. Then retry the same submit request.

To see whether the revision is only published (can_accept_new_jobs: true, ready_for_jobs: false) or already warm (ready_for_jobs: true), read GET /api/v1/projects/my-project/status.

Step 5: Read results

curl -sS "$BASE_URL/api/v1/results/<job_id>" \
  -H "Authorization: Bearer $TOKEN"

What things cost

Every submit response includes summary.billing_estimate, the authoritative quote for that job. The unused hold is released at settlement. Indicative rates and worked examples are on Pricing.