Modal Sandboxes

Disposable remote containers (with optional GPU) via Modal, paid per call in USDC. No Modal account, no GPU procurement.

Base only in wallet mode; fine on an API key. sol.blockrun.ai carries the /v1/modal/* routes but has no Modal backend configured, so every action — create, exec, status, terminate — answers 503. That reads as "the sandbox service is down" rather than "wrong chain", which is exactly the wrong conclusion to act on: retrying will not help. The tool checks the active chain first and says so. Switch with blockrun_wallet action:"chain" chain:"base". Prices below are Base prices and include its per-transaction fee.

READ THIS BEFORE SETTING timeout

timeout is the BILLED lifetime, charged upfront in full, and never refunded — not an idle timeout. Above 300s the price switches from a flat rate to per-hour billing for the entire duration you ask for, whether you use it or not. Terminating early refunds nothing.

That makes timeout the single most expensive field in this MCP:

A 24h H100 sandbox costs $216 upfront, non-refundable, even if your job finishes in a minute.

So: ask for the time you need, not a safe-looking ceiling. Need 20 minutes of H100? timeout: 1200 is $3.00, not $216. Keep timeout ≤ 300, cpu ≤ 1 and memory ≤ 1024 and you stay on the flat rate.

How to Call from MCP

// 1. Create — timeout: 300 keeps you on the FLAT rate ($0.0110, or $0.2010 with A100).
//    Anything above 300, or above 1 CPU or 1024 MB, bills hourly for the full requested lifetime, no refund.
blockrun_modal({ path: "sandbox/create", body: {
  image: "python:3.11",
  gpu: "A100",
  timeout: 300,
  setup_commands: ["pip install torch transformers"]
}})
// returns { sandbox_id, ... }

// 2. Exec
blockrun_modal({ path: "sandbox/exec", body: {
  sandbox_id: "sb_abc...",
  command: ["python", "-c", "import torch; print(torch.cuda.get_device_name(0))"]
}})

// 3. Terminate
blockrun_modal({ path: "sandbox/terminate", body: { sandbox_id: "sb_abc..." } })

Endpoint Catalog

sandbox/create pricing is bimodal

timeout ≤ 300s, cpu ≤ 1 and memory ≤ 1024 — flat rate, charged once:

Anything longer or larger — per-hour × the full requested lifetime, never below the flat rate, upfront, no refund:

Hours are exact, not rounded up — timeout: 1800 on A100 is 0.5h = $3.5010. Every figure above includes the $0.001 flat transaction fee. Max timeout is 86400 (24h).

Field Reference

Worked Examples

1. Quick Python eval

const { structuredContent: sb } = await blockrun_modal({ path: "sandbox/create", body: {} })
await blockrun_modal({ path: "sandbox/exec", body: {
  sandbox_id: sb.sandbox_id,
  command: ["python", "-c", "import numpy; print(numpy.__version__)"]
}})
await blockrun_modal({ path: "sandbox/terminate", body: { sandbox_id: sb.sandbox_id } })

Cost: $0.0150 — create $0.0110 + exec $0.0020 + terminate $0.0020. Every call carries the $0.001 transaction fee, so three calls pay it three times; batch your work into one exec rather than several.

2. GPU inference, A100, with deps pre-installed

blockrun_modal({ path: "sandbox/create", body: {
  gpu: "A100",
  timeout: 1200,
  memory: 16384
}})

Then sandbox/exec with your inference command.

Cost: $1.3383 — create $1.3343 + exec $0.0020 + terminate $0.0020. timeout: 1200 is above the 300s flat tier, so the A100 bills hourly for the full 20 minutes you asked for: $4.00/h × (1200/3600) = $1.3333, + the $0.001 fee. It is charged upfront and never refunded — it does NOT auto-evict when idle, and terminating after 30 seconds still costs the full $1.3343. Ask for the time you actually need.

3. Test untrusted code Claude generated

blockrun_modal({ path: "sandbox/exec", body: {
  sandbox_id,
  command: ["bash", "-c", "<the generated script>"],
  timeout: 60
}})

Output is captured. No risk to your local machine.

When NOT to Use Modal

• Normal repo edits / dev work — use local tools, Modal adds latency and cost
• Long-running services — sandboxes are ephemeral, not server hosts
• Anything you'd run hundreds of times per minute — payment overhead dominates at high QPS

Notes

• sandbox_id is returned by create and required by every other endpoint
• exec is sync — blocks until command finishes or hits its timeout
• terminate is cheap; call it to free the sandbox even if timeout would expire shortly
• The free-tier nvidia/* LLM models in blockrun_chat are different infrastructure — Modal is for your arbitrary code

Reference

• Endpoints: POST /v1/modal/sandbox/{create,exec,status,terminate}
• Upstream: Modal