MPPtempoquality 0.62

Async sandboxed code execution in 60+ languages via Judge0, paid per call on Tempo L2.

Price
$0.006 / call
Protocol
mpp
Verified
no
Endpoint
https://judge0.mpp.paywithlocus.com/judge0/submit-code

What it does

This MPP endpoint wraps the Judge0 code-execution engine behind a Locus MPP payment gateway. The `/judge0/submit-code` path accepts source code and a language ID, queues execution asynchronously in a sandboxed environment, and returns a submission token. You then poll `/judge0/get-submission` with that token to retrieve results (stdout, stderr, execution time, memory usage, status). Over 60 programming languages are supported, including Python 3 (language_id 71), JavaScript (63), C++ (54), and Java (62). Optional parameters let you set CPU/wall-clock time limits, memory and stack limits, compiler flags, command-line arguments, stdin, expected output for auto-grading, stderr redirection, and network access.

Payment is settled via the Tempo method (pathUSD on Tempo L2). The submit-code and execute-code endpoints cost 6,000 base units per call. Assuming the currency uses 6 decimals (consistent with pathUSD), that works out to $0.006 per submission. The get-submission, list-languages, and list-statuses helper endpoints cost 5,000 base units ($0.005) each. All endpoints use POST and require the MPP payment flow — the probe returned 404 on HEAD/GET, which is expected since these are POST-only routes.

The service is part of the broader Locus MPP ecosystem. Additional documentation is referenced at https://beta.paywithlocus.com/mpp/judge0.md and the upstream Judge0 docs at https://judge0.com. No response schema is documented in the OpenAPI spec, so consumers should expect Judge0-standard response shapes (token on submit, full submission object on get).

Capabilities

async-code-executionsandboxed-runtime60-plus-languagescompiler-flagsstdin-supportexpected-output-matchingcpu-time-limitmemory-limitnetwork-togglempp-tempo-payment

Use cases

  • —AI agents that need to run generated code and verify output
  • —Automated grading systems comparing program output against expected results
  • —Interactive coding platforms requiring sandboxed multi-language execution
  • —CI/CD pipelines that need lightweight, pay-per-use code compilation and testing
  • —LLM tool-use workflows where the model writes and tests code iteratively

Fit

Best for

  • —Agents needing on-demand code execution without managing infrastructure
  • —Pay-per-call usage patterns where a subscription to Judge0 is overkill
  • —Multi-language support (Python, JS, C++, Java, and 60+ others)
  • —Async submission workflows where you submit then poll for results

Not for

  • —Long-running or GPU-intensive workloads (sandbox limits apply)
  • —High-throughput batch execution where a self-hosted Judge0 instance would be cheaper
  • —Workflows requiring persistent filesystem or database state between calls

Quick start

curl -X POST https://judge0.mpp.paywithlocus.com/judge0/submit-code \
  -H "Content-Type: application/json" \
  -d '{"source_code": "print(\"Hello, World!\")", "language_id": "71"}'

Example

Request

{
  "stdin": "",
  "language_id": "71",
  "source_code": "print('Hello, World!')",
  "memory_limit": 128000,
  "cpu_time_limit": 5,
  "redirect_stderr_to_stdout": true
}

Response

{
  "token": "a1b2c3d4-e5f6-7890-abcd-ef1234567890"
}

Endpoint

URLhttps://judge0.mpp.paywithlocus.com/judge0/submit-code
Transporthttp
Protocolmpp
CurrencypathUSD

Quality

0.62/ 1.00

The OpenAPI spec provides a clear request schema, pricing info, and language examples for all five endpoints. However, no response schema is documented, the probe returned 404 because it used HEAD/GET on a POST-only route (not a sign of failure but liveness is unconfirmed via 402 challenge), and crawled pages returned no useful documentation. The example response is inferred from Judge0's standard behavior, not directly sourced.

Warnings

  • —Probe did not capture a 402 MPP challenge because HEAD/GET were used on a POST-only endpoint; liveness not fully confirmed
  • —No response schema in the OpenAPI spec — response shape is inferred from upstream Judge0 conventions
  • —Price assumes pathUSD with 6 decimals; the currency address 0x20c0… is not independently verified
  • —External docs at https://beta.paywithlocus.com/mpp/judge0.md were not crawled

Citations

Provenance

Indexed frommpp_dev
Enriched2026-04-19 17:39:10Z · anthropic/claude-opus-4.6 · v2
First seen2026-04-18
Last seen2026-04-22

Agent access

JSONhttps://clawmart.sh/api/listings/vUTAtj