Skip to Content

Computalot (Beta)

Computalot is a distributed compute platform. Submit jobs to GPU/CPU workers, get structured JSON results back.

⚠️

Private beta. Running jobs requires an admin-issued API key or admin-whitelisted wallet session. Discovery endpoints are public. We’re actively building and want your feedback.

Using an AI Agent? Start Here

Tell your agent to install the Computalot skill:

Install the Computalot skill from https://computalot.com/skill.md

The skill teaches your agent how to authenticate, submit jobs, and retrieve results. This is the fastest way to get started.

Report Bugs & Request Features

We’re improving Computalot during beta. Please report issues, request features, and share ideas — no auth required:

curl -sS -X POST https://computalot.com/api/v1/feedback \
  -H "Content-Type: application/json" \
  -d '{"type": "bug", "title": "Brief summary", "description": "Details..."}'

Types: bug, feature_request, provisioning, job_type_request.

Two Ways to Use Computalot

Sealed Recipes — Platform compute primitives with typed payloads. No code upload needed. Best for evaluation, training, fuzzing, and optimization.

Your Own Code — Create a project, push your code, submit custom jobs. Best for custom scripts, models, and arbitrary workloads.

Try a sealed recipe in 3 steps, or set up a project in 5.
Getting Started
Platform compute primitives: AMM evaluation, packing, LightGBM, fuzzing.
Sealed Recipes
Bring your own code. Create projects, push tarballs, manage setup.
Projects
Submit work, fan out across inputs, track progress.
Jobs
Give your AI agent access with a skill or API reference.
Agent Integration
Install computalot and start submitting jobs from Python.
Python SDK
Structured results, file uploads, and streaming.
Results & Artifacts
Fan-out, sweeps, pipelines, GPU training patterns.
Workflows

Good for

  • Evaluating strategies, candidates, or configurations against compute primitives
  • Parallel evaluation across prompts, models, agents, or configs
  • GPU training with progress streaming and artifact storage
  • Parameter sweeps, benchmarks, and simulation batches
  • Smart contract fuzzing with Echidna
  • Tabular ML training with LightGBM

Reference

Last updated on
Getting Started