Rigmark

Rigmark is an open-source benchmarking tool for local AI that measures performance using agent-like code, prose, and structured output workloads.

RigMark provides a reproducible serving benchmark for OpenAI-compatible local AI appliances. It simulates the way coding agents interact with models by testing prose generation, code creation, and structured output. The tool generates a structured JSON receipt and a shareable terminal card that details the benchmark revision, prefill timings, and concurrent load results. Users execute the benchmark via a CLI, targeting any OpenAI-compatible endpoint. It includes mechanisms for cache isolation and verified output gates to ensure measurements are consistent across different model serving stacks.

Rigmark at a glance

Best for: Software engineer
Runs on: CLI
Pricing: Open source
Released: October 2026
Category: AI coding tool
Website: github.com/alexellis/rigmark

Key features

Measures performance using agent-shaped prose and code
Validates structured JSON output with specific ceilings
Tests concurrent request behavior with custom load
Generates content-hashed JSON receipts for auditability
Produces screenshot-ready terminal summary cards
Runs untrusted generated code in isolated Docker containers
Supports cache-busting and immediate-replay prefill tests

Use it for

Comparing performance between different local AI model serving stacks
Sharing reproducible benchmark results for local hardware recipes
Testing throughput under concurrent request load
Validating output consistency across different model versions

Companions

Tools that articles about Rigmark recommend using with it, each with its job.

Context7 — Provides up-to-date documentation for AI agents and code editors.
Ollama — Run local models at zero marginal cost, with code kept on your machine.
OpenRouter — Access over 500 models and 80 providers through one unified API.
Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps
Oh My OpenCode Slim — Lightweight multi-agent orchestration and background tasks.

Alternatives to Rigmark

Self Bench — Evaluates coding agents on internal repository tasks instead of generic public datasets.
Strata — Runs large AI models locally on consumer gaming PCs without server reliance.
pstack — Improves AI coding agent outcomes with rigorous workflows.
jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
GPUStack — Unifies AI model serving and GPU compute management across heterogeneous hardware.
AI Deploy Bench — Measures AI agent success, time, tokens, cost, and human steps for cloud deployments.
OpenCode — Works with 75+ model providers, local models, and existing subscriptions
OpenViking — Unifies agent memory, knowledge RAG, and skills into a single directory structure.
Daytona — It provisions sandboxes in under 90 milliseconds with full security isolation.
Flyte — Resumes failed tasks automatically from the exact point of interruption.
RigCheck — Calculates exact model memory limits and file requirements based on developer specifications.
BenchGen — It enables stress-testing AI agents in safe, simulated operational environments.

All Rigmark alternatives