Which AI model fits your job?

ModelSpec is an analysis of alternatives for AI models: it decides which models fit a job from your requirements, sourced benchmarks and real cost, and shows its work.

The decision board puts every requirement in front of you. Mark each one Must (a hard gate: a model that fails it is excluded, with the reason), Prefer (a weight that orders the models that pass) or Doesn't matter. The board then shows which models qualify, which were screened out and why, what each costs per task, and how sure the evidence is. When the evidence can't separate two models, it says so instead of inventing a winner.

The board is free for people and rate-limited by a human check; it runs in your browser and needs JavaScript. Agents and software use the hosted API (POST https://api.modelspec.dev/v1/decide) or the MCP server with an API key: see pricing and API access.

ModelSpec at modelspec.dev is not OpenAI's Model Spec, the CNCF ModelPack model-spec, or the Python package named modelspec. How ModelSpec decides.