EVERYTHING XIAOMI
Get listed
Inspect AI logo

Inspect AI

Inspect AI is a free, open-source framework for evaluating large language models across coding, agent tasks, reasoning, knowledge, behavior and multimodal un…

Place
#18 of 29
Score
5.8
Inspect AI's website

What Inspect AI is

Inspect AI is a free, open-source framework for evaluating large language models across coding, agent tasks, reasoning, knowledge, behavior and multimodal understanding. Evaluations bring together datasets, solvers and scorers, and the framework includes more than 200 pre-built evaluations. Inspect View provides a web interface for monitoring and visualizing results, while a VS Code extension supports log browsing, task execution and debugging. Inspect has built-in support for more than 20 model providers, plus local inference with Hugging Face, vLLM and SGLang. Its tools include web search, code execution, text editing, computer use and web browsing, and it supports MCP tools. Teams can integrate agents made with OpenAI Agents SDK, Pydantic AI and LangChain, as well as command-line agents such as Claude Code, Codex CLI and Gemini CLI. Untrusted model code can run in environments including Docker, Kubernetes, Modal, Proxmox and Vagrant. Inspect is self-hosted. Installation uses pip, and developing or running evaluations also requires access to a model.

Good

  • Includes more than 200 pre-built evaluations.
  • Supports more than 20 model providers.
  • Offers a VS Code extension for logs and debugging.
  • Can sandbox untrusted model code.

Watch out

  • Evaluation development and runs require model access.
  • The framework is self-hosted.

Who it suits

Inspect suits developers and teams evaluating language models across different task types. It is relevant for those who want to run evaluations themselves and need integrations with model providers, agents or sandbox environments.

Verdict

Inspect AI offers a broad framework for building and running model evaluations, with supporting tools for viewing results and working in VS Code. Factor in model access and self-hosting when planning its use.

Plans and prices

From Inspect AI's own pricing page, checked 4 Oct 2026.

All about pricing
PlanPriceWhat it gives you
Inspect AIFreeOpen-source framework for large language model evaluations

The facts

Each one read from Inspect AI's own pages, with the date it was checked.

What it does

Agent integrations
Inspect can integrate agents built with OpenAI Agents SDK, Pydantic AI, and LangChain, as well as CLI agents such as Claude Code, Codex CLI, and Gemini CLI.checked 4 Oct 2026
Editor integration
The Inspect VS Code extension provides tools for browsing evaluation logs, running and debugging tasks, and configuring options.checked 4 Oct 2026
Evaluation results
Inspect View is a web-based tool for monitoring and visualizing evaluations.checked 4 Oct 2026
Installation requirement
The getting-started instructions install Inspect from PyPI with pip install inspect-ai and say evaluation development and runs also require access to a model.checked 4 Oct 2026
Model integrations
Inspect has built-in support for more than 20 model providers and local inference with Hugging Face, vLLM, and SGLang.checked 4 Oct 2026
Provider examples
Documented provider groups include OpenAI, Anthropic, Google, AWS Bedrock, Azure AI, Hugging Face, vLLM, Ollama, and SGLang.checked 4 Oct 2026
Purpose
Inspect is an open-source framework for evaluating large language models across coding, agentic tasks, reasoning, knowledge, behavior, and multimodal understanding.checked 4 Oct 2026
Reusable components
Evaluations combine datasets, solvers, and scorers, and can use more than 200 pre-built evaluations.checked 4 Oct 2026
Sandboxing
The framework supports running untrusted model code in Docker, Kubernetes, Modal, Proxmox, Vagrant, and other systems through an extension API.checked 4 Oct 2026
Tools and MCP
Inspect includes standard tools for web search, code execution, text editing, computer use, and web browsing, and supports MCP tools.checked 4 Oct 2026

The company

Founded
2024checked 28 Sep 2026

Alternatives to Inspect AI

All alternatives
  1. 1DeepEval logo

    DeepEval

    Free planOpen sourceWebAPILinuxmacOS
    8.3From $200 a month
  2. 2Promptfoo logo

    Promptfoo

    Free planOpen sourceWebAPILinuxmacOS
    7.8Free plan
  3. 3Braintrust logo

    Braintrust

    Free planWebAPISelf-hosted
    7.7From $249 a month
  4. 4Giskard logo

    Giskard

    Free planOpen sourceWebAPILinuxSelf-hosted
    7.6Free plan
  5. 5

    TruLens

    Free planOpen sourceAPISelf-hosted
    7.2Free plan
  6. 6Arena (formerly Chatbot Arena) logo

    Arena (formerly Chatbot Arena)

    Free planOpen sourceWebAPISelf-hosted
    7.1Free plan