Humanloop

Humanloop is a web-based collaboration platform for prompt engineering, evaluation, and deployment of LLM features.

Its edge is the team workflow around prompts: it supports multiple LLM providers including OpenAI and Anthropic Claude models, gives you evaluation tools to score and compare outputs, and adds prompt versioning and A/B testing in a shared workspace.

It fits teams building LLM features who want to version and test prompts together, and it supports bringing your own key. It's a weaker fit if you want a CLI or IDE extension — it's web-only — if you can't send data to third-party LLM providers via their APIs, or if you need a free option; the listing indicates a paid tier with none noted.

Among LLM-ops tools, Humanloop leans toward collaborative prompt evaluation and deployment rather than raw model hosting or inference.