Galileo is an evaluation and observability platform for generative AI applications and agents, oriented toward testing, monitoring, and guardrailing model behavior in production. It provides built-in and custom evaluation metrics such as context adherence, correctness, toxicity, and PII detection, with tracing of application runs. A free tier includes custom evals and metrics; the Pro plan is priced monthly with annual billing, and enterprise deployments are quote-based. The platform is proprietary and aimed at product and ML teams standardizing LLM quality checks.
evalsllm-tracingguardrails
Overview
- Founded
- 2021
- Pricing tier
- High
- Startup-friendly
- Yes
- Enterprise-ready
- Yes
LLM observability
- Request tracing
- Yes
- Quality evals
- Yes
- Prompt management
- No
- LLM gateway
- No
- Open-source
- No
- Self-hostable
- No
- Free tier
- Yes: Free tier with custom evals and metrics