Product / UX / Telemetry & observability

Galileo

AI observability and reliability platform that combines offline evals, online production guardrails, and tracing, powered by their proprietary Luna-2 small judge models that detect hallucinations, tool errors, and policy violations at claimed 97% lower cost than LLM-as-judge. Positions as 'don't just monitor AI failures, stop them' and is consistently ranked in top-5 LLM observability platforms (Maxim, Firecrawl roundups). Series B-funded, with enterprise customers across financial services and healthcare.

Galileo AI observability & eval-engineering platform; verified live June 2026 at galileo.ai. Proprietary SaaS / VPC / on-prem; Luna-2 small eval models. Not open source.

Openness

1 high confidence

1.0

license: proprietary
source: closed
deploy: SaaS/VPC/on-prem(managed)

Proprietary platform; recipe lists Galileo explicitly as a closed example. No OSI-licensed core product.

https://galileo.ai/ recorded 2026-06-04
proprietary platform; SaaS/VPC/on-prem deploy, no open source core product

Adoption

3 low confidence

3.0

Vendor states 'trusted by leading enterprises' but discloses no named customers or quantitative user/usage figures on the homepage. Level 3 reflects established commercial traction without a verified count; treat as directional.

https://galileo.ai/ recorded 2026-06-04
'Trusted by leading enterprises' marketing claim, no quantitative figures

Capability

4 medium confidence

4.0

Strong feature coverage across the recipe matrix; the offline-eval-to-production-guardrail loop and purpose-built Luna eval models are differentiators. OTel support not confirmed on the homepage. Capped at 4 (MLflow/Langfuse anchors hold C4); not a frontier-definer over the OSS anchors.

https://galileo.ai/ recorded 2026-06-04
evals, guardrails, tracing, cost optimization (Luna), data capture features

Unchanged since 2026-06-24 (last edited, not re-checked)