Product / UX / Telemetry & observability

LangWatch

Open-source platform for LLM evaluations and AI agent testing with production monitoring capabilities. Combines observability traces with automated quality checks and guardrails. Provides dashboards for cost, latency, and quality metrics. Positioned at the intersection of evaluation and production observability. 3.3K GitHub stars.

langwatch/langwatch monorepo active June 2026 (skills component v0.6.0, 3 Jun 2026). Confirmed live on GitHub.

Openness

4 high confidence

4.0

license: Apache-2.0(core)+MIT(SDKs)
source: public
core-gated: gated

Explicit open-core: Apache-2.0 core + MIT SDKs, with an /ee enterprise directory (SCIM, audit logs, billing) commercially licensed, same pattern as the Langfuse O4 anchor.

https://github.com/langwatch/langwatch recorded 2026-06-04
Apache-2.0 core, MIT SDKs, langwatch/ee/ enterprise modules requiring commercial license

Adoption

2 low confidence

2.0

No verified download/active-user figure located this run; ~3.3k GitHub stars is the only honest signal, so capped at level 2 via stars fallback (cannot exceed 3 on stars regardless).

https://github.com/langwatch/langwatch recorded 2026-06-04
~3.3k GitHub stars; no disclosed usage numbers

Capability

4 high confidence

4.0

Full feature matrix across all six recipe dimensions plus distinctive agent-simulation/scenario testing and a built-in gateway; on par with Langfuse C4. Held at 4 (no frontier-definer claim over the anchors).

https://github.com/langwatch/langwatch recorded 2026-06-04
feature list: tracing, evals, simulations, prompt mgmt, datasets/annotation, AI Gateway, OTel

Unchanged since 2026-07-30 (last edited, not re-checked)