Technology · Agentic → Observability · Open Core · wiki:deep
Opik (catalog folder comet-optik) is Comet’s open-source LLM observability and evaluation platform: agent/LLM tracing, datasets and experiments, LLM-as-a-judge metrics, prompt/agent optimization, production dashboards, online evaluation rules, optional guardrails, and PyTest CI hooks. Apache-2.0; free to self-host the full platform, with Comet-hosted options. Folder name keeps the landscape spelling “Optik”; the product/repo is Opik.
Same cell job as Langfuse/Phoenix: see and score agent behavior. Prefer Opik when Comet MLOps affinity or self-host Apache-2.0 + high-scale production monitoring is the draw. Prefer Langfuse for ClickHouse-backed OSS LLM engineering; Phoenix for OpenInference/Arize; Helicone for gateway-first logging.
SDK/integrations log traces and spans; UI supports annotation, playground, datasets/experiments, and online eval rules. Production paths emphasize high-volume ingest and dashboards.
Related: arize-phoenix · helicone · langfuse · topics/09-self-monitoring
Scroll inside the canvas to pan
Catalog backlinks — what points here (wiki “what links here”).
Claims below are backed by science sources on disk.
Product / tracing role
comet-ml/opik · MODERATE
“Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.”
Docs / lifecycle
Opik documentation · MODERATE
“Langfuse is an open-source LLM engineering platform that helps teams collaboratively debug, analyze, and iterate on their LLM applications.”
Contrast vs Langfuse
Langfuse contrast · MODERATE
“Langfuse is an open source LLM engineering platform. It helps teams collaboratively develop, monitor, evaluate, and debug AI applications.”
Features and peers linked from the catalog map — not a second product surface.
No DM vendor crosswalk edges yet.
Primary repo github.com/comet-ml/opik · Open Core
technologies/comet-optik/README.md
9 tags · 14 out · 15 in · 3 artifacts · 1 gaps · 31 corpus docs
map_edge · 8
tech_features · 1
tech_quote · 8
tech_readme · 1
tech_science_source · 3
tech_section · 10