[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"glossary-llmops::en":3,"gloss-cluster-llmops::en":23,"gloss-next-llmops::en":9},{"slug":4,"category":5,"name":6,"definition":7,"meta_desc":8,"faq":9,"schema_markup":9,"related":10},"llmops","mlops","LLMOps","LLMOps is the operational discipline for running LLM-powered features in production — the LLM-era counterpart to MLOps. Where MLOps centers on training pipelines and model registries, LLMOps centers on what teams actually change day to day: prompts, retrieval configuration, model choice, and guardrails. Its core practices are versioning prompts like code, running eval suites before every change, tracing each request end to end (prompt, retrieved context, output, tokens, cost), monitoring quality drift after provider model updates, and enforcing per-feature cost budgets. The failure mode it prevents is familiar to anyone who shipped an AI feature: someone edits a prompt in a dashboard, quality quietly regresses, and nobody notices until customers complain. For a SaaS team, a minimal LLMOps stack is a prompt registry, a golden dataset with automated evals in CI, and request-level tracing — enough to make model and prompt changes as safe as code deploys.","LLMOps is the practice of running LLM features in production: prompt versioning, evals, tracing, cost budgets, and drift monitoring.",null,[11,14,17,20],{"slug":12,"name":13},"eval-harness","Eval Harness",{"slug":15,"name":16},"model-registry","Model Registry",{"slug":18,"name":19},"observability","Observability",{"slug":21,"name":22},"prompt-versioning","Prompt Versioning",[24,28,32,35,38,42,45,48,51,54,57,58],{"slug":25,"category":5,"name":26,"updated_at":27},"annotation-guidelines","Annotation Guidelines","2026-08-24T03:30:02+00:00",{"slug":29,"category":5,"name":30,"updated_at":31},"baseline-model","Baseline Model","2026-08-24T02:46:38+00:00",{"slug":33,"category":5,"name":34,"updated_at":31},"batch-inference","Batch Inference",{"slug":36,"category":5,"name":37,"updated_at":31},"canary-prompt","Canary Prompt",{"slug":39,"category":5,"name":40,"updated_at":41},"champion-challenger","Champion-Challenger (A\u002FB Model Testing)","2026-08-24T02:46:37+00:00",{"slug":43,"category":5,"name":44,"updated_at":31},"class-imbalance","Class Imbalance",{"slug":46,"category":5,"name":47,"updated_at":31},"continuous-batching","Continuous Batching",{"slug":49,"category":5,"name":50,"updated_at":31},"cross-validation","Cross-Validation",{"slug":52,"category":5,"name":53,"updated_at":31},"data-labeling","Data Labeling",{"slug":55,"category":5,"name":56,"updated_at":41},"drift-detection","Drift Detection",{"slug":12,"category":5,"name":13,"updated_at":41},{"slug":59,"category":5,"name":60,"updated_at":41},"experiment-tracking","Experiment Tracking"]