Axomaly vs PyOD

Axomaly is a cloud service that learns patterns in SaaS product and business metrics and flags meaningful changes without fixed thresholds. It accepts timestamped time-series values, with optional metadata, for business, operational, and application metrics. Before creation, Smart Metric Creation prepares a detection configuration and baseline requirements for review. Values are submitted through a REST API, one timestamped value per POST request. Detection begins once the required history is available; the amount depends on metric cadence and detector configuration. Results include an anomaly decision, confidence, expected range, direction of change, and detector evidence, which can feed existing workflows alongside logs, traces, dashboards, and alerting. The alpha starts with one metric and includes guided onboarding and direct support. It is free during alpha with no credit card required. During that period, data and results are hosted in the EU, isolated by organisation, encrypted in transit using HTTPS/TLS, and kept for up to 12 months before rolling deletion.

PyOD is a free Python library for anomaly detection, with documented detectors for tabular, time-series, graph, text, image, and audio data. Its documentation lists 61 detectors available through a common API. Users can work with the classic detector API or ADEngine, which profiles data, selects benchmark-backed detectors, runs multiple detectors in parallel, computes consensus scores, and reports diagnostics. ADEngine also works as a standalone Python API without an LLM. PyOD includes an agentic investigation workflow and activation paths for Claude Code, Codex, and MCP-compatible agents, including an od-expert skill and optional MCP server. Installation is available through pip, conda-forge, or source, and requires Python 3.9 or higher. Optional pip extras enable capabilities including PyTorch, graph detectors, embeddings, audio, and MCP support. The project serves academic research and commercial products. Its model persistence guidance warns that pickle and joblib can execute arbitrary Python code during deserialization, requiring callers to pass trusted=True when loading artifacts. ADEngine describes its quality verdict as a heuristic rather than a guarantee and recommends validation against held-out labels or domain review.