Arize AI vs W&B Weave (2026)

More than 40 integrations with models, frameworks and AI tools instrument the customer's application to send traces into Arize, code evaluators run in E2B, Daytona, Vercel or Modal sandboxes, and Data Fabric syncs trace data to warehouses. These move telemetry in and data out, and no connector lets an agent take authenticated actions in outside systems.

SourceArize, arize.com homepage and pricing and the Phoenix README on github.com/Arize-airead 2026-09-21

Integrations with LLM providers, agent SDKs and harnesses such as the OpenAI Agents SDK, Claude Agent SDK and Google ADK, and frameworks such as LangChain, CrewAI and LlamaIndex let Weave trace calls, and it traces activity between MCP clients and servers, while alerts go out by Slack and email. These bring trace data in and send notifications out, and none connects an agent to a real system to take actions.

SourceWeights & Biases, docs.wandb.ai Weave index (integrations overview, agent integrations, MCP) and wandb.ai/site/pricingread 2026-09-21

Agents that Arize observes, evaluates and improves run elsewhere, and Arize has no workflows that sequence, branch or retry an agent's steps, or mix deterministic nodes with agent steps.

SourceArize, arize.com pricing and homepageread 2026-09-21

Weave traces the customer's own multi-step and multi-agent workflows, including sub-agent delegations, and automations trigger actions from monitor metrics and trace activity. Weave itself runs no workflows that sequence, branch or retry steps, or that combine deterministic nodes with agent steps.

SourceWeights & Biases, docs.wandb.ai Weave index (trace sub-agents, automations)read 2026-09-21

Online evals run on traces and spans as they arrive, custom monitors fire on production conditions, and Signal's agents surface failure patterns from production telemetry, trace root causes and generate proposed fixes, so Arize's work starts from incoming telemetry with no person initiating each run.

SourceArize, arize.com homepage and pricingread 2026-09-21

Monitors passively score production traffic, built-in signals score agent turns and tag quality and safety issues as they arrive, and automations trigger actions from monitor metrics and trace activity, with Slack and email alerts on the Pro plan. Scoring and actions start on incoming traffic without a person asking.

SourceWeights & Biases, docs.wandb.ai Weave index (monitors, signals, automations) and wandb.ai/site/pricingread 2026-09-21

RAG pipelines built elsewhere are traced and evaluated in Arize, but Arize offers no retrieval structure over the customer's own documents that grounds an agent's answers.

SourceArize, arize.com homepage and pricingread 2026-09-21

RAG applications can be evaluated with LLM judges, and Weave traces the context a customer's retrieval returns. There is no document ingestion, index, retrieval layer or knowledge API that grounds an agent in company data.

SourceWeights & Biases, docs.wandb.ai Weave index (evaluate RAG applications)read 2026-09-21

Arize stores agent trajectories and sessions for tracing and evaluation, and there is no session, conversation or long-term memory that an agent reads back as context.

SourceArize, arize.com homepage and pricingread 2026-09-21

Conversations, threads and turns of the customer's agent are recorded, and datasets, prompts and models are versioned as tracked objects. There is no session, workflow or long-term memory that an agent reads and writes.

SourceWeights & Biases, docs.wandb.ai Weave index (trace threads, track and version objects)read 2026-09-21

Signal generates review-ready fixes for a person to review instead of applying them, so changes Arize's agents propose pass through human review. There is no approval step, escalation rule or pause before an agent acts at run time, and no autonomy modes that treat low risk and high risk actions differently.

SourceArize, arize.com homepage and pricingread 2026-09-21

Guardrails run inline scorers on a user's input or a model's output in real time, before outputs reach users, and can block or modify a response when a score crosses a threshold, with built-in and custom scorers and AWS Bedrock Guardrails as examples. Every guardrail result is stored as a monitor. Annotation queues route traces to domain experts for structured feedback.

SourceWeights & Biases, docs.wandb.ai/weave/guides/evaluation/guardrails and the Weave index (annotation queues)read 2026-09-21

Arize reports SOC 2 Type II, ISO 27001, PCI DSS, HIPAA and GDPR certification, with a Trust Center, and its plans include organization- and space-level RBAC, service accounts, Enterprise SSO and audit logs.

SourceArize, arize.com homepage and pricingread 2026-09-21

Users sign in with SSO through Google, GitHub or enterprise providers such as Okta and Azure Active Directory over OIDC, are organized into teams and projects with a restricted visibility scope, get role-based access at team or project level, automate with scoped service accounts, and are managed through the SCIM API. Multi-tenant Cloud and Dedicated Cloud are both SOC 2 Type II compliant with reports on the W&B Security Portal, Dedicated Cloud is HIPAA compliant, and paid plans include audit logs and customizable data retention.

SourceWeights & Biases, docs.wandb.ai/weave/guides/platform and wandb.ai/site/pricingread 2026-09-21

Agents are traced in Arize AX on OpenTelemetry with path and graph visualizations of each agent trajectory, token, latency and cost tracking, session and multi-modal tracing, custom dashboards, metrics, monitors and views on traces, audit logs on Enterprise, retention by plan (15 days Free, 30 days Pro, custom Enterprise), and Data Fabric sync of trace data to BigQuery, Databricks or Snowflake.

SourceArize, arize.com pricing and homepageread 2026-09-21

The Agents view shows conversations, turns, LLM calls, tool calls and sub-agent delegations with their cost, and the trace view follows nested execution paths. Calls can be filtered and exported through the Python SDK, REST API or UI, PII can be redacted from traces, and paid plans include audit logs and customizable data retention.

SourceWeights & Biases, docs.wandb.ai Weave index (agents view, trace tree, query and export calls, redact PII) and wandb.ai/site/weave and pricingread 2026-09-21

Every Arize AX plan chooses a US, EU or CA data region, Enterprise adds self-hosted deployments, and open-source Phoenix runs locally, in containers with Docker and Helm, or in the customer's cloud.

SourceArize, arize.com pricing and the Phoenix README on github.com/Arize-airead 2026-09-21

Weave runs on W&B Multi-tenant Cloud in a North America region of W&B's Google Cloud account, or on W&B Dedicated Cloud on AWS, Google Cloud or Azure, with data in a dedicated ClickHouse cluster in the cloud and region of the customer's choice, IP allowlisting and private connectivity. It also runs on Self-Managed instances the customer hosts. W&B fully manages the two cloud options.

SourceWeights & Biases, docs.wandb.ai/weave/guides/platformread 2026-09-21

Ready-made agents a customer switches on include Alyx, an AI engineering agent that runs evals, debugs issues and improves the customer's agents, and Signal, whose agents find production issues and propose fixes, each working without the other, alongside managed agents on Enterprise and packaged skills that teach coding agents to trace, evaluate and debug with Phoenix. Those are role-specific agents and ready-made workflows.

SourceArize, arize.com homepage and pricing and the Phoenix README on github.com/Arize-airead 2026-09-21

W&B Skills install into a coding agent to teach it to train models, build agents and analyze experiments on the W&B platform, a packaged, ready-to-install starting point for another agent. Weave has no ready-made agents, packaged employees or deployable workflows for a customer's own work. The Claude Code, Codex and OpenClaw plugins trace sessions and are integrations, not packaged agents.

SourceWeights & Biases, docs.wandb.ai/llms.txt (W&B Skills) and the Weave index, and wandb.ai/site/weaveread 2026-09-21

The AI provider behind Arize's own agents and evaluators can be customized, with the customer's own key alongside Arize-managed models on Enterprise, and the Phoenix playground compares models and replays traced LLM calls against them. Customers control the models Arize's own AI features run on, though bringing their own key requires the Enterprise tier.

SourceArize, arize.com pricing and the Phoenix README on github.com/Arize-airead 2026-09-21

Customers can connect OpenAI-compatible inference endpoints to a project as custom runtimes, managed from the UI or the Python and TypeScript SDKs, compare models in the Playground and Evaluation Playground, and run open source models through W&B Serverless Inference. Customers choose models and bring their own endpoints.

SourceWeights & Biases, docs.wandb.ai Weave index (custom runtimes, Playground, Serverless Inference)read 2026-09-21

Arize publishes Python and TypeScript client, OpenTelemetry and evals packages, a GraphQL API, a CLI that fetches traces, datasets and experiments, and a remote MCP server at each Phoenix instance's /mcp endpoint that MCP clients use to query traces, datasets and experiments, all on the OpenInference and OpenTelemetry standards it maintains. That is a stable API and SDK for Arize's own platform, with MCP served.

SourceArize, the Phoenix README on github.com/Arize-ai and arize.com homepageread 2026-09-21

Weave offers Python and TypeScript SDKs, a service REST API with an OpenAPI description, and an OpenTelemetry endpoint that accepts traces from any pipeline without the SDK. The W&B MCP server lets an IDE or agent query W&B data and documentation, W&B Skills teach coding agents to use the platform, and the Pro plan adds CI/CD automations.

SourceWeights & Biases, docs.wandb.ai/llms.txt and the Weave index (OpenTelemetry, query and export calls), and wandb.ai/site/pricingread 2026-09-21

Testing covers online evals on traces and spans and offline evals on datasets and experiments, with trace evals of agent trajectories, session evals of multi-turn conversations, agent-as-a-judge and custom code evaluators, evaluator alignment, human annotations and labeling queues, agent experimentation in the playground, and Prompt Learning optimization. Phoenix adds open-source evals, datasets and experiments. That tests the customer's agent against datasets before production and scores quality over time.

SourceArize, arize.com pricing and the Phoenix README on github.com/Arize-airead 2026-09-21

Evaluations run applications against versioned datasets with built-in, local and custom scorers, including LLM judges, for single-turn and multi-turn agents. Evaluations are compared side by side to spot regressions and ranked on leaderboards, monitors passively score production traffic, and the Pro plan adds CI/CD automations.

SourceWeights & Biases, docs.wandb.ai Weave index (evaluations, scorers, compare evaluations, monitors) and wandb.ai/site/weave and pricingread 2026-09-21

Arize observes, evaluates and improves agents, and its agents have no browser, desktop or computer control.

SourceArize, arize.com homepageread 2026-09-21

Weave covers tracing, evaluation, monitoring, guardrails, annotation and deployment, and no agent controls a browser, desktop or computer. The CoreWeave Sandboxes listed beside Weave are isolated environments for running agents, and they execute code instead of controlling a browser, desktop or computer.

SourceWeights & Biases, docs.wandb.ai Weave index and wandb.ai/site/weave and pricing; wandb.ai/site/pricingread 2026-09-21