# CoreWeave Docs: English Forge > Learn how to deploy, manage, and observe your AI training and inference workloads on CoreWeave. ## Forge - [English / Forge / Weights & Biases (658 pages)](https://docs.coreweave.com/_llms/en/forge/weights-and-biases.md): Documentation for English / Forge / Weights & Biases. - [English / Forge / Post-Training (129 pages)](https://docs.coreweave.com/_llms/en/forge/post-training.md): Documentation for English / Forge / Post-Training. ### Forge Home - [Forge Home](https://docs.coreweave.com/forge-home.md): Overview of CoreWeave Forge products for developing, evaluating, and serving AI models and agents - [What moved in Forge](https://docs.coreweave.com/forge-whats-moved.md): Where to find your projects, settings, and account in Forge if you already use Weights & Biases - [Where the Weights & Biases docs moved](https://docs.coreweave.com/forge-docs-whats-moved.md): A map from the previous Weights & Biases documentation to its current location - [Platform](https://docs.coreweave.com/redirects/forge/platform.md): Forge platform documentation: configuration, security and access, monitoring and usage, deployment options, and Launch #### Forge - [ARIA](https://docs.coreweave.com/redirects/forge/aria.md): AI research agent that analyzes experiments, explains results, and builds reports - [Weights & Biases](https://docs.coreweave.com/redirects/forge/wandb.md): W&B Models for experiment tracking, sweeps, artifacts, and reports; W&B Weave for tracing and evaluating LLM apps - [Post-Training](https://docs.coreweave.com/redirects/forge/training.md): Post-train and fine-tune LLMs with Serverless RL and SFT, and distill specialized models from real traffic - [Agent Lens](https://docs.coreweave.com/redirects/forge/agent-lens.md): Overview of Agent Lens on CoreWeave - [Serverless Inference](https://docs.coreweave.com/redirects/forge/inference.md): Open-source foundation models through an OpenAI-compatible API - [Serverless Sandbox](https://docs.coreweave.com/redirects/forge/sandboxes.md): On-demand, isolated compute environments you create and discard from Python - [Notebooks](https://docs.coreweave.com/redirects/forge/notebooks.md): Overview of Notebooks on CoreWeave - [Registry](https://docs.coreweave.com/redirects/forge/registry.md): Centralized, versioned hub for models and datasets with lineage and deployment handoff - [HiveMind](https://docs.coreweave.com/products/wandb/hivemind.md): A shared dashboard for AI coding sessions. Track activity, spend, and outcomes across Claude Code, Cursor, Codex, Gemini CLI, and more with leaderboards, team views, and efficiency insights. ### Platform #### Forge platform ##### Settings - [Settings](https://docs.coreweave.com/products/wandb/platform/app/settings-page.md): Use the Weights & Biases Settings page to customize your individual user profile or team settings. - [Manage user settings](https://docs.coreweave.com/products/wandb/platform/app/settings-page/user-settings.md): Manage your profile information, account defaults, alerts, participation in public preview features, account activation, and create teams in your user settings. - [Manage billing settings](https://docs.coreweave.com/products/wandb/platform/app/settings-page/billing-settings.md): View plan details, monitor usage, and configure usage and spending alerts for your Weights & Biases organization. - [Manage email settings](https://docs.coreweave.com/products/wandb/platform/app/settings-page/emails.md): Add, verify, and delete the email addresses on your Forge account, and choose your primary email. - [Team settings](https://docs.coreweave.com/products/wandb/platform/app/settings-page/teams.md): Collaborate with your colleagues, share results, and track all the experiments across your team. Manage team settings including members, alerts, and privacy. - [Configure privacy settings](https://docs.coreweave.com/products/wandb/platform/hosting/privacy-settings.md): Enforce privacy settings for team visibility, project access, and code saving defaults in Weights & Biases. - [Manage secrets](https://docs.coreweave.com/products/wandb/platform/secrets.md): Overview of Weights & Biases secrets, how they work, and how to get started using them. - [Manage storage](https://docs.coreweave.com/products/wandb/platform/app/settings-page/storage.md): Manage Weights & Biases data storage consumption using reference artifacts, external buckets, and TTL deletion policies. ##### Identity and access management (IAM) - [Identity and access management (IAM)](https://docs.coreweave.com/products/wandb/platform/hosting/iam/org_team_struct.md): Learn how IAM scopes (organizations, teams, and projects) map to your company. - [Automate user and team management](https://docs.coreweave.com/products/wandb/platform/hosting/iam/automate_iam.md): Automate your deployment's user and team management at scale using the SCIM API and the Python SDK API. - [Manage users, groups, and roles with SCIM](https://docs.coreweave.com/products/wandb/platform/hosting/iam/scim.md): Use the SCIM API to automate provisioning of users, groups, and custom roles in an organization. - [Advanced IAM configuration](https://docs.coreweave.com/products/wandb/platform/hosting/iam/advanced_env_vars.md): Configure advanced Weights & Biases IAM settings with environment variables for SSO, session length, OIDC, and LDAP. ###### Authentication - [Configure SSO with LDAP](https://docs.coreweave.com/products/wandb/platform/hosting/iam/ldap.md): Configure LDAP-based SSO authentication for W&B Server including connection parameters and environment variables. - [Configure SSO with OIDC](https://docs.coreweave.com/products/wandb/platform/hosting/iam/sso.md): Configure Weights & Biases SSO with OpenID Connect providers such as Okta, Azure AD, and AWS Cognito. - [Use federated identities with SDK](https://docs.coreweave.com/products/wandb/platform/hosting/iam/identity_federation.md): Use identity federation with JSON Web Tokens (JWTs) to authenticate with the W&B SDK and CLI without API keys. - [Manage API keys](https://docs.coreweave.com/products/wandb/platform/hosting/iam/api-keys.md): Find, create, and delete Forge API keys, and restrict how members of your organization create and use them. - [Use service accounts to automate workflows](https://docs.coreweave.com/products/wandb/platform/hosting/iam/service-accounts.md): Manage automated or non-interactive workflows using org and team scoped service accounts ###### Access management - [Access management](https://docs.coreweave.com/products/wandb/platform/hosting/iam/access-management-intro.md): Manage users, teams, roles, and project visibility as an organization or team admin in Weights & Biases. - [Manage your organization](https://docs.coreweave.com/products/wandb/platform/hosting/iam/access-management/manage-organization.md): Manage users, teams, roles, seats, and billing within a Forge organization as an organization admin. - [Manage access control for projects](https://docs.coreweave.com/products/wandb/platform/hosting/iam/access-management/restricted-projects.md): Manage project access using visibility scopes and project-level roles ##### Data security - [Bring your own bucket (BYOB)](https://docs.coreweave.com/products/wandb/platform/hosting/data-security/secure-storage-connector.md): Store W&B artifacts and data in your own cloud storage buckets using the Bring Your Own Bucket (BYOB) feature. - [Manage bucket storage and costs](https://docs.coreweave.com/products/wandb/platform/hosting/managing-bucket-storage.md): Learn how Weights & Biases uses object storage, how deletion affects bucket bytes, and how to reduce usage and costs. - [Access BYOB using pre-signed URLs](https://docs.coreweave.com/products/wandb/platform/hosting/data-security/presigned-urls.md): Learn how Weights & Biases uses pre-signed URLs for blob storage, team-level access control, and audit logging. - [Configure IP allowlisting for Dedicated Cloud](https://docs.coreweave.com/products/wandb/platform/hosting/data-security/ip-allowlisting.md): Restrict access to a W&B Dedicated Cloud instance to authorized IP addresses using IP allowlisting. - [Configure private connectivity to Dedicated Cloud](https://docs.coreweave.com/products/wandb/platform/hosting/data-security/private-connectivity.md): Connect to a W&B Dedicated Cloud instance over a private network using AWS PrivateLink, GCP, or Azure Private Link. - [Data encryption in Dedicated Cloud](https://docs.coreweave.com/products/wandb/platform/hosting/data-security/data-encryption.md): Learn how Weights & Biases encrypts Dedicated Cloud data with cloud-native keys and handles customer-managed keys. ##### Security & Admin - [Security & Access](https://docs.coreweave.com/redirects/forge/security.md): Choose between CoreWeave IAM and Weights & Biases access management, and find the platform security model ##### Billing & Pricing - [Infrastructure and Forge billing](https://docs.coreweave.com/redirects/forge/billing.md): Choose between CoreWeave Infrastructure billing and Forge billing, and find where each invoice lives ##### Monitoring & Usage - [Track user activity with audit logs](https://docs.coreweave.com/products/wandb/platform/hosting/monitoring-usage/audit-logging.md): Fetch and analyze Weights & Biases audit logs across deployment types, including the log schema and tracked actions. - [Configure Slack alerts](https://docs.coreweave.com/products/wandb/platform/hosting/monitoring-usage/slack-alerts.md): Create and configure a Slack application to receive W&B Server alerts, notifications, and monitoring updates. - [View organization activity](https://docs.coreweave.com/products/wandb/platform/hosting/monitoring-usage/org_dashboard.md): View user status, activity, and usage trends in the organization dashboard across deployment types. - [W&B Mobile App (iOS)](https://docs.coreweave.com/products/wandb/platform/hosting/monitoring-usage/mobile-app.md): Explore Weights & Biases projects, training runs, console logs, line plots, and system metrics on your iPhone or iPad. ##### Deployment Options - [Deployment options overview](https://docs.coreweave.com/products/wandb/platform/hosting.md): Compare Weights & Biases deployment options: Forge, Dedicated Cloud, and Self-Managed installations. - [Deployment options](https://docs.coreweave.com/products/wandb/platform/hosting/hosting-options.md): Choose a Weights & Biases deployment type: Forge, Dedicated Cloud, or Self-Managed. - [Forge](https://docs.coreweave.com/products/wandb/platform/hosting/hosting-options/multi_tenant_cloud.md): Learn about Forge, a fully managed deployment on Google Cloud with built-in compliance and data security. ###### Dedicated Cloud - [Dedicated Cloud](https://docs.coreweave.com/products/wandb/platform/hosting/hosting-options/dedicated-cloud.md): Learn about W&B Dedicated Cloud deployment features including compliance, data security, IAM, and maintenance policies. - [Rate limits](https://docs.coreweave.com/products/wandb/platform/hosting/hosting-options/dedicated-cloud/rate-limits.md): Default rate limits on Dedicated Cloud and how to request changes - [Supported Dedicated Cloud regions](https://docs.coreweave.com/products/wandb/platform/hosting/hosting-options/dedicated-cloud/regions.md): View all supported AWS, Google Cloud, and Azure regions available for W&B Dedicated Cloud instances. - [Export data from Dedicated Cloud](https://docs.coreweave.com/products/wandb/platform/hosting/hosting-options/dedicated-cloud/export-data.md): Export runs, metrics, artifacts, and reports from a W&B Dedicated Cloud instance using the Python SDK API. ###### Self-Managed - [W&B Self-Managed deployment overview](https://docs.coreweave.com/products/wandb/platform/hosting/hosting-options/self-managed.md): Deploy W&B Self-Managed on cloud or on-premises infrastructure - [Reference architecture](https://docs.coreweave.com/products/wandb/platform/hosting/self-managed/ref-arch.md): Review self-managed Weights & Biases architecture covering Kubernetes, MySQL, object storage, and networking. - [Self-Managed infrastructure requirements](https://docs.coreweave.com/products/wandb/platform/hosting/self-managed/requirements.md): Infrastructure and software requirements for W&B Self-Managed deployments - [Rate limits](https://docs.coreweave.com/products/wandb/platform/hosting/self-managed/rate-limits.md): Optional rate limits on Self-Managed instances for stability - [Deploy Weights & Biases with Kubernetes Operator](https://docs.coreweave.com/products/wandb/platform/hosting/self-managed/operator.md): Deploy W&B Platform with Kubernetes Operator on cloud or on-premises - [Configure environment variables](https://docs.coreweave.com/products/wandb/platform/hosting/env-vars.md): Configure a self-managed W&B Server installation using environment variables for database, storage, Redis, and IAM settings. - [Deploy on Air-Gapped Kubernetes](https://docs.coreweave.com/products/wandb/platform/hosting/self-managed/on-premises-deployments/kubernetes-airgapped.md): Deploy W&B Platform in air-gapped and disconnected Kubernetes environments - [Enterprise licenses](https://docs.coreweave.com/products/wandb/platform/hosting/enterprise-licenses.md): Learn about Weights & Biases Enterprise licenses, what features they include, and how to obtain and configure them. - [Update Weights & Biases license and version](https://docs.coreweave.com/products/wandb/platform/hosting/server-upgrade-process.md): Guide for updating Weights & Biases version and license across different installation methods. - [Disable automatic updates for W&B Server](https://docs.coreweave.com/products/wandb/platform/hosting/self-managed/disable-automatic-app-version-updates.md): Learn how to disable automatic updates for W&B Server. ### ARIA - [ARIA](https://docs.coreweave.com/products/aria.md): Analyze and run experiments, explain results, and build reports with W&B's AI research agent. - [Chat with ARIA](https://docs.coreweave.com/products/aria/chat.md): Start and organize ARIA chats, queue messages, add context, grant network access, and give feedback. - [Use ARIA for autoresearch](https://docs.coreweave.com/products/aria/autoresearch.md): Learn how to use ARIA, W&B's AI Research and Iteration Agent, to analyze results and run experiments. - [Secrets in ARIA](https://docs.coreweave.com/products/aria/secrets.md): Store personal secrets, such as API tokens, and let ARIA use them as environment variables in its sandbox. - [Governance and security](https://docs.coreweave.com/products/aria/governance.md): Learn how W&B Admins can control how your organization's data is used by W&B's AI features like ARIA. ### Agent Lens #### Agent Lens - [Agent Lens](https://docs.coreweave.com/products/agent-lens.md): Overview of Agent Lens on CoreWeave - [What is Agent Lens?](https://docs.coreweave.com/products/agent-lens/what-is-agent-lens.md): Learn about CoreWeave Agent Lens and how it helps you trace, monitor, and improve LLM agents ##### Get started - [Choose an agent integration](https://docs.coreweave.com/products/agent-lens/get-started/integrations.md): Pick a built-in Agent Lens integration for your LLM provider SDK, agent SDK, or harness. - [Quickstart: Set up custom agent observability](https://docs.coreweave.com/products/agent-lens/get-started/custom-agents.md): Instrument a custom agent with Agent Lens to capture existing OTel spans. - [Send OpenTelemetry spans to Agent Lens](https://docs.coreweave.com/products/agent-lens/get-started/opentelemetry.md): Send OTLP trace data from any OpenTelemetry pipeline to the Agent Lens endpoint. No SDK integration required. #### Guides ##### Trace your agents - [Trace your agents](https://docs.coreweave.com/products/agent-lens/tracing/instrument.md): Use the Agent Lens SDK to instrument multi-turn agentic applications and view their activity in the Agent Lens. - [Set attributes on agent spans](https://docs.coreweave.com/products/agent-lens/tracing/attributes.md): Attach custom attributes to agent spans to filter and analyze activity in Agent Lens. - [Batch logging for your agent](https://docs.coreweave.com/products/agent-lens/tracing/batch-logging.md): Manually log agent traces for frameworks that have already completed the LLM call and need to record it. - [Trace sub-agents](https://docs.coreweave.com/products/agent-lens/tracing/sub-agents.md): Use Agent Lens's sub-agent span to trace sub-agent delegations and view nested agent invocations. ##### Conversations - [View agent activity](https://docs.coreweave.com/products/agent-lens/conversations/view-activity.md): Use Agent Lens to analyze your agent's conversations and performance metrics. - [Annotate conversation turns](https://docs.coreweave.com/products/agent-lens/conversations/annotate-turns.md): Add reactions and notes to conversation turns to record what your team noticed while reviewing traces. ##### Insights - [View agent insights](https://docs.coreweave.com/products/agent-lens/insights.md): Review your agent's failures and user intents by category, with metrics and example turns for each. ##### Tags - [Create a custom tag](https://docs.coreweave.com/products/agent-lens/tags/create-tag.md): Create a custom Agent Lens tag and turn on AI evaluation to label agent turns automatically. #### Integrations ##### LLM provider SDKs - [OpenAI](https://docs.coreweave.com/products/agent-lens/integrations/openai.md): Trace OpenAI SDK calls in Agent Lens with the CoreWeave Forge SDK. - [Anthropic](https://docs.coreweave.com/products/agent-lens/integrations/anthropic.md): Trace Anthropic SDK calls in Agent Lens with the CoreWeave Forge SDK. - [Google Gen AI](https://docs.coreweave.com/products/agent-lens/integrations/google-genai.md): Trace Google Gen AI SDK calls to Gemini in Agent Lens with the CoreWeave Forge SDK. ##### Agent SDKs - [Google ADK](https://docs.coreweave.com/products/agent-lens/integrations/google-adk.md): Trace an agent built with Google's Agent Development Kit (ADK) using Agent Lens. - [OpenAI Agents SDK](https://docs.coreweave.com/products/agent-lens/integrations/openai-agents-sdk.md): Trace an agent built with the OpenAI Agents SDK using Agent Lens. - [Claude Agent SDK](https://docs.coreweave.com/products/agent-lens/integrations/claude-agents-sdk.md): Trace an agent built with the Claude Agent SDK using Agent Lens. ##### Agent harnesses - [Claude Code plugin](https://docs.coreweave.com/products/agent-lens/integrations/claude-code.md): Track Claude Code sessions in Agent Lens for observability and debugging. - [Codex plugin](https://docs.coreweave.com/products/agent-lens/integrations/codex.md): Trace Codex agentic sessions, LLM calls, and tool executions in Agent Lens. - [OpenClaw plugin](https://docs.coreweave.com/products/agent-lens/integrations/openclaw.md): Track OpenClaw agent sessions in Agent Lens for observability and debugging. - [Pi extension](https://docs.coreweave.com/products/agent-lens/integrations/pi.md): Trace Pi agentic sessions, LLM calls, and tool executions in Agent Lens. #### Reference - [Agent Lens SDK reference](https://docs.coreweave.com/products/agent-lens/reference.md): Reference for the Python and TypeScript SDKs that send agent traces to Agent Lens. ##### Python SDK - [Python SDK](https://docs.coreweave.com/products/agent-lens/reference/python-sdk.md): Reference for the CoreWeave Forge Python SDK, which sends agent traces to Agent Lens. - [Setup and shutdown](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/setup.md): Start, flush, and stop tracing with init, force_flush, shutdown, and TracingSession. - [Conversation](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/conversation.md): The Conversation class groups turns under one conversation ID. - [Turn](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/turn.md): The Turn class records one user-agent exchange as an invoke_agent span. - [LLM](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/llm.md): The LLM class records one LLM API call as a chat span. - [Tool](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/tool.md): The Tool class records one tool execution as an execute_tool span. - [SubAgent](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/subagent.md): The SubAgent class records a delegated agent invocation as a nested invoke_agent span. - [Span functions](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/span-functions.md): Module-level functions that start, end, and look up the current conversation, turn, and LLM call. - [Batch logging](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/batch-logging.md): Emit complete turns and conversations after the fact with log_turn and log_conversation. - [Message types](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/message-types.md): Messages, message parts, token usage, reasoning, and media attachments recorded on spans. - [Integrations](https://docs.coreweave.com/products/agent-lens/reference/python-sdk/integrations.md): Patch the OpenAI, Anthropic, and Google Gen AI SDKs so their calls are traced automatically. ##### TypeScript SDK - [TypeScript SDK](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk.md): Reference for the Agent Lens tracing and provider integration APIs in @coreweave/forge-sdk. ###### Functions - [endConversation](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/endconversation.md): End the current Conversation. - [endLLM](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/endllm.md): End the current LLM. - [endTurn](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/endturn.md): End the current Turn. - [forceFlush](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/forceflush.md): Reference for the forceFlush function in the Agent Lens TypeScript SDK. - [getCurrentConversation](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/getcurrentconversation.md): Reference for the getCurrentConversation function in the Agent Lens TypeScript SDK. - [getCurrentLLM](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/getcurrentllm.md): Reference for the getCurrentLLM function in the Agent Lens TypeScript SDK. - [getCurrentTurn](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/getcurrentturn.md): Reference for the getCurrentTurn function in the Agent Lens TypeScript SDK. - [init](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/init.md): Reference for the init function in the Agent Lens TypeScript SDK. - [runIsolated](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/runisolated.md): Give one request or agent run its own tracing state, including across awaits. - [shutdown](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/shutdown.md): Reference for the shutdown function in the Agent Lens TypeScript SDK. - [startConversation](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/startconversation.md): Start a new Conversation and install it as the current conversation. - [startLLM](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/startllm.md): Start an LLM span as a child of the current Turn. - [startSubagent](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/startsubagent.md): Start a SubAgent span. - [startTool](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/starttool.md): Start a Tool span. - [startTurn](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/functions/startturn.md): Start a new Turn. ###### Classes - [Conversation](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/classes/conversation.md): A Conversation groups Turns under a single gen_ai.conversation.id. - [LLM](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/classes/llm.md): An LLM call. - [SubAgent](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/classes/subagent.md): Reference for the SubAgent class in the Agent Lens TypeScript SDK. - [Tool](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/classes/tool.md): A tool invocation. - [Turn](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/classes/turn.md): An agent invocation. ###### Interfaces - [ConversationInit](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/conversationinit.md): Reference for the ConversationInit interface in the Agent Lens TypeScript SDK. - [LLMInit](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/llminit.md): Reference for the LLMInit interface in the Agent Lens TypeScript SDK. - [Message](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/message.md): Reference for the Message interface in the Agent Lens TypeScript SDK. - [Reasoning](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/reasoning.md): Reference for the Reasoning interface in the Agent Lens TypeScript SDK. - [SubAgentInit](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/subagentinit.md): Reference for the SubAgentInit interface in the Agent Lens TypeScript SDK. - [ToolEndOptions](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/toolendoptions.md): Reference for the ToolEndOptions interface in the Agent Lens TypeScript SDK. - [ToolInit](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/toolinit.md): Reference for the ToolInit interface in the Agent Lens TypeScript SDK. - [TracingInitOptions](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/tracinginitoptions.md): Reference for the TracingInitOptions interface in the Agent Lens TypeScript SDK. - [TurnInit](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/turninit.md): Reference for the TurnInit interface in the Agent Lens TypeScript SDK. - [Usage](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/interfaces/usage.md): Reference for the Usage interface in the Agent Lens TypeScript SDK. ###### Type aliases - [JsonObject](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/type-aliases/jsonobject.md): An object whose properties are representable in JSON. - [JsonValue](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/type-aliases/jsonvalue.md): A value representable in JSON. - [MessagePart](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/type-aliases/messagepart.md): A message part in the GenAI semconv parts model. - [Modality](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/type-aliases/modality.md): Reference for the Modality type in the Agent Lens TypeScript SDK. - [Role](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/type-aliases/role.md): Reference for the Role type in the Agent Lens TypeScript SDK. ###### Integrations - [instrumentAnthropic](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/instrumentanthropic.md): Reference for the instrumentAnthropic function in the Agent Lens TypeScript SDK. - [instrumentGoogleGenAI](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/instrumentgooglegenai.md): Register automatic ESM and CommonJS instrumentation for @google/genai. - [instrumentOpenAI](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/instrumentopenai.md): Reference for the instrumentOpenAI function in the Agent Lens TypeScript SDK. - [patchAnthropicModule](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/patchanthropicmodule.md): Apply the Anthropic integration to an ESM module namespace. - [patchGoogleGenAIModule](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/patchgooglegenaimodule.md): Apply the Google Gen AI integration to an ESM or CommonJS namespace. - [patchOpenAIModule](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/patchopenaimodule.md): Apply the OpenAI integration to an ESM module namespace. - [wrapAnthropic](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/wrapanthropic.md): Wrap an Anthropic client explicitly. - [wrapGoogleGenAI](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/wrapgooglegenai.md): Wrap a Google Gen AI client explicitly. - [wrapOpenAI](https://docs.coreweave.com/products/agent-lens/reference/typescript-sdk/integrations/wrapopenai.md): Wrap an OpenAI client explicitly. ### Serverless Inference - [Serverless Inference](https://docs.coreweave.com/products/inference/serverless.md): Access open-source foundation models through an OpenAI-compatible API - [Available models](https://docs.coreweave.com/products/inference/serverless/models.md): Browse the foundation models available through Serverless Inference - [Model lifecycle](https://docs.coreweave.com/products/inference/serverless/lifecycle.md): Learn about Serverless Inference model lifecycle and retirement - [Try models in the Playground UI](https://docs.coreweave.com/products/inference/serverless/ui-guide.md): Access Serverless Inference models through the web interface - [Bring your own LoRA adapter](https://docs.coreweave.com/products/inference/serverless/lora.md): Bring your own custom LoRA for serving fine-tuned models on Serverless Inference. - [Usage information and limits](https://docs.coreweave.com/products/inference/serverless/usage-limits.md): Understand pricing, usage limits, and account restrictions for Serverless Inference - [Usage examples](https://docs.coreweave.com/products/inference/serverless/examples.md): Learn how to use Serverless Inference with practical code examples - [Support: Serverless Inference](https://docs.coreweave.com/products/inference/serverless/support-inference.md) #### Response Settings - [Enable JSON mode](https://docs.coreweave.com/products/inference/serverless/response-settings/json-mode.md): Configure JSON mode in Serverless Inference to get structured JSON output from model responses for easier parsing. - [View reasoning information](https://docs.coreweave.com/products/inference/serverless/response-settings/reasoning.md): How to return and view reasoning in your Serverless Inference responses - [Enable streaming responses](https://docs.coreweave.com/products/inference/serverless/response-settings/streaming.md): Enable streaming output with Serverless Inference to receive model responses incrementally. - [Prefix caching](https://docs.coreweave.com/products/inference/serverless/response-settings/prefix-caching.md): Reduce latency for repeated prompts with prefix caching, and isolate cache reuse with cache_salt when needed. - [Enable structured output](https://docs.coreweave.com/products/inference/serverless/response-settings/structured-output.md): How to configure structured output in Serverless Inference responses - [Call tools](https://docs.coreweave.com/products/inference/serverless/response-settings/tool-calling.md): Use function and tool calling with Serverless Inference to let models invoke external tools and APIs during generation. #### Tutorials - [Cline with Serverless Inference](https://docs.coreweave.com/products/inference/serverless/tutorials/integration-cline.md): Learn how to configure the Cline coding agent to use Serverless Inference. - [Creating a fine-tuned LoRA](https://docs.coreweave.com/products/inference/serverless/tutorials/creating-lora.md): Learn how to create a fine-tuned LoRA to use with Serverless Inference. #### API Reference - [Serverless Inference API reference](https://docs.coreweave.com/products/inference/serverless/api-reference.md): Explore the complete REST API reference for the Serverless Inference service, including endpoints and authentication. - [Chat completions](https://docs.coreweave.com/products/inference/serverless/api-reference/chat-completions.md): Create chat completions using the OpenAI-compatible endpoint - [List models](https://docs.coreweave.com/products/inference/serverless/api-reference/list-models.md): Retrieve a list of available Serverless Inference models and their IDs using the models endpoint. ### Serverless Sandbox #### Sandboxes - [About Serverless Sandbox](https://docs.coreweave.com/products/sandboxes/serverless.md): Run agent tools, evaluations, and training workloads in isolated serverless environments. - [Get started with serverless sandboxes](https://docs.coreweave.com/products/sandboxes/serverless/get-started.md): Run your first serverless sandbox with a W&B API key. - [Use sandboxes with W&B Dedicated Cloud](https://docs.coreweave.com/products/sandboxes/serverless/wandb-dedicated.md): Authenticate serverless sandboxes with an API key from your W&B Dedicated Cloud instance. - [Support](https://docs.coreweave.com/redirects/sandboxes/serverless-support.md) ##### Concepts - [Sandbox lifecycle](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/sandbox-lifecycle.md): Understand sandbox states, creation patterns, waiting, and shutdown behavior. - [Sync versus async patterns](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/sync-vs-async.md): Choose between synchronous and asynchronous patterns in the CoreWeave sandbox Python SDK. ##### Tutorials - [Configuration](https://docs.coreweave.com/products/sandboxes/serverless/client/tutorial/configuration.md): Configure sandbox resources, images, and environment settings. - [Run commands in a sandbox environment](https://docs.coreweave.com/products/sandboxes/serverless/client/tutorial/running-commands.md): Execute commands inside a sandbox and capture output. - [Stream real-time output from a sandbox](https://docs.coreweave.com/products/sandboxes/serverless/client/tutorial/streaming.md): Stream command output in real time from a sandbox. - [File operations](https://docs.coreweave.com/products/sandboxes/serverless/client/tutorial/files.md): Tutorial for uploading and downloading files to and from CoreWeave sandbox instances - [Multiple sandboxes](https://docs.coreweave.com/products/sandboxes/serverless/client/tutorial/multiple-sandboxes.md): Tutorial for managing multiple CoreWeave sandbox instances using sessions with shared defaults - [Remote functions](https://docs.coreweave.com/products/sandboxes/serverless/client/tutorial/remote-functions.md): Run Python functions inside sandboxes with automatic serialization. - [Cleanup](https://docs.coreweave.com/products/sandboxes/serverless/client/tutorial/cleanup.md): Terminate sandboxes and manage cleanup with context managers, explicit stops, and global atexit handlers. - [Run a computer-use agent in a sandbox](https://docs.coreweave.com/products/sandboxes/serverless/tutorials/computer-use.md): Run a Linux desktop in a serverless sandbox and let an agent operate it through screenshots and input. - [RL training](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/rl-training.md): Use sandboxes for reinforcement learning training environments. - [SWE-bench](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/swebench.md): Run SWE-bench evaluations using CoreWeave sandboxes. #### Use sandboxes ##### Integrate agents - [Run agents on Sandboxes](https://docs.coreweave.com/products/sandboxes/serverless/agents.md): Run interactive and autonomous agents with longer sessions, parallel work, and tools on CoreWeave. - [Claude Code](https://docs.coreweave.com/products/sandboxes/serverless/agents/claude-code.md): Run Claude Code in a CoreWeave sandbox and interact from your terminal, Claude apps, or claude.ai. - [Run OpenCode in a Sandbox](https://docs.coreweave.com/products/sandboxes/serverless/agents/opencode.md): Run OpenCode in a sandbox with W&B Serverless Inference or another model provider. - [Run Muse Code in a Sandbox](https://docs.coreweave.com/products/sandboxes/serverless/agents/muse-code.md): Run Muse Code CLI on a repository in a sandbox, retrieve the result, and stop compute. - [Run Pi in a Sandbox](https://docs.coreweave.com/products/sandboxes/serverless/agents/pi.md): Run Pi from an attached terminal with optional W&B Serverless Inference, then retrieve your work. - [Use Sandboxes with Claude Managed Agents](https://docs.coreweave.com/products/sandboxes/serverless/agents/claude-managed-agents.md): Connect a self-hosted sandbox worker to Claude Managed Agents and verify tool execution on CoreWeave. - [Run OpenAI Agents API on Sandboxes](https://docs.coreweave.com/products/sandboxes/serverless/agents/openai-agents-api.md): Connect OpenAI-managed agents to a CoreWeave sandbox for tool execution and shared-workspace analysis. - [Run Codex in a Sandbox](https://docs.coreweave.com/products/sandboxes/serverless/agents/codex.md): Run Codex CLI in a sandbox, connect a local CLI, or use desktop and mobile apps with a remote project. - [Run Cursor in a Sandbox](https://docs.coreweave.com/products/sandboxes/serverless/agents/cursor.md): Run Cursor CLI interactively or automate coding tasks in a CoreWeave sandbox. - [Run Devin Outposts on Sandboxes](https://docs.coreweave.com/products/sandboxes/serverless/agents/devin-outposts.md): Let Devin plan in Devin Cloud while its commands run in a CoreWeave sandbox you control. ##### Configure - [Sandbox configuration](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/sandbox-configuration.md): Configure sandbox resources, images, networking, timeouts, and storage. - [Run a GPU sandbox](https://docs.coreweave.com/products/sandboxes/serverless/gpu-sandboxes.md): Request GPUs for a serverless sandbox, which runs in a virtual machine. - [Sessions](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/sessions.md): Group and manage multiple sandboxes with sessions. - [Create a sandbox from a Compose file](https://docs.coreweave.com/products/sandboxes/serverless/create-from-file.md): Start a multi-container sandbox by uploading a Docker Compose file to the sandbox API. - [Implement a warm pool](https://docs.coreweave.com/products/sandboxes/serverless/warm-pools.md): Keep sandboxes ready for incoming work using a Python warm pool recipe. - [Run long-running sandboxes](https://docs.coreweave.com/products/sandboxes/serverless/long-running-sandboxes.md): Configure serverless sandbox lifetimes up to 30 days. - [Cleanup patterns](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/cleanup-patterns.md): Patterns for reliable sandbox cleanup and resource management. ###### Variables and secrets - [Environment variables](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/environment-variables.md): Set and manage environment variables in CWSandbox instances, including session-level defaults - [Environment variables](https://docs.coreweave.com/products/sandboxes/serverless/environment-variables.md): Environment variables CoreWeave sets inside every sandbox, and how to override them. - [Use W&B secrets](https://docs.coreweave.com/products/sandboxes/serverless/secrets.md): Inject W&B team secrets into serverless sandboxes with Python or TypeScript. ##### Run code - [Command execution](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/execution.md): Run commands in sandboxes with streaming, process control, and error handling. - [Interactive shells and TTY](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/interactive-shells.md): Use interactive TTY sessions with shell() for terminal applications and raw byte streaming. - [Remote functions](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/remote-functions.md): Execute Python functions inside sandboxes with the function decorator. - [Sandbox logging](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/logging.md): Stream container logs (stdout and stderr) from a sandbox's main process with stream_logs(). ##### Files and storage - [File operations](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/file-operations.md): Read, write, and manage files within CoreWeave sandbox instances using the CWSandbox client - [Volumes](https://docs.coreweave.com/products/sandboxes/serverless/volumes.md): Give a sandbox writable scratch space that lasts as long as the sandbox. - [File system snapshots](https://docs.coreweave.com/products/sandboxes/serverless/file-system-snapshots.md): Snapshot a sandbox's working directory to object storage, then restore or fork it into new sandboxes. ##### Networking - [Public endpoints](https://docs.coreweave.com/products/sandboxes/serverless/public-endpoints.md): Give a service inside a sandbox a public address, with platform TLS or TLS passthrough. - [Direct sandbox connections](https://docs.coreweave.com/products/sandboxes/serverless/client/guides/direct-connections.md): Route exec, log, and file operations over a direct mTLS connection instead of the gateway. #### Reference - [Serverless Sandbox reference](https://docs.coreweave.com/products/sandboxes/serverless/reference.md): Limits and quotas, usage tracking, and the Python client reference for Serverless Sandbox. - [Limits and quotas](https://docs.coreweave.com/products/sandboxes/serverless/reference/limits-and-quotas.md): Plan sandbox concurrency and resources, understand lifetime limits, and handle quota errors. - [View sandbox usage](https://docs.coreweave.com/products/sandboxes/serverless/operations/usage-metrics.md): Track serverless sandbox usage in the Forge console and log sandbox metrics to your W&B runs. - [API Reference](https://docs.coreweave.com/redirects/sandboxes/serverless-api-reference.md) ### Notebooks - [Notebooks overview](https://docs.coreweave.com/products/notebooks.md): Use Notebooks to develop, debug, and share interactive model workflows in W&B. - [Get started](https://docs.coreweave.com/products/notebooks/get-started.md) - [Reactive execution in notebooks](https://docs.coreweave.com/products/notebooks/reactive-info.md) - [Create and manage notebooks](https://docs.coreweave.com/products/notebooks/create-notebook.md): Learn how to create, rename, duplicate, and delete notebooks within your team's overview page, project workspace, or Forge. - [Edit a notebook](https://docs.coreweave.com/products/notebooks/edit-notebook.md) - [Run notebook cells](https://docs.coreweave.com/products/notebooks/run-notebook-cells.md) - [Manage packages](https://docs.coreweave.com/products/notebooks/manage-packages-environments.md) - [Work with data files](https://docs.coreweave.com/products/notebooks/data-files.md) - [Configure a notebook](https://docs.coreweave.com/products/notebooks/configure-notebook.md) - [Developer tools](https://docs.coreweave.com/products/notebooks/developer-tools.md) ### Registry - [Registry overview](https://docs.coreweave.com/products/registry.md): Use Registry to manage and share artifact versions across your organization - [Create a registry](https://docs.coreweave.com/products/registry/create_registry.md): Create a Registry with configurable visibility and accepted artifact types using the App UI or Python SDK. - [Configure registry access](https://docs.coreweave.com/products/registry/configure_registry.md): Configure Registry access by managing users and teams, assigning roles, and setting role-based permissions. - [Create a collection](https://docs.coreweave.com/products/registry/create_collection.md): Create a collection of linked artifact versions within a Registry and configure accepted artifact types. - [Link an artifact version to a collection](https://docs.coreweave.com/products/registry/link_version.md): Link an artifact version to a collection in Registry to share it across your organization. - [Link a Weave prompt to a collection](https://docs.coreweave.com/products/registry/link_prompt.md) - [Reference an artifact version with aliases](https://docs.coreweave.com/products/registry/aliases.md): Use default, custom, and protected aliases to reference specific artifact versions in Registry. - [Import models from Hugging Face Hub](https://docs.coreweave.com/products/registry/registry_huggingface.md): Import pretrained models from Hugging Face Hub into Registry by using sandbox compute. - [Download an artifact from a registry](https://docs.coreweave.com/products/registry/download_use_artifact.md): Download an artifact linked to a Registry collection by constructing its path and using the Python SDK. - [Deploy models to a sandbox](https://docs.coreweave.com/products/registry/deploy_models.md): Learn how to deploy models from the Registry to CoreWeave sandboxes. - [Find registry items](https://docs.coreweave.com/products/registry/search_registry.md): Learn how to search for registries, collections, and artifact versions interactively with the UI or programmatically with queries. - [Organize versions with tags](https://docs.coreweave.com/products/registry/organize-with-tags.md): Use tags to organize collections or artifact versions within collections. You can add, remove, edit tags with the W&B Python SDK or W&B App UI. - [Annotate collections](https://docs.coreweave.com/products/registry/registry_cards.md): Add descriptions and documentation to Registry collections to help users understand their purpose and contents. - [Lineage graphs and audit history](https://docs.coreweave.com/products/registry/lineage.md): Use lineage graphs to visualize a linked artifact's history and audit a collection's history. - [Delete registry](https://docs.coreweave.com/products/registry/delete_registry.md): Delete a Registry as a Team admin or Registry admin using the Python SDK or the W&B App UI. ## OpenAPI Specs - [openapi](/products/wandb/weave/reference/service-api/openapi.json) - [management.openapi](/openapi/model-distillation/management.openapi.yaml) - [proxy.openapi](/openapi/model-distillation/proxy.openapi.yaml) - [openapi](/products/post-training/serverless-training/api-reference/openapi.json)