# Weights & Biases Documentation: English Products W&B Weave ## W&B Weave - [W&B Weave](https://docs.wandb.ai/weave.md): Track, test, and improve language model apps with W&B Weave ### Get Started - [Choose an agent integration](https://docs.wandb.ai/weave/agent-integration-quickstart.md): Pick a built-in Weave integration for your agent SDK or harness and start sending conversations, turns, LLM calls, and tool calls to the Agents view. - [Quickstart: Set up custom agent observability](https://docs.wandb.ai/weave/custom-agents-quickstart.md): Trace a multi-turn agent with the Weave SDK. Conversations, turns, LLM calls, and tool calls render in the Agents view of your project. - [Evaluate your AI agent with Weave](https://docs.wandb.ai/weave/agent-evals.md): Evaluate single-turn and multi-turn AI agents in Weave using the agents workflow and EvaluationLogger, scored with an LLM judge. - [Send OpenTelemetry spans to the Agents view](https://docs.wandb.ai/weave/guides/tracking/trace-agents-otel.md): Send OTLP trace data from any OpenTelemetry pipeline to the Weave agents endpoint, with no Weave SDK required. - [Quickstart: Instrument and trace functions](https://docs.wandb.ai/weave/quickstart.md): Get started with W&B Weave by adding tracing to your LLM application to debug and monitor model interactions. ### Guides - [What is Weave?](https://docs.wandb.ai/weave/concepts/what-is-weave.md): Learn about W&B Weave and how it helps you build, evaluate, and improve LLM applications - [Manage Weave projects](https://docs.wandb.ai/weave/guides/platform/weave-projects.md): Use Weave projects to organize related assets like traces, prompts, evaluations, models, and dashboards. - [Configure Weave environment variables](https://docs.wandb.ai/weave/guides/core-types/env-vars.md): Configure Weave's behavior at runtime using environment variables #### Trace your agents - [Trace your agents](https://docs.wandb.ai/weave/guides/tracking/trace-agents.md): Use the Weave SDK to instrument multi-turn agentic applications and view them in the Agents tab. - [Send OpenTelemetry spans to the Agents view](https://docs.wandb.ai/weave/guides/tracking/trace-agents-otel.md): Send OTLP trace data from any OpenTelemetry pipeline to the Weave agents endpoint, with no Weave SDK required. - [View agent activity](https://docs.wandb.ai/weave/guides/tracking/view-agent-activity.md): Use W&B Weave's Agents view to understand what your agent did, how much it cost, and exactly where things went right or wrong. - [Monitor your agents with signals](https://docs.wandb.ai/weave/guides/tracking/view-agent-signals.md): Use W&B Weave signals to score agent turns, apply tags, and surface quality and safety issues in production. - [Create a custom signal](https://docs.wandb.ai/weave/guides/tracking/create-custom-signal.md): Create a custom W&B Weave signal to tag agent turns for a behavior you define. - [Set attributes and events on agent spans](https://docs.wandb.ai/weave/guides/tracking/trace-agents-attributes.md): Attach custom attributes and record events on agent spans (Turn, LLM, Tool, and SubAgent) to filter and analyze agent activity in Weave. - [Batch logging for your agent](https://docs.wandb.ai/weave/guides/tracking/trace-agents-batch.md): Manually log agent traces for frameworks that have already completed the LLM call and need to record it. - [Trace sub-agents](https://docs.wandb.ai/weave/guides/tracking/trace-sub-agents.md): Use Weave's sub-agent span to trace sub-agent delegations and view nested agent invocations. - [Reference media in your own bucket (BYOB) using agent spans](https://docs.wandb.ai/weave/guides/tracking/agents-byob-references.md): Render images and video that live in your own cloud bucket in the Weave Agents view by returning their URIs from agent tool spans, without importing the bytes into Weave. ##### Integrate Weave with agents - [Google ADK](https://docs.wandb.ai/weave/guides/integrations/agents/google-adk.md): Trace an agent built with Google's Agent Development Kit (ADK) using Weave. - [OpenAI Agents SDK](https://docs.wandb.ai/weave/guides/integrations/agents/openai-agents-sdk.md): Trace an agent built with the OpenAI Agents SDK using Weave. - [Claude Agent SDK](https://docs.wandb.ai/weave/guides/integrations/agents/claude-agents-sdk.md): Trace an agent built with the Claude Agent SDK using Weave. - [Claude Code plugin](https://docs.wandb.ai/weave/guides/integrations/agents/claude-code-harness.md): Track Claude Code sessions in W&B Weave for observability and debugging. - [Codex plugin](https://docs.wandb.ai/weave/guides/integrations/agents/codex-harness.md): Trace Codex agentic sessions, LLM calls, and tool executions in W&B Weave. - [OpenClaw plugin](https://docs.wandb.ai/weave/guides/integrations/agents/openclaw-harness.md): Track OpenClaw agent sessions in W&B Weave for observability and debugging. - [Pi extension](https://docs.wandb.ai/weave/guides/integrations/agents/pi-dev-harness.md): Trace Pi agentic sessions, LLM calls, and tool executions in Weave. #### Trace functions with Ops and Calls - [Track costs](https://docs.wandb.ai/weave/guides/tracking/costs.md): Understand automatic and custom cost tracking for LLM operations in Weave - [Create and manage saved views](https://docs.wandb.ai/weave/guides/tools/saved-views.md): Customize how you interact with traced function calls and evaluations - [Use trace plots](https://docs.wandb.ai/weave/guides/tracking/trace-plots.md): Visualize trace-level metrics with interactive charts in Weave to explore latency, cost, and token usage patterns for LLM applications. - [Link a W&B run to trace function calls](https://docs.wandb.ai/weave/guides/tracking/trace-to-run.md): Associate Weave traces with W&B runs for experiment tracking - [Use Weave with W&B training runs](https://docs.wandb.ai/weave/guides/tools/weave-in-workspaces.md): Integrate Weave traces with W&B training runs to view function execution details alongside ML metrics in workspace dashboards. ##### Integrate with your LLM provider and frameworks - [Integrations overview](https://docs.wandb.ai/weave/guides/integrations.md): Trace agents and LLM applications with Weave's integrations for agent SDKs, harnesses, LLM providers, and orchestration frameworks. - [Control automatic LLM call tracking](https://docs.wandb.ai/weave/guides/integrations/autopatching.md): Control how W&B Weave automatically records calls to OpenAI, Anthropic, and other LLM libraries - [TypeScript SDK: third-party integration guide](https://docs.wandb.ai/weave/guides/integrations/js.md): Integrate third-party libraries with the Weave TypeScript SDK - [Local Models](https://docs.wandb.ai/weave/guides/integrations/local_models.md): Run open source models locally with Ollama, LMStudio, or GPT4All and trace them with Weave using OpenAI SDK compatibility. ###### LLM Providers - [Serverless Inference](https://docs.wandb.ai/weave/guides/integrations/inference.md): Use Serverless Inference with Weave to trace and monitor open source foundation model calls through the OpenAI-compatible API. - [Bedrock](https://docs.wandb.ai/weave/guides/integrations/bedrock.md): Track and monitor Amazon Bedrock LLM calls with Weave, capturing foundation model interactions and converse API usage. - [Anthropic](https://docs.wandb.ai/weave/guides/integrations/anthropic.md): Use Weave to automatically track and log LLM calls made with the Anthropic SDK - [Cerebras](https://docs.wandb.ai/weave/guides/integrations/cerebras.md): Use Weave to trace and log LLM calls made via the Cerebras Cloud SDK - [Cohere](https://docs.wandb.ai/weave/guides/integrations/cohere.md): Use Weave to automatically track and log LLM calls made through the Cohere Python library - [Google](https://docs.wandb.ai/weave/guides/integrations/google.md): Use Weave to trace and log Google GenAI model calls - [Groq](https://docs.wandb.ai/weave/guides/integrations/groq.md): Track and monitor Groq LPU inference with Weave, capturing model calls, performance metrics, and function chains for LLM applications. - [Hugging Face Hub](https://docs.wandb.ai/weave/guides/integrations/huggingface.md): Integrate Hugging Face Hub with W&B Weave to track and analyze your machine learning applications - [LiteLLM](https://docs.wandb.ai/weave/guides/integrations/litellm.md): Automatically track and log LLM calls made through LiteLLM - [Microsoft Azure](https://docs.wandb.ai/weave/guides/integrations/azure.md): Integrate Weave with Microsoft Azure OpenAI services - [MistralAI](https://docs.wandb.ai/weave/guides/integrations/mistral.md): Track and monitor MistralAI model calls with Weave's automatic tracing for chat completions, function calling, and model interactions. - [NVIDIA NIM](https://docs.wandb.ai/weave/guides/integrations/nvidia_nim.md): Use Weave to trace and log LLM calls made through the ChatNVIDIA library - [OpenAI](https://docs.wandb.ai/weave/guides/integrations/openai.md): Integrate OpenAI with Weave for tracing, evaluation, and monitoring - [OpenRouter](https://docs.wandb.ai/weave/guides/integrations/openrouter.md): Use OpenRouter's unified interface for many LLMs with automatic Weave integration - [Together AI](https://docs.wandb.ai/weave/guides/integrations/together_ai.md): Track Together AI's open-source LLMs with Weave using OpenAI SDK compatibility for seamless tracing and evaluation of model calls. ###### Frameworks - [OpenAI Realtime API](https://docs.wandb.ai/weave/guides/integrations/openai-realtime-audio.md): Use Weave to automatically trace your calls to the OpenAI Realtime API. - [Bedrock Agents](https://docs.wandb.ai/weave/guides/integrations/bedrock_agents.md): Trace Amazon Bedrock Agents invocations with Weave, capturing agent inputs, foundation model usage, and completion output. - [LangChain](https://docs.wandb.ai/weave/guides/integrations/langchain.md): Use Weave to track and log all calls made through the LangChain Python library - [LlamaIndex](https://docs.wandb.ai/weave/guides/integrations/llamaindex.md): Trace and debug LlamaIndex applications with Weave, capturing LLM calls, RAG pipelines, agent steps, and evaluations automatically. - [DSPy](https://docs.wandb.ai/weave/guides/integrations/dspy.md): Use Weave to automatically track and log calls made using DSPy modules and functions - [Instructor](https://docs.wandb.ai/weave/guides/integrations/instructor.md): Trace structured data extraction from LLMs with Weave's Instructor integration, capturing Pydantic validation and retry logic. - [CrewAI](https://docs.wandb.ai/weave/guides/integrations/crewai.md): Integrate CrewAI with Weave for monitoring and tracing multi-agent applications - [Smolagents](https://docs.wandb.ai/weave/guides/integrations/smolagents.md): Track Smolagents applications with Weave's automatic tracing, capturing tool calls, model inferences, and multi-step agent workflows. - [PydanticAI](https://docs.wandb.ai/weave/guides/integrations/pydantic_ai.md): Trace PydanticAI agent and tool calls in Weave using OpenTelemetry - [Vercel AI SDK](https://docs.wandb.ai/weave/guides/integrations/vercel_ai_sdk.md): Trace Vercel AI SDK calls in Weave using OpenTelemetry - [Haystack](https://docs.wandb.ai/weave/guides/integrations/haystack.md): Trace Deepset Haystack pipelines with W&B Weave using the WeaveConnector integration. - [Agno](https://docs.wandb.ai/weave/guides/integrations/agno.md): Trace Agno multi-agent systems with Weave using OpenTelemetry to capture agent interactions, tool calls, and multi-modal workflows. - [Koog](https://docs.wandb.ai/weave/guides/integrations/koog.md): Trace Kotlin-based Koog agents with Weave using OpenTelemetry to capture tool calls, prompts, and completions for AI applications. - [AutoGen](https://docs.wandb.ai/weave/guides/integrations/autogen.md): Use Weave to track and monitor AutoGen agents and multi-agent systems - [Verdict](https://docs.wandb.ai/weave/guides/integrations/verdict.md): Use Verdict evaluation framework with Weave to trace and monitor your LLM evaluation pipelines - [Verifiers](https://docs.wandb.ai/weave/guides/integrations/verifiers.md): Track Verifiers RL environments and LLM agent training with Weave, capturing multi-round conversations and evaluation rollouts. - [VERL](https://docs.wandb.ai/weave/guides/integrations/verl.md): Trace VERL rollouts in Weave to inspect multi-turn conversations, tool calls, and reward scoring during RL fine-tuning. ###### Protocols - [Model Context Protocol (MCP) and Weave](https://docs.wandb.ai/weave/guides/integrations/mcp.md): Trace activity between your MCP client and MCP server with Weave - [Send OpenTelemetry Traces to Weave](https://docs.wandb.ai/weave/guides/tracking/otel.md): Import OpenTelemetry compatible trace data through a dedicated endpoint ##### Tracing basics - [Quickstart: Instrument and trace functions](https://docs.wandb.ai/weave/quickstart.md): Get started with W&B Weave by adding tracing to your LLM application to debug and monitor model interactions. - [Understand Ops, Calls, and Traces](https://docs.wandb.ai/weave/guides/tracking/tracing.md): Learn how Ops, Calls, and Traces create the foundation of W&B Weave's tracing system. - [Trace your code](https://docs.wandb.ai/weave/guides/tracking/create-call.md): Instrument your running code so its execution appears as detailed traces in W&B Weave. - [Navigate the Weave Trace view](https://docs.wandb.ai/weave/guides/tracking/trace-tree.md): Use Weave's Trace view to make sense of complex execution paths in your LLM and agentic apps. - [Compare traces and other logged information](https://docs.wandb.ai/weave/guides/tools/comparison.md): Visually compare and diff code, traces, prompts, models, and configurations - [Query and export Calls](https://docs.wandb.ai/weave/guides/tracking/querying-calls.md): Filter, sort, and export Weave call data using the Python SDK, REST API, or the Weave UI for custom analysis. ##### Advanced Ops - [Trace generator functions](https://docs.wandb.ai/weave/guides/tracking/trace-generator-func.md): Track sync and async generator functions with W&B Weave tracing - [Trace nested functions](https://docs.wandb.ai/weave/tutorial-tracing_2.md): Learn how to track deeply nested Call structures with W&B tracing - [Trace threads](https://docs.wandb.ai/weave/guides/tracking/threads.md): Trace and analyze multi-turn conversations in your LLM applications using threads. - [Customize Ops](https://docs.wandb.ai/weave/guides/tracking/ops.md): Learn how to color your Ops for better visibility, how to modify what's logged, and how to control the sampling rate - [Define and log attributes](https://docs.wandb.ai/weave/guides/tools/attributes.md): Use attributes to add metadata to your traces and evaluations. - [Log media](https://docs.wandb.ai/weave/guides/core-types/media.md): Log media returned in your traces, such as images and videos. - [Reference media in your own bucket (BYOB) using Weave Op](https://docs.wandb.ai/weave/guides/tracking/byob-references.md): Render images and video that live in your own cloud bucket in Weave traces by logging their URIs, without importing the bytes into Weave. - [View and customize trace display](https://docs.wandb.ai/weave/guides/tracking/view-call.md): View Calls in the UI and customize how trace data is displayed - [Disable tracing](https://docs.wandb.ai/weave/guides/tracking/trace-disable.md): Learn options to disable or conditionally turn off W&B Weave tracing - [Write-ahead log](https://docs.wandb.ai/weave/guides/tracking/write-ahead-log.md): Improve the resilience of W&B Weave trace data capture with the write-ahead log ##### Work with Calls - [Update and delete Calls](https://docs.wandb.ai/weave/guides/tracking/update-call.md): Modify display names, add feedback, and delete Calls in W&B Weave - [Call schema reference](https://docs.wandb.ai/weave/guides/tracking/call-schema-reference.md): Reference for the Call object structure and properties - [Get a handle to the Call object during execution](https://docs.wandb.ai/weave/guides/tracking/get-call-object.md): Access the W&B Weave `Call` object at runtime for feedback, display names, and other metadata - [Set Call display name](https://docs.wandb.ai/weave/guides/tracking/set-call-display.md): Set or override the display name for a Call in W&B Weave tracing #### Evaluate your agents and functions - [Evaluations overview](https://docs.wandb.ai/weave/guides/core-types/evaluations.md): Evaluation-driven LLM application development to systematically improve applications - [Build an evaluation](https://docs.wandb.ai/weave/tutorial-eval.md): Learn how to build an evaluation pipeline with Weave Models and Evaluations - [Evaluate RAG applications](https://docs.wandb.ai/weave/tutorial-rag.md): Build and evaluate RAG applications using Weave with LLM judges - [Collect and track datasets](https://docs.wandb.ai/weave/guides/core-types/datasets.md): Organize, collect, track, and version examples for LLM application evaluation - [Scoring overview](https://docs.wandb.ai/weave/guides/evaluation/scorers.md): Evaluate AI outputs and return evaluation metrics with Weave Scorers - [Use builtin scorers](https://docs.wandb.ai/weave/guides/evaluation/builtin_scorers.md): Use Weave's predefined scorers for evaluating your AI applications - [Evaluate using local scorers](https://docs.wandb.ai/weave/guides/evaluation/weave_local_scorers.md): Small language models that run locally to evaluate AI system safety and quality - [Log evaluation data from your code](https://docs.wandb.ai/weave/guides/evaluation/evaluation_logger.md): Flexible, incremental way to log evaluation data from Python and TypeScript code - [Export evaluation data](https://docs.wandb.ai/weave/guides/evaluation/export_eval.md): Programmatically export evaluation results using the Evaluation REST API. - [Compare evaluations](https://docs.wandb.ai/weave/guides/evaluation/compare_evals.md): Visually compare two or more evaluations to spot regressions, improvements, and scoring differences across runs - [Compare and rank models](https://docs.wandb.ai/weave/guides/core-types/leaderboards.md): Compare and rank different model versions based on evaluation metrics - [Map columns in datasets](https://docs.wandb.ai/weave/guides/tools/column-mapping.md): Map columns in datasets to different names. This helps you align the column names in your dataset with the column names expected by the scorer. - [Create dynamic Leaderboards in Evaluations](https://docs.wandb.ai/weave/guides/evaluation/dynamic_leaderboards.md): Dynamic Leaderboards let you configure, customize, save, and update Leaderboard views directly from an evaluation. #### Experiment with prompts and models - [Use the Playground to experiment with prompts](https://docs.wandb.ai/weave/guides/tools/playground.md): Simplify the process of iterating on LLM prompts and responses - [Compare model performance using the Evaluation Playground](https://docs.wandb.ai/weave/guides/tools/evaluation_playground.md): Compare and evaluate model performance without code using Weave's Evaluation Playground with custom datasets and LLM judges. - [Add custom runtimes](https://docs.wandb.ai/weave/guides/tools/custom-runtimes.md): Connect OpenAI-compatible inference endpoints to a Weave project as custom runtimes, and manage them from the UI or the Python and TypeScript SDKs. - [Learn Weave with Serverless Inference](https://docs.wandb.ai/weave/quickstart-inference.md): Learn Weave fundamentals using Serverless Inference to trace model calls, compare outputs, and run evaluations. #### Version your work - [Tutorial: App versioning](https://docs.wandb.ai/weave/tutorial-weave_models.md): Learn how to use Weave Model to track and version your application and its parameters - [Track application versions with models](https://docs.wandb.ai/weave/guides/core-types/models.md): Track versions of your application with structured models that combine data and code. - [Create prompt objects](https://docs.wandb.ai/weave/guides/core-types/prompts.md): Create, publish, and use prompt objects for LLM applications - [Store and track versions of prompts](https://docs.wandb.ai/weave/guides/core-types/prompts-version.md): Retrieve and manage versions of your prompts for LLM applications - [Track and version objects](https://docs.wandb.ai/weave/guides/tracking/objects.md): Track and version any JSON-serializable object in W&B Weave #### Monitor and collect feedback - [Redact PII from traces](https://docs.wandb.ai/weave/guides/tracking/redact-pii.md): Automatically redact Personally Identifiable Information from traces - [Monitor using built-in signals](https://docs.wandb.ai/weave/guides/evaluation/monitors.md): W&B provides built-in scoring solutions for agents in production - [Set up custom monitors](https://docs.wandb.ai/weave/guides/evaluation/custom-monitors.md): Passively score production traffic to surface trends and issues - [Set up automations](https://docs.wandb.ai/weave/guides/evaluation/automations.md): Create event-driven automations that trigger actions based on monitor metrics and trace activity. - [Set up guardrails](https://docs.wandb.ai/weave/guides/evaluation/guardrails.md): Ensure LLM safety and measure output quality in production applications ##### Annotate - [Set up annotation queues](https://docs.wandb.ai/weave/guides/tracking/annotation-queues.md): Create annotation queues, route traces to domain experts, and export structured feedback. - [Review items in an annotation queue](https://docs.wandb.ai/weave/guides/tracking/annotation-review.md): Evaluate trace items and submit structured feedback using a simplified review interface. - [Collect feedback and use annotations](https://docs.wandb.ai/weave/guides/tracking/feedback.md): Collect and analyze feedback for LLM applications through UI and SDK #### Deploy and scale - [Deployment options and security features](https://docs.wandb.ai/weave/guides/platform.md): Learn about Weave's deployment options, IAM management, and data security features. - [Set up a self-managed W&B Weave instance](https://docs.wandb.ai/weave/guides/platform/weave-self-managed.md): Deploy and manage Weave on your own infrastructure - [Configure ingest sampling for Weave Self-Managed](https://docs.wandb.ai/weave/guides/platform/ingest-sampling.md): Keep only a share of incoming traces to control storage and LLM scoring costs on a self-managed Weave instance - [Server response caching](https://docs.wandb.ai/weave/guides/platform/server-caching.md): Improve performance with Weave server response caching for repeated queries and limited bandwidth ### Cookbooks - [Overview](https://docs.wandb.ai/weave/cookbooks.md): Practical guides and examples for building with Weights & Biases Weave #### Weave fundamentals - [Introduction to Traces](https://docs.wandb.ai/weave/cookbooks/Intro_to_Weave_Hello_Trace.md): Learn how to use introduction to traces with W&B Weave - [Introduction to Evaluations](https://docs.wandb.ai/weave/cookbooks/Intro_to_Weave_Hello_Eval.md): Learn how to use introduction to evaluations with W&B Weave #### Evaluations & Datasets - [Leaderboard Quickstart](https://docs.wandb.ai/weave/cookbooks/leaderboard_quickstart.md): Learn how to use leaderboard quickstart with W&B Weave - [HuggingFace Dataset Evaluations](https://docs.wandb.ai/weave/cookbooks/hf_dataset_evals.md): Learn how to use HuggingFace dataset evaluations with W&B Weave - [Import from CSV](https://docs.wandb.ai/weave/cookbooks/import_from_csv.md): Import datasets from CSV files into W&B Weave for use in evaluations, tracing, and model comparison workflows. #### Models & Prompts - [Use Weave with W&B Models](https://docs.wandb.ai/weave/cookbooks/Models_and_Weave_Integration_Demo.md): Interactive notebook that walks through tracking experiments in W&B Models alongside LLM traces and evaluations in Weave. - [Chain Of Density](https://docs.wandb.ai/weave/cookbooks/chain_of_density.md): Implement chain-of-density summarization techniques with W&B Weave for iterative text compression and evaluation. - [DSPy Prompt Optimization](https://docs.wandb.ai/weave/cookbooks/dspy_prompt_optimization.md): Learn how to use dspy prompt optimization with W&B Weave - [Not Diamond custom routing](https://docs.wandb.ai/weave/cookbooks/notdiamond_custom_routing.md): Learn how to use Not Diamond custom routing with W&B Weave #### Advanced Topics - [Multi-agent structured output](https://docs.wandb.ai/weave/cookbooks/multi-agent-structured-output.md): Learn how to use multi-agent structured output with W&B Weave - [Codegen](https://docs.wandb.ai/weave/cookbooks/codegen.md): Build and evaluate code generation pipelines with W&B Weave to trace prompts, outputs, and quality metrics. - [Trace and Evaluate a Computer Vision Pipeline with Weave](https://docs.wandb.ai/weave/cookbooks/ocr-pipeline.md): Learn how to use trace and evaluate a computer vision pipeline with weave with W&B Weave - [Audio with Weave](https://docs.wandb.ai/weave/cookbooks/audio_with_weave.md): Track and evaluate audio data in your LLM applications using W&B Weave's audio logging and tracing features. #### Production & Monitoring - [Online Monitoring](https://docs.wandb.ai/weave/cookbooks/online_monitoring.md): Set up online monitoring for LLM applications in production using W&B Weave to track performance and quality metrics. - [Feedback Prod](https://docs.wandb.ai/weave/cookbooks/feedback_prod.md): Collect and analyze user feedback on LLM outputs in production using W&B Weave's feedback tracking features. - [Scorers as guardrails](https://docs.wandb.ai/weave/cookbooks/scorers_as_guardrails.md): Learn how to use scorers as guardrails with W&B Weave - [Custom Model Cost](https://docs.wandb.ai/weave/cookbooks/custom_model_cost.md): Define and track custom model costs in W&B Weave to monitor spending across different LLM providers and models. - [PII Data Handling](https://docs.wandb.ai/weave/cookbooks/pii.md): Handle personally identifiable information (PII) in LLM applications using W&B Weave's data redaction and filtering tools. ### Reference - [Python SDK](https://docs.wandb.ai/weave/ref-link-python.md) - [TypeScript SDK](https://docs.wandb.ai/weave/ref-link-typescript.md) - [Service API](https://docs.wandb.ai/weave/ref-link-service-api.md) ### Details & Support - [Limits and expected behaviors](https://docs.wandb.ai/weave/details/limits.md): A list of Weave's limitations, known issues, and expected behaviors - [Support: Weave](https://docs.wandb.ai/weave/details/support-weave.md)