wandb.ai

To comply with applicable regulations, starting on September 1st, 2025, our services will no longer be accessible from your location.

Learn More Learn more For more information or if you need help , please contact Weights & Biases Customer Support at retrieving your data support@wandb.com Skip to content AI is to easy productionize The AI developer platform to build AI agents, applications, and models with confidence W&B Weave: Build agentic AI applications import weave weave.init("quickstart") @weave.op() def llm_app(prompt): pass # Track LLM calls, document retrieval, agent steps get started with weave W&B Models: Build AI models import wandb run = wandb.init(project="my-model-training-project") run.config = {"epochs": 1337, "learning_rate": 3e-4} run.log({"metric": 42}) my_model_artifact = run.log_artifact("./my_model.pt", type="model") get started with Models REQUEST DEMO The world’s leading AI teams trust Weights & Biases Meet our customers Weights & Biases AI developer platform Models Build and manage AI models Training Fine-tune AI models on agentic tasks Inference Serve hosted & fine-tuned AI models Weave Iterate, evaluate and monitor agents Registry Datasets | Models | Prompts | Code | Metadata Core Reports | Automations | SDK | Skills for coding agents Secure deployment Saas | Dedicated | Customer-managed Models Build and manage AI models Training Fine-tune with serverless RL Inference Access and explore hosted AI models Weave Iterate, evaluate and monitor agents Registry Datasets | Models | Prompts | Code | Metadata Core Reports | Automations | SDK Secure deployment Saas | Dedicated | Customer-managed Models Build and manage AI models Inference Access and explore hosted AI models Training Fine-tune with serverless RL Weave Iterate, evaluate and monitor agents Registry Datasets | Models | Prompts | Code | Metadata Core Reports | Automations | SDK Secure deployment SaaS | Dedicated | Customer-managed Compliance-ready for the enterprise Weights & Biases AI development platform is certified under ISO/IEC 27001:2022, ISO/IEC 27017:2015, and ISO/IEC 27018:2019, and is compliant with SOC 2 and HIPAA standards.

Our platform also helps customers comply with NIST 800-53 and we are in alignment with GDPR requirements for processing personal information.

Learn more about compliance with different deployment options: | Multi-tenant Dedicated Certified under ISO 27001:2022 ISO 27017:2015 ISO 27018:2019 Compliant with SOC 2 HIPAA NIST 800-53 GDPR Get started with one line of code “I love Weave for a number of reasons.

The fact that I could just add a library to our code and all of a sudden I’ve got a whole bunch of information about the GenAI portion of our product, in Weights & Biases, which I was already using and very familiar with.

All those things that I’m watching for the performance of our AI, I can now report on quickly and easily with Weave.” Mike Maloney, Co-founder and CDO Neuralift INTEGRATE QUICKLY LANGCHAIN LLAMAINDEX PyTorch HF Transformers Lightning TensorFlow Keras Scikit-LEARN XGBoost import openai, weave weave.init("weave-intro") @weave.op def correct_grammar(user_input): client = openai.

Start a new run run = wandb.init(project="gpt5") # 2.

Save model inputs and hyperparameters config = run.config config.dropout = 0.01 # 3.

Log gradients and model parameters run.watch(model) for batch_idx, (data, target) in enumerate(train_loader): ... if batch_idx % args.log_interval == 0: # 4.

Log metrics to visualize performance run.log({"loss": loss}) import wandb ‍ # 1.

Define which wandb project to log to and name your run run = wandb.init(project="gpt-5", run_name="gpt-5-base-high-lr") ‍ # 2.