Weights & Biases

Description

Provide tools for experiment tracking, dataset and model versioning, prompt evaluation, production monitoring, and AI application debugging to help teams develop machine-learning systems.

Weights & Biases is located in the San Francisco ecosystem.

Related

• MercorProvides human data, expert evaluations, and workflow-data monetization tools for AI developers and enterprises deploying and improving…
• Snorkel AIBuilds datasets, evaluation frameworks, environments, and custom AI systems for product and development teams needing domain-specific data…
• Datology AIBuilds a data refinery that helps AI model builders clean, synthesize, and compose customer data into training datasets for open and…
• RekaBuilds AI models and infrastructure for physical-world intelligence, including tools to train and deploy models and process large video…
• Hugging FaceHosts and supports collaboration on public machine learning models, datasets, and applications, while offering enterprise compute, unified…
• Flapping AirplanesResearches data-efficient AI training methods to help models learn from fewer text tokens and reduce dependence on large internet-scale…
• BasetenProvides infrastructure and engineering support to train, deploy, serve, and scale open-source, custom, and fine-tuned AI models for…
• Modal LabsBuilds cloud infrastructure for teams developing, training, and serving AI applications, with GPU tools, model inference, and isolated…
• RebellionsBuilds AI accelerators and full-stack infrastructure for production AI inference, helping data center operators run large models with…
• Scale AIProvides training data, annotation, RLHF, model evaluations, red-teaming, and full-stack systems for enterprises and governments building,…
• Fireworks AIBuilds infrastructure to train, customize, and serve open or customer models, supporting developers and enterprises with model libraries,…
• AnyscaleBuilds Ray-based software for AI teams to scale distributed training, data curation, batch inference, and serving workloads for foundation…