Skip to main content

Weights & Biases Announces LLM Monitoring

Weights & Biases, the AI developer platform, today unveiled LLM Monitoring, a new out-of-the-box capability that allows organizations to centrally monitor their team’s LLM activities in real-time.
Created on November 30|Last edited on November 30
San Francisco – Nov. 30, 2023 – Weights & Biases, the AI developer platform, today unveiled LLM Monitoring, a new out-of-the-box capability that allows organizations to centrally monitor their team’s LLM activities in real-time. Enterprises can now track and manage prompts usage statistics like cost per prompt request, number of prompt tokens, and prompt latency within Weights & Biases, the ecosystem they’re already using across the ML lifecycle.
“As enterprises make significant investments in LLMs and workflows become more complex, there is a pressing need for a single system of record monitoring LLM-related activities,” said Lukas Biewald, CEO and Co-Founder of Weights & Biases. “Now with a simple two-line change or an environment variable, organizations can monitor all of their LLM usage across teams in a single place. LLM monitoring fits a real market need that we’ve been hearing about and we’re excited to roll this out to the market.”
With LLM Monitoring, individuals can filter and drill down to analyze prompt performance, observe all inputs sent to the prompt, and track experiments over time to see which prompts perform best. By monitoring all activities across any variable, including throughput, token usage, cost, latency, errors, and more, organizations can see usage breakdowns across users, teams, or prompts and analyze trade offs between performance, latency, and costs to optimize project ROI.
LLM Monitoring is available now. Those interested can visit https://wandb.ai/site/monitoring and follow the tutorials to get started.
About Weights & Biases
Weights & Biases is the leading developer-first MLOps platform that provides enterprise-grade, end-to-end MLOps workflow to accelerate ML activities. Used by over 900k ML practitioners including teams at OpenAI, Co:here, Aleph Alpha, and Toyota, Weights & Biases is part of the new standard of best practices for machine learning.
###

Tags: ML News
Iterate on AI agents and models faster. Try Weights & Biases today.