Evaluate Your AI Agents with Fireworks AI Models in Datadog Agent Observability
Datadog, Monday, November 2nd, 2026
Datadog Agent Observability now supports Fireworks AI as a model provider for LLM-as-a-judge evaluations and Playground experiments.
Datadog announced that Agent Observability supports Fireworks AI, which hosts open models such as GPT-OSS, Kimi, GLM, Qwen and MiniMax.
Teams can bring their own Fireworks AI API key to get visibility into agents built on those models, from traces and token usage to quality evaluations.
Fireworks AI models can power LLM-as-a-judge evaluators that score application inputs and outputs for issues like hallucinations or off-topic answers, and can be tested in the Agent Observability Playground.
Because evaluations run on the customer's Fireworks AI account, usage follows existing billing and governance controls.