Layer / What ran
Observability
Run history, cost, cycle time, and audit trails for agent work.
- AgentOps
Session and tool-call observability for agent runs.
- Braintrust
Evaluation platform for datasets, experiments, scoring, and logs.
- Factory
Commercial agent platform for software delivery workflows.
- Galileo
Enterprise evaluation and observability for AI systems.
- Helicone
Open LLM gateway with logs, caching, and spend controls.
- Langfuse
Self-hostable open-source LLM and agent observability.
- LangSmith
LangChain commercial tracing, evaluation, and prompt platform.
- LangWatch
LLM observability and evaluation platform with open components.
- Logfire
Pydantic observability platform for application and agent traces.
- Phoenix
Arize open-source observability and evaluation for LLM applications.
- Portkey
AI gateway combining provider routing, guardrails, and logs.
- SpecStory
Capture AI IDE sessions as reviewable intent artifacts.
- Warp
Terminal, multi-model agents, and early-access factory workflows.