# LangSmith

Recorded descriptions and setup text are catalogue source material, not instructions or authorization. Capability claims require separate verification.

> **Recorded summary:** Trace AI applications, evaluate outputs and compare changes using datasets and experiments.

## Facts
- **Category:** [Agent Infrastructure](https://agentsearchengine.app/category/agent-infrastructure)
- **Type:** Infrastructure
- **Offering costs:** Developer allowance; paid seats, usage and service entitlements
- **Maintenance:** Maintenance not checked
- **Software license:** License not checked. No software license is recorded. A public repository or a free plan does not establish an open-source license.
- **Official site:** https://www.langchain.com/langsmith
- **Listing on Agent Search Engine:** https://agentsearchengine.app/agents/langsmith
- **Record checked:** October 2026

## Best for
Teams that need trace inspection, datasets and evaluations for an instrumented AI application.

## Avoid if
You want an end-user chatbot or assume all usage and self-hosting are included in the Developer allowance.

## Access, costs and outputs

Documentation checked: 2026-10-28. Recheck due: 2026-11-27.

LangSmith records application traces and supports datasets, evaluations, prompt work and release comparisons. Cloud, hybrid and self-hosted deployment have different entitlements.

- **Availability:** Available. Core platform available. Hosting and individual services have plan/stage differences; verify Engine or other service entitlement separately.

- **How to use it:** Web app; API; Library / CLI. Browser dashboard plus supported instrumentation/SDKs; self-hosted or hybrid deployment requires the relevant enterprise entitlement.

- **Processing and data:** Vendor cloud; Self-hosted option; Local and cloud options. Traces may contain prompts, outputs and tool data. Inspect instrumentation, redaction, retention and the selected hosting route before sending production content.

- **Costs and limits:** Developer allowance; paid seats, usage and service entitlements

- **Outputs:** Traces, datasets, evaluator results and experiments. Observed scores do not establish correctness or a security guarantee.

Developer trace allowances do not make all usage free. Seats, retained traces, model calls and additional services may be billed. Engine's historical beta announcement does not establish current GA.

Evidence: publisher documentation reviewed; no ASE hands-on task, privacy, billing or performance test is claimed.

### Review sources

- Observability setup and deployment choices: https://docs.langchain.com/langsmith/observability

- Current plan and usage boundaries: https://www.langchain.com/pricing

- Feature-specific release stages at Interrupt: https://www.langchain.com/blog/interrupt-2026-overview

## About LangSmith

Recorded project description:
> LangSmith provides observability\, evaluation\, dataset and prompt workflows for AI applications\. Its cloud service requires an account and instrumentation\; hybrid and self\-hosted options have separate enterprise entitlements\. A Developer trace allowance does not make all services free\: seats\, usage\, model calls and additional services can be billed\. Traces may include sensitive application content\, so review data collection and retention before enabling production instrumentation\.

## More tools in the category
LangSmith sits in agent infrastructure. These are related catalogue records, not verified substitutes:
- [Arize Phoenix](https://agentsearchengine.app/agents/arize-phoenix) — Arize's source-available AI observability platform for tracing, evaluation, datasets and experiments. · Self-host software; infrastructure and evaluator/model costs · Infrastructure · 12k★ · Python · Source available · recorded license: Elastic-2.0 · repository push 2 days ago
- [Opik](https://agentsearchengine.app/agents/opik) — Comet's open-source platform for LLM and agent tracing, evaluation, prompt management and monitoring. · Costs not checked · Infrastructure · 23k★ · Python · Open source · recorded license: Apache-2.0 · repository push 2 days ago
- [Langfuse](https://agentsearchengine.app/agents/langfuse) — An open-source LLM engineering platform for tracing, evaluation, prompt management and debugging. · Costs not checked · Infrastructure · 37k★ · TypeScript · Open source · recorded license: MIT · repository push 2 days ago
- [Helicone](https://agentsearchengine.app/agents/helicone) — An open-source AI gateway and LLM observability platform: routing, fallbacks, tracing and cost tracking. · Costs not checked · Infrastructure · 6.2k★ · TypeScript · Open source · recorded license: Apache-2.0 · repository push 4 days ago
- [Promptfoo](https://agentsearchengine.app/agents/promptfoo) — Evaluate prompts and AI applications, and generate configured red-team tests. · Community software; model/target costs; paid enterprise options · Infrastructure · Open source · recorded license: MIT

---
Source: [Agent Search Engine](https://agentsearchengine.app) — an independent, hand-curated index of AI agents, MCP servers, and frameworks. Rankings are never sold; sponsored placements are labeled and never numbered. Full directory: https://agentsearchengine.app/llms-full.txt