LangSmith
LangChain's platform for tracing, evaluating and deploying agents
Overview
LangSmith is LangChain Inc.'s commercial agent-engineering platform: tracing and monitoring, offline and online evaluations, datasets, annotation queues and prompt management, plus agent deployment, no-code agents, sandboxes and an LLM gateway. It works with any framework through Python, TypeScript, Go and Java SDKs and OpenTelemetry.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 14 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- Up to 5k base traces a month, then pay as you go
- Community support
- Google or GitHub sign-in
- Up to 10k base traces a month, then pay as you go
- 1 free small serverless deployment; access to Deployment and Engine
- Add unlimited seats
- Self-hosted and hybrid deployment
- Custom SSO, ABAC and RBAC
- Support SLA, custom seats and workspaces
- 0.005 LSU per base trace (14-day retention)
- 0.0025 LSU to upgrade a trace to 180-day retention
- Deployment, sandbox and Fleet usage metered in LSUs
Use cases
What you can produce with LangSmith
- Traces of every LLM call, tool use and agent step, from LangChain or any framework via SDK or OpenTelemetry
- Evaluation datasets built from curated examples or production traces, with scored experiments
- Online evaluators, monitoring and alerts on live production traffic
- Human feedback collected through annotation queues
- Prompt versions iterated and compared in the Playground and Prompt Hub
- Agents deployed on managed serverless or dedicated infrastructure, debuggable in LangSmith Studio
- Automated failure clustering and suggested fixes from LangSmith Engine
LangSmith is the most complete option for teams already on LangChain or LangGraph, where tracing works almost out of the box, and it is no longer tied to that stack: SDKs and OpenTelemetry bring in any framework. Evals, annotation queues, a prompt hub and managed agent deployment sit in one product.
The trade-offs are cost and hosting. Seats are $39 a month on Plus and traces beyond the allowance cost about $5 per thousand, and self-hosting needs an Enterprise contract, with cloud data in the US or EU only. Open-source alternatives win on control; LangSmith wins on breadth.
What we found
LangSmith is the commercial platform of LangChain Inc., the San Francisco company behind the open-source LangChain, LangGraph and deepagents frameworks. It opened in closed beta in July 2023 and reached general availability in February 2024, alongside a Series A led by Sequoia. In October 2025 LangChain raised $125 million at a $1.25 billion valuation, led by IVP with Sequoia, Benchmark, Amplify, CapitalG and Sapphire; no later primary round was found. LangChain says its frameworks have 90 million combined monthly downloads and that LangSmith trace volume grew twelvefold in a year.
The product has grown well beyond tracing. Observability and evaluation remain the core: traces, monitoring, an Insights agent that clusters agent behaviour, online and offline evals, datasets, annotation queues for human feedback and a prompt hub with playground. Around them now sit Deployment for long-running agents, Fleet for no-code agents, Engine for finding failures and proposing fixes, sandboxes for agent-generated code and an LLM gateway. It is framework-agnostic, with Python, TypeScript, Go and Java SDKs and OpenTelemetry ingest, so code need not use LangChain at all.
Developer is free for one seat with 5,000 base traces a month. Plus costs $39 per seat a month with 10,000 base traces and one free small serverless deployment. Enterprise is custom and is the only plan with hybrid or self-hosted deployment, custom SSO and a support SLA. Beyond the allowance, usage is billed in LangChain Standard Units at $1 each: a base trace costs 0.005 LSU, about $5 per thousand, with 14-day retention, and upgrading a trace to 180-day extended retention adds 0.0025 LSU. Startups can apply for up to $10,000 in credits.
Cloud data is held in the US or EU, so ASEAN teams with residency requirements are pushed toward an Enterprise self-hosted contract. Earlier versions of this page cited a price comparison against Langfuse and unnamed outside reviews; neither could be traced, and both were removed.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including LangSmith's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with LangSmith unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to LangSmith directly →
Spotted something out of date? Suggest an update →
More in Code & Dev Tools
