- Home
- Prompting Tool
- LangSmith
LangSmith
San Francisco, California, United States
What it is
LangSmith is a platform for building, evaluating, deploying, monitoring, and improving AI agents. Teams can assess curated datasets offline and score production interactions in real time. Evaluation options include human review, heuristic checks, LLM-as-judge, pairwise comparison, and custom evaluators written in Python or TypeScript. Annotation queues let teams send runs to subject-matter experts and apply feedback to evaluations, prompts, or datasets. The Playground supports prompt experiments and comparisons across versions or model providers; LangSmith chat can also improve prompts. For observability, it provides agent tracing, cost tracking, trajectory monitoring, online LLM and code evaluations, and webhook or PagerDuty alerts. Traces can come through an SDK or API from LangGraph, custom Python, or other stacks, and evaluation workflows can connect with pytest, Vitest, and GitHub workflows. Hosting options include Cloud, Hybrid, and Self-Hosted for Enterprise. The Developer plan is free; Plus costs $39/user/mo. Base traces are retained for 14 days, with 180-day extended retention available for an additional fee.
For and against
For
- Supports offline and real-time evaluations
- Custom evaluators can use Python or TypeScript
- Annotation queues incorporate subject-matter expert feedback
- Integrates with pytest, Vitest, and GitHub workflows
- Enterprise offers cloud, hybrid, and self-hosted options
Against
- Base traces are retained for 14 days
- Extended trace retention costs extra
- Plus costs $39/user/mo
The verdict
LangSmith brings agent evaluation, prompt work, and observability into one platform, with a free Developer plan and paid Plus tier. Teams should consider trace-retention limits and compare the available hosting and support options with their needs.
Who it suits
LangSmith suits teams developing AI agents that need evaluation, prompt tools, tracing, and production monitoring. Its framework-agnostic trace intake and hosting options may fit teams using varied application stacks or deployment models.
Plans and prices
Developer
1 seat · 5,000 base traces/month · community support
Read 15 Oct 2026 · langchain.com
Plus
10,000 base traces/month · unlimited seats · 1 free small serverless deployment
Read 15 Oct 2026 · langchain.com
Enterprise
self-hosted or hybrid deployment · custom SSO, ABAC and RBAC · support SLA
Read 15 Oct 2026 · langchain.com
On the record
- API and CLI
- LangSmith exposes a REST API and a CLI that manages projects, traces, runs, datasets, evaluators, experiments, threads and sandboxes.github.com · read 15 Oct 2026
- CI/CD
- The evaluation workflow integrates with pytest, Vitest and GitHub workflows for automated regression checks.langchain.com · read 15 Oct 2026
- CLI installation
- The CLI provides installation paths for macOS, Linux and Windows through shell scripts, PowerShell, Homebrew, Scoop, Go or downloadable binaries.github.com · read 15 Oct 2026
- Company origins
- LangChain says the company began in early 2023 and that LangSmith launched first as a commercial observability product before expanding into evaluation and deployment.langchain.com · read 15 Oct 2026
- Company profile
- LinkedIn lists LangChain as privately held, headquartered in San Francisco, with 201–500 employees and 469 employees shown on the page.linkedin.com · read 15 Oct 2026
- Data residency
- Cloud data is hosted in GCP US or EU regions, while self-hosted data remains in the customer's VPC.langchain.com · read 15 Oct 2026
- Data use
- LangChain states that LangSmith data is not used to train models and remains owned by the customer.langchain.com · read 15 Oct 2026
- Deployment
- Deployment provides managed or self-hosted agent servers with streaming, cron scheduling, authentication, MCP support and more than 30 API endpoints.langchain.com · read 15 Oct 2026
- Documentation
- LangChain publishes reference documentation, integration guides, deployment instructions, a CLI guide and a REST API reference.docs.langchain.com · read 15 Oct 2026
- Evaluation
- Evaluation supports offline datasets, online production evaluations, human annotation queues and side-by-side experiment comparison.langchain.com · read 15 Oct 2026
- Evaluators
- Its evaluation framework supports human review, heuristic checks, LLM-as-judge evaluators, pairwise comparisons, and custom evaluators in Python or TypeScript.langchain.com · read 5 Oct 2026
- Evaluator types
- Evaluators can be human, heuristic, LLM-as-judge, pairwise or custom Python and TypeScript code.langchain.com · read 15 Oct 2026
- Framework support
- LangSmith is framework-agnostic and accepts traces through its SDK or API from LangGraph, custom Python, or other application stacks.langchain.com · read 5 Oct 2026
- Funding
- LangChain announced a $125 million Series B at a $1.25 billion valuation on October 20, 2025.langchain.com · read 15 Oct 2026
- Hosting
- The pricing page lists Cloud hosting for Developer and Plus, with Cloud, Hybrid, or Self-Hosted options for Enterprise.langchain.com · read 5 Oct 2026
- Human feedback
- Annotation queues let teams assign runs to subject-matter experts and use feedback to calibrate evaluations, improve prompts, or augment datasets.langchain.com · read 5 Oct 2026
- Integrations
- Documented integrations cover OpenAI, Anthropic, Bedrock, Gemini, Mistral, LiteLLM, OpenTelemetry, CrewAI, Vercel AI SDK, PydanticAI, AutoGen and many other frameworks and providers.docs.langchain.com · read 15 Oct 2026
- Legal entity
- Crunchbase identifies the maker as LangChain, Inc., a privately held San Francisco company founded by Harrison Chase and Ankush Gola.crunchbase.com · read 15 Oct 2026
