Plans for teams of any size

Our Developer plan is a great choice for personal projects. You will have 1 free seat with access to LangSmith (5k base traces/month included).  

The Plus plan is for teams that want to self-serve with moderate usage and collaboration needs. You can purchase unlimited seats with access to LangSmith (10k base traces/month included). You will be able to ship agents with our managed LangSmith Deployment service, with 1 free small serverless deployment included. 

The Enterprise plan is for teams that need more advanced administration, security, support, or deployment options. Contact our sales team to learn more.

Yes! We offer a Startup Plan for LangSmith, designed for early-stage companies building agentic applications. You’ll get discounted rates and generous free trace allotments to build with confidence from day one.

‍Apply here to get started with startup pricing.

For the Developer or Plus Plan, seats are billed monthly on the 1st or pro-rated if added mid-month (no credit for removed seats); traces are billed monthly in arrears for your usage. Enterprise plans are invoiced annually upfront.

No. LangSmith does not use your data to train models. Your traces, prompts, and outputs remain private to your organization. See LangSmith Terms of Service for more information.

A seat is a distinct user inside your organization. We consider the total number of users (including invited users) to determine the number of seats to bill.

Yes, Developer plan users can easily upgrade to the Plus plan on the Plans and Billing page. For the Enterprise plan, please contact our sales team to discuss your needs.

LangChain Compute Unit (LCU): A LCU is a normalized unit of work done on the LangSmith Platform and is used for pricing purposes. Different services in LangSmith that do work and consume compute, tokens, or processing power all meter at different rates and are measured in LCUs. Examples of LangSmith products that consume LCUs include Engine, Fleet, deployments, and sandboxes. 1 LCU = $1.50.

LangChain Storage Unit (LSU): A LSU is a normalized unit of data stored or managed by the LangSmith Platform and is used for pricing purposes. Different services in LangSmith that store data in a database or in memory all meter at different rates and are measured in LSUs. Examples of LangSmith products that consume LSUs include traces, deployments, and sandboxes. 1 LSU = $1.00

A trace represents a single execution of your application—whether it’s an agent, evaluator, or playground session. It can include many individual steps, such as LLM calls and other tracked events. Here's an example of a single trace.

Base traces have a shorter retention period of 14 days. Extended traces have a longer retention period of 180 days. You can "upgrade" base traces to extended traces for an additional fee.

Base traces are short-lived (14-day retention) and ideal for quick debugging or ad-hoc analysis. They’re priced for volume and short-term utility.

Extended traces are retained for 180 days. This is useful when traces include valuable feedback associated with them, such as from users, evaluators, or human labelers. This feedback makes them valuable for ongoing improvement and model tuning.

LangSmith lets you choose the right retention for each trace, helping you balance cost and value.

Plus plans include 1 free small serverless deployment. If you spin up additional serverless or dedicated deployments, you’ll be charged on usage (resource time).

Uptime is the duration your deployment’s database is live and persisting state. Uptime will be tracked as soon as your deployment is live and ends when you shut it down. Dev agent deployments are typically short-lived (used during iteration, then deleted) – whereas Production agent deployments stay live and are updated via revisions (rather than being deleted).

Serverless deployments scale to zero when idle and are intended for testing, background agents, or internal agents. Dedicated deployments are always on and intended for production workloads that require autoscaling, backups, and high availability. We recommend using the dedicated deployment for any customer-facing agent.

Use Serverless for background or latency-tolerant agents (it scales to zero, so you pay only while it runs) and Dedicated for customer-facing agents that need consistent uptime. Start at Small and size up to Medium or Large as traffic grows, you can change size anytime. Full specs by size are in the docs.

Yes. Fleet pricing now includes underlying model costs. Fleet models are available in 3 performance tiers - Fast, Pro & Max.

Fleet includes built-in tools and supports connecting third-party tools via a remote MCP server. Third-party tools typically require you to authenticate with your account, and any fees or usage charges are billed by the third-party provider. More information on setting up tool keys in our docs.

Yes. Every Fleet run is automatically traced in LangSmith. These traces are attributed to the user’s account and count toward usage-based billing under the customer’s LangSmith plan.

Yes. Contact our sales team to learn more.

Fleet usage is measured in LangChain Compute Units (LCUs). LCU usage depends on the amount of work your Fleet agent performs, your selected model tier (Fast, Pro, Max) and the amount of content it processes & generates. Developer & Plus plans include a free usage allowance, shared across your organization and resets monthly:

• Developer: 5 LCUs included per organization each month. After the allowance is used, Fleet pauses new runs until the allowance resets or the organization upgrades to Plus.
• Plus: 25 LCUs included per organization each month. Additional usage is billed.

Fast, Pro, and Max are performance tiers, not fixed model names. Fleet selects and LangChain maintains the underlying models for each tier, based on specific task performance and cost optimization. Users select the tier based on the performance requirements for each workload.

• Fast prioritizes speed and cost efficiency for everyday work.
• Pro provides stronger reasoning for more complex work.
• Max provides the highest capability for the most demanding work and typically costs more and takes longer.

No. LangChain manages model-provider access for Fast, Pro, and Max. Your organization is charged for Fleet usage in LCUs.

Engine is metered in LCUs (LangChain Compute Units), which are consumed during each Engine run. 1 LCU costs $1.50.

An LCU represents a unit of work done by Engine. This includes the compute, infrastructure, and underlying model usage needed to analyze traces, identify issues, and propose fixes to your agent.

An Engine run can consume ~5-30 LCUs based on the volume of traces analyzed, agent complexity, the number and types of issues surfaced, and the suggested fixes.

These ranges are estimates. Actual usage depends on the amount of work Engine performs, including the volume of traces analyzed, the number of issues surfaced, the number of proposed fixes generated, and the complexity of your application.

Engine is scheduled to run once every 6 hours.

Yes. Engine pricing includes the underlying model usage required for analysis and fix generation.

You can view your organization’s LCU usage in your LangSmith billing console by visiting Settings > Billing. Learn more.

Plus plans: LCUs are an additional charge on top of your monthly license fee and any other product usage.

Enterprise plans: LCUs are charged against your committed LangChain credits.

Ask your Organization Admin to turn off Engine in the Settings page by following instructions here. Once Engine is disabled, it will stop running on its regular schedule and will no longer consume LCUs.

Sandboxes are designed as ephemeral compute environments and are billed per second. LangSmith Sandboxes have configurable TTLs so inactive sandboxes shut down automatically.

Not currently. Sandboxes are available for SaaS Cloud deployments.