Skip to content
Talk to Us

Blocks.ai publishes this comparison and sells in this market. Every claim about another product links to a public source.

Gemini Enterprise Agent Platform vs LangSmith

Gemini Enterprise Agent Platform

Google Cloud platform to build, deploy, govern, and optimize AI agents

LangSmith

Platform for observing, evaluating, and deploying agents

Short answer

Both are agent platforms: Gemini Enterprise Agent Platform is Google Cloud’s platform to build, deploy, govern, and optimize agents, and LangChain calls LangSmith its platform for observing, evaluating, and deploying them.⁠Source 1, Source 2, Source 3 Agent Platform runs on Google Cloud and applies policy at Agent Gateway; LangSmith runs in LangChain’s cloud or, on Enterprise, in your own infrastructure.⁠Source 2, Source 4, Source 5, Source 6

By Blocks.ai teamLast checked

Where each one sits

Six layers of running AI agents at a company, and what each product’s own public sources say it covers.

Public sources · checked 2 October 2026
  • Offered
  • Not publicly documented

Gemini Enterprise Agent Platform

  • Build agents: OfferedAgent Studio low-code canvas⁠Source 1
  • Host and run agents: OfferedAgent Runtime⁠Source 4
  • Identity and access: OfferedSPIFFE-based Agent Identity⁠Source 7
  • Registry and governance: OfferedAgent Registry⁠Source 8
  • Traffic between agents, tools, and models: OfferedAgent Gateway⁠Source 4
  • Agents across organizations: OfferedGateway calls to outside agents⁠Source 4

LangSmith

  • Build agents: OfferedFleet agent builder⁠Source 9
  • Host and run agents: OfferedLangSmith Deployment runtime⁠Source 10
  • Identity and access: OfferedFleet per-agent permissions⁠Source 11
  • Registry and governance: OfferedCentralized registry in Deployment⁠Source 12
  • Traffic between agents, tools, and models: OfferedFleet forwards MCP tool calls⁠Source 13
  • Agents across organizations: Not publicly documented

At a glance

TopicGemini Enterprise Agent PlatformLangSmith
Where agents runGoogle Cloud runtimes such as Agent Runtime, which keeps agent data at rest in the selected location.⁠Source 4, Source 14 Agents hosted elsewhere can be registered manually.⁠Source 15Managed by LangChain on AWS and GCP (Plus plan or above).⁠Source 5, Source 10 On Enterprise: BYOC on AWS, self-hosted in your Kubernetes cluster, or hybrid.⁠Source 6, Source 10, Source 16, Source 17 Agents elsewhere can send traces.⁠Source 6
Agent identity and accessAgents on supported services can each have a SPIFFE-based identity.⁠Source 7 By default, Agent Gateway blocks agents’ outbound calls unless an IAM access policy grants them.⁠Source 4Hosted deployments check a LangSmith API key by default, or your own auth handler.⁠Source 18 Fleet can share each agent with Clone, Run, or Edit permissions.⁠Source 11
Building agentsAgent Studio, a low-code canvas, and the open-source Agent Development Kit.⁠Source 1, Source 19 Agent Garden has prebuilt agents and templates.⁠Source 1Fleet builds agents without code, from a description or a template.⁠Source 9 Code-first agents can be deployed with the LangGraph CLI.⁠Source 10
Pricing modelBy use: Agent Runtime compute is $0.085 per vCPU-hour; Agent Gateway egress is $0.085 per 15,000 requests.⁠Source 20 Agent Registry is free; skill scanning is billed from January 2027.⁠Source 21Seats and usage: one free Developer seat, Plus at $39 per seat a month, and usage metered in LangChain Standard Units at $1 each.⁠Source 17 Enterprise: contact sales.⁠Source 17
Generally availableAgent Registry and Agent Gateway since 18 June 2026.⁠Source 22 Agent Identity is generally available; the Agent Identity API is in preview.⁠Source 22LangChain announced GA of LangGraph Platform, now Deployment, on 14 May 2025, and of Agent Builder, now Fleet, on 13 January 2026.⁠Source 12, Source 23, Source 24, Source 25 Self-hosted Fleet is in beta.⁠Source 9

What each one is

Gemini Enterprise Agent Platform

Gemini Enterprise Agent Platform is Google Cloud’s platform to build, deploy, and govern agents, an evolution of Vertex AI.⁠Source 1, Source 2 Agent Registry catalogs agents, MCP servers, and tools; Agent Identity can give agents on supported services an identity; and Agent Gateway checks traffic against policy.⁠Source 4, Source 7, Source 8

LangSmith

LangChain describes LangSmith as a framework-agnostic platform for observing, evaluating, and deploying agents.⁠Source 3 LangSmith Deployment runs agents in production, and LangSmith Fleet creates and manages agents without code.⁠Source 9, Source 10 Agents hosted elsewhere can send traces to LangSmith.⁠Source 6 Role-based access control and audit logs are Enterprise features.⁠Source 26, Source 27

The differences that matter

  1. Controlling agent access

    Gemini Enterprise Agent Platform

    By default, Agent Gateway blocks agents’ outbound calls unless an IAM policy grants them, and can limit which clients reach Agent Runtime agents.⁠Source 4

    LangSmith

    Hosted deployments check a LangSmith API key by default, or your own auth handler; self-hosted Agent Servers have no default authentication.⁠Source 18

    Model Armor filters attached to Agent Gateway can scan prompts and tool responses; LangSmith Fleet agents can pause for human approval before specific actions.⁠Source 4, Source 28

  2. Where it runs

    Gemini Enterprise Agent Platform

    On Google Cloud: Agent Gateway is managed and regional, and agent data at rest stays in a supported location you pick.⁠Source 14, Source 29

    LangSmith

    LangSmith Cloud runs on GCP or AWS; on Enterprise, LangSmith can be self-hosted, hybrid, or BYOC, generally available on AWS.⁠Source 5, Source 6, Source 16, Source 17

    Self-hosting Agent Registry or Agent Gateway is not publicly documented. For air-gapped use, LangSmith’s docs have you host Deployment’s control plane and Agent Servers yourself.⁠Source 10

  3. Tracing and audit logs

    Gemini Enterprise Agent Platform

    Agent Platform provides metrics, traces, and logs, with traces from agents’ OpenTelemetry data.⁠Source 30 Registry admin changes are audit-logged.⁠Source 31

    LangSmith

    Each deployment gets a tracing project, and LangSmith accepts OpenTelemetry traces from any compatible app.⁠Source 32, Source 33 On Enterprise, audit logs record administrative and configuration actions.⁠Source 27

    LangSmith audit logs, kept up to 400 days, can be pulled hourly into a SIEM; Cloud Logging can route logs to Pub/Sub for third-party tools.⁠Source 27, Source 34

For security teams

What a security review asks, answered from each vendor’s public documentation.

TopicGemini Enterprise Agent PlatformLangSmith
Network exposureAgent Gateway manages mTLS connections and applies policy checks to traffic; its inbound mode supports only Agent Runtime agents.⁠Source 4Agent Server exposes A2A and MCP endpoints.⁠Source 35, Source 36 Enterprise: private access to LangSmith Cloud via AWS PrivateLink or Private Service Connect.⁠Source 5
IdentityAgents on supported services can have a SPIFFE-based identity and authenticate as themselves or for an end user.⁠Source 7Fleet agents can use fixed credentials or the user’s own.⁠Source 37 Cloud Enterprise customers can configure SAML SSO and SCIM.⁠Source 38
Access changes and revocationDeny rules override allow rules.⁠Source 39 Deleting an agent leaves its IAM bindings as inactive grants, to be removed by hand.⁠Source 7Fleet sharing can be revoked at any time.⁠Source 28 Admins can deactivate any member’s access token, effective within a minute.⁠Source 40
Audit trailRegistry admin changes are audit-logged; other calls only if Data Access logs are on.⁠Source 31, Source 41 Gateway logs record access requests.⁠Source 42Enterprise audit logs record administrative and configuration actions, kept up to 400 days and returned in OCSF format.⁠Source 27
ComplianceListed in scope for ISO 27001, SOC 1, 2, and 3, and PCI DSS, and in Google Cloud’s HIPAA BAA.⁠Source 43, Source 44LangChain’s Trust Center lists SOC 2 Type II, ISO 27001:2022, GDPR, and HIPAA for LangSmith.⁠Source 45

Full comparison

18 criteria in five groups. Every cell links to its source, or says no public source answers it.

Gemini Enterprise Agent Platform and LangSmith compared on 18 criteria
Gemini Enterprise Agent PlatformLangSmith
What it is
What it is and who it’s forGoogle Cloud platform to build, deploy, govern, and optimize AI agents, described as an evolution of Vertex AI, for developers and technical teams.⁠Source 1, Source 2LangChain calls it a framework-agnostic platform for observing, evaluating, and deploying agents.⁠Source 3 It positions its no-code part, Fleet, for non-technical teams.⁠Source 3, Source 9
MaturityGoogle announced it on 22 April 2026.⁠Source 46 Registry and Gateway GA 18 June 2026.⁠Source 22 Agent Identity is generally available; the Agent Identity API is in preview.⁠Source 22LangChain announced GA of LangGraph Platform (now Deployment) on 14 May 2025 and Agent Builder (now Fleet) on 13 January 2026.⁠Source 12, Source 23, Source 24, Source 25 Self-hosted Fleet is in beta.⁠Source 9
Control
Agent registry and discoveryAgent Registry: a per-project catalog of agents, MCP servers, and tools.⁠Source 8, Source 47 Agents on supported runtimes can be registered automatically, others manually.⁠Source 48, Source 49LangChain says Deployment manages agents in a centralized registry with versioning and rollbacks.⁠Source 12, Source 50 Listing agents hosted elsewhere is not publicly documented.
Identity and access controlSPIFFE-based identity on supported services.⁠Source 7 By default, Agent Gateway blocks outbound calls without an IAM grant and can limit who reaches Runtime agents.⁠Source 4Hosted deployments check a LangSmith API key by default, or your auth handler.⁠Source 18 Fleet: per-agent Clone, Run, and Edit.⁠Source 11 Enterprise: RBAC; on Cloud, SAML SSO.⁠Source 26, Source 38
Ownership, policy, and revocationAllow and deny access policies (deny overrides allow), enforce and dry-run modes, and Model Armor filters.⁠Source 4, Source 39 An agent owner field is not publicly documented.Fleet agent cards show the owner, sharing can be revoked at any time, and pausing an agent’s channels pauses it.⁠Source 25, Source 28, Source 51 Enterprise: custom roles.⁠Source 26
Audit log and observabilityRegistry admin changes are audit-logged; reads only if Data Access logs are on.⁠Source 31, Source 41 Gateway traffic shows in Cloud Logging.⁠Source 4 Traces need agents’ OpenTelemetry data.⁠Source 30A tracing project per deployment, and OpenTelemetry traces accepted.⁠Source 32, Source 33 Enterprise audit logs of administrative and configuration actions, kept up to 400 days.⁠Source 27
Connection
How agents connectAgent Gateway: inbound for Agent Runtime agents, outbound for them and the Gemini Enterprise app.⁠Source 4 Outside agents are registered by endpoint or agent card.⁠Source 15Agent Server has A2A and MCP endpoints.⁠Source 35, Source 36 For deployment updates, the data plane polls the control plane, which never connects to it directly.⁠Source 32
Agents across organizationsRuntime agents can call outside agents via Agent Gateway.⁠Source 4 A cross-organization access model is not publicly documented.Fleet can share agents in your workspace; Agent Server agents can talk to other A2A agents.⁠Source 28, Source 35 A cross-organization access model is not publicly documented.
Protocol supportAgent Registry catalogs MCP servers and A2A agents; Agent Gateway carries MCP and A2A traffic.⁠Source 4, Source 49, Source 52 A2A agents on Agent Runtime are in preview.⁠Source 53Agent Server: A2A over JSON-RPC only, with documented gaps; MCP over Streamable HTTP.⁠Source 35, Source 36 Fleet discovers tools from remote MCP servers.⁠Source 13
Frameworks, models, and clouds supportedBuild with the Agent Development Kit or any other open-source framework, using Gemini or other Model Garden models.⁠Source 19 Outside agents can be registered manually.⁠Source 15Deployment runs Google ADK, CrewAI, AutoGen, and other agents via a wrapper package or the LangGraph Functional API.⁠Source 54 Agents hosted elsewhere can send traces.⁠Source 6
Operations
Deployment options and data residencyAgent Gateway is managed and regional.⁠Source 29 Agent data at rest stays in a supported location you pick.⁠Source 14 Self-hosting the registry or gateway: not publicly documented.Cloud on GCP (US, EU, APAC) or AWS (US).⁠Source 5 On Enterprise: BYOC (GA on AWS), self-hosted, or hybrid.⁠Source 6, Source 16, Source 17 Moving between Cloud regions isn’t supported.⁠Source 55
Compliance attestationsListed for ISO 27001, 27017, and 27018, SOC 1, 2, and 3, and PCI DSS, and in the HIPAA BAA.⁠Source 43, Source 44 Access Transparency isn’t available for Agent Registry.⁠Source 22Trust Center lists SOC 2 Type II, ISO 27001:2022, GDPR, and HIPAA.⁠Source 45 On the SaaS platform, data is encrypted with AES-256 at rest and TLS 1.2+ in transit.⁠Source 56
Support and SLAGoogle Cloud support packages, some with 24/7 coverage.⁠Source 57 An uptime SLA naming Agent Registry, Gateway, Identity, or Runtime is not publicly documented.Base, Standard, and Premium support plans.⁠Source 58 Reasonable commercial efforts toward 99.5% API uptime for SaaS, measured quarterly.⁠Source 58
Time and effort to get runningA Google Cloud project with the Agent Registry API enabled, plus the Identity-Aware Proxy API for gateway policy.⁠Source 47 Google recommends dry-run mode in staging.⁠Source 39Cloud deployment needs a Plus plan or above and an API key.⁠Source 59 Self-hosting the Deployment control plane needs Kubernetes and Enterprise.⁠Source 10
Pricing model and public pricesAgent Registry is free; skill scanning is billed from January 2027.⁠Source 21 Gateway egress: $0.085 per 15,000 requests; runtime: $0.085 a vCPU-hour.⁠Source 20 Monthly free tier.⁠Source 20Developer: one free seat.⁠Source 17 Plus: $39 per seat a month, with one free small serverless deployment.⁠Source 17 Enterprise: contact sales.⁠Source 17 Usage metered in LSUs, $1 each.⁠Source 17
Building
Agent building toolsAgent Studio, a low-code canvas; the Agent Development Kit; and Agent Garden prebuilt agents and templates.⁠Source 1 A Managed Agents API is in preview.⁠Source 1Fleet builds agents from a description or a template, without code.⁠Source 9 Code-first agents can be deployed with the LangGraph CLI and app templates.⁠Source 10
Model accessGoogle says Model Garden offers more than 200 models, including Gemini, third-party models such as Anthropic’s Claude, and open-source models.⁠Source 1, Source 46LangChain says closed or open models work.⁠Source 3 Fleet’s managed tiers (Fast, Pro, Max) may change models over time; custom models aren’t in that picker.⁠Source 60
Integrations and ecosystemAgents in Agent Registry can be made available to users of the Gemini Enterprise app.⁠Source 29 Google’s own remote MCP servers are registered automatically.⁠Source 52Fleet has built-in tools for Gmail, Slack, Google Calendar, GitHub, and others.⁠Source 60 LangChain says Fleet agents work in Slack, Teams, and Gmail.⁠Source 61

Which to choose

Choose Gemini Enterprise Agent Platform if

  • You want to build, deploy, and govern agents on one Google Cloud platform, with Agent Studio and the open-source Agent Development Kit.⁠Source 1, Source 2, Source 19
  • You want outbound gateway calls blocked by default unless an IAM policy allows them, with Model Armor scanning prompts and tool responses.⁠Source 4
  • You want a catalog of agents, MCP servers, and tools that can include agents hosted outside Google Cloud, registered manually.⁠Source 8, Source 15
  • You want a wide choice of models: Google says Model Garden offers more than 200, including Anthropic’s Claude.⁠Source 46

Choose LangSmith if

  • You need the platform itself in your own infrastructure: on Enterprise, LangSmith can be self-hosted, which LangChain describes for air-gapped use.⁠Source 6, Source 10
  • You want tracing for agents from LangChain or other frameworks, including agents you host yourself, through integrations or OpenTelemetry.⁠Source 6, Source 33, Source 62
  • You want people to build agents without code and share them with Clone, Run, or Edit permissions.⁠Source 9, Source 11
  • You want Fleet agents to pause for human approval before specific actions, with review from one place.⁠Source 28

Questions buyers ask

Does LangSmith work only with LangChain agents?

No. LangChain calls LangSmith framework-agnostic.⁠Source 3 LangSmith Deployment runs agents built with frameworks such as CrewAI and AutoGen, and agents hosted elsewhere can send traces to LangSmith.⁠Source 6, Source 54

How is each one priced?

Agent Registry is free; skill scanning is billed from January 2027.⁠Source 21 Agent Runtime is $0.085 per vCPU-hour plus memory.⁠Source 20 LangSmith: one free Developer seat, Plus at $39 per seat a month, and usage metered in LangChain Standard Units, $1 each.⁠Source 17

Do they support MCP and A2A?

Yes. Agent Registry catalogs MCP servers and A2A agents, Agent Gateway carries both, and hosting A2A agents on Agent Runtime is in preview.⁠Source 4, Source 49, Source 52, Source 53 LangSmith’s Agent Server has MCP endpoints and A2A ones, with documented gaps, and Fleet discovers tools from remote MCP servers.⁠Source 13, Source 35, Source 36

Can either one connect agents across organizations?

Agent Runtime agents can call outside agents through Agent Gateway.⁠Source 4 Google says preview A2A agents handling their own auth can be exposed to untrusted callers.⁠Source 63 LangSmith’s Agent Server agents can talk to A2A agents.⁠Source 35 Neither publicly documents a cross-organization access model.

How we compare

Read the full method

Every claim on this page links to a public source. Where none answers a question, the page says so.

We re-check every fact at least every 90 days. This page was last checked .

Something wrong or out of date? Tell us and we’ll correct it.

Sources

70 public sources, each with the date we checked it. Every one opens in a new tab.

  1. Source 1: Agent Platform overview Google · checked Back:abcdefghij

  2. Source 2: Gemini Enterprise Agent Platform (formerly Vertex AI) Google · checked Back:abcde

  3. Source 3: LangSmith: The Agent Engineering Platform LangChain · checked Back:abcdef

  4. Source 4: Agent Gateway overview Google · checked Back:abcdefghijklmnopqrs

  5. Source 5: Cloud (SaaS) LangChain · checked Back:abcde

  6. Source 6: Set up LangSmith LangChain · checked Back:abcdefghij

  7. Source 7: Agent Identity overview Google · checked Back:abcdef

  8. Source 8: Agent Registry overview Google · checked Back:abcd

  9. Source 9: No-code agents with LangSmith Fleet LangChain · checked Back:abcdefgh

  10. Source 10: LangSmith Deployment LangChain · checked Back:abcdefghi

  11. Source 11: Agent platform comparison LangChain · checked Back:abcd

  12. Source 12: LangSmith Deployment LangChain · checked Back:abcd

  13. Source 13: Remote MCP servers LangChain · checked Back:abc

  14. Source 14: Supported locations for agents in Agent Platform Google · checked Back:abc

  15. Source 15: Use manual registration Google · checked Back:abcde

  16. Source 16: Bring Your Own Cloud (BYOC) LangChain · checked Back:abc

  17. Source 17: LangSmith Plans and Pricing LangChain · checked Back:abcdefghij

  18. Source 18: Authentication & access control LangChain · checked Back:abc

  19. Source 19: Build with Gemini Enterprise Agent Platform Google · checked Back:abcd

  20. Source 20: Gemini Enterprise Agent Platform pricing Google · checked Back:abcd

  21. Source 21: Agent Registry pricing Google · checked Back:abc

  22. Source 22: Gemini Enterprise Agent Platform release notes Google · checked Back:abcde

  23. Source 23: LangGraph Platform is now Generally Available: Deploy & manage long-running, stateful Agents LangChain · checked Back:ab

  24. Source 24: Now GA: LangSmith Agent Builder LangChain · checked Back:ab

  25. Source 25: LangSmith Fleet changelog LangChain · checked Back:abc

  26. Source 26: Role-based access control LangChain · checked Back:abc

  27. Source 27: Audit logs LangChain · checked Back:abcde

  28. Source 28: Access & oversight LangChain · checked Back:abcde

  29. Source 29: Import A2A agents from Agent Registry Google · checked Back:abc

  30. Source 30: Observability overview Google · checked Back:ab

  31. Source 31: Agent Registry audit logging Google · checked Back:abc

  32. Source 32: LangSmith control plane LangChain · checked Back:abc

  33. Source 33: Trace with OpenTelemetry LangChain · checked Back:abc

  34. Source 34: Route log entries Google · checked Back to text

  35. Source 35: A2A endpoint in Agent Server LangChain · checked Back:abcdefg

  36. Source 36: MCP endpoint in Agent Server LangChain · checked Back:abcd

  37. Source 37: Agent identity LangChain · checked Back to text

  38. Source 38: Authentication methods LangChain · checked Back:ab

  39. Source 39: IAM Access policies overview Google · checked Back:abc

  40. Source 40: Overview (administration) LangChain · checked Back to text

  41. Source 41: Cloud Audit Logs overview Google · checked Back:ab

  42. Source 42: Monitor traffic through Agent Gateway Google · checked Back to text

  43. Source 43: Google Cloud Platform Services in Scope by Compliance Program Google · checked Back:ab

  44. Source 44: HIPAA compliance on Google Cloud Google · checked Back:ab

  45. Source 45: Trust Center - LangChain LangChain · checked Back:ab

  46. Source 46: Introducing Gemini Enterprise Agent Platform, powering the next wave of agents Google · checked Back:abc

  47. Source 47: Set up Agent Registry Google · checked Back:ab

  48. Source 48: Use automatic registration Google · checked Back to text

  49. Source 49: Register agents Google · checked Back:abc

  50. Source 50: Assistants LangChain · checked Back to text

  51. Source 51: Manage agent settings LangChain · checked Back to text

  52. Source 52: Register MCP servers Google · checked Back:abc

  53. Source 53: Create an Agent2Agent agent Google · checked Back:ab

  54. Source 54: Deploy other frameworks LangChain · checked Back:abc

  55. Source 55: Regions FAQ LangChain · checked Back to text

  56. Source 56: LangSmith shared responsibility model LangChain · checked Back to text

  57. Source 57: Getting help for agents Google · checked Back to text

  58. Source 58: Support Plans LangChain · checked Back:ab

  59. Source 59: Deploy your app to cloud LangChain · checked Back to text

  60. Source 60: Essentials LangChain · checked Back:ab

  61. Source 61: LangSmith Fleet LangChain · checked Back to text

  62. Source 62: LangSmith Observability LangChain · checked Back to text

  63. Source 63: Share an agent Google · checked Back to text

  64. Source 64: Why Blocks? Blocks.ai · checked Back to text

  65. Source 65: What is Blocks? Blocks.ai · checked Back to text

  66. Source 66: Your company's private network Blocks.ai · checked Back to text

  67. Source 67: Network requirements Blocks.ai · checked Back to text

  68. Source 68: Solutions: Agent sprawl Blocks.ai · checked Back to text

  69. Source 69: Solutions: Partner networks Blocks.ai · checked Back to text

  70. Source 70: Pricing Blocks.ai · checked Back to text