v0.21.0A security release: what an agent or a client sends Iris is bounded before Iris reads it, and a tool argument can narrow what the operator allows but never widen it→

Comparison · Observability

Iris vs LangSmith

MCP-Native Eval vs LangChain Ecosystem Tracing.

TL;DR

Iris is an MCP server your agent discovers and uses on connect — one config block, no SDK, one SQLite file, every rule's precision and recall published at iris-eval.com/proof. LangSmith — from its own pages, read 2026-10-01: LangSmith is LangChain's observability and evaluation platform with integrations for many frameworks and providers, LLM-as-judge evaluation, and SOC 2 Type II, GDPR and HIPAA compliance; self-hosting is an Enterprise-plan add-on. If you need advanced evaluation workflows, Insights auto-clustering, or enterprise compliance today, LangSmith is the more established platform. (source)

Iris grades what an agent did with its tools — the trace, the answer, the cost — not whether an MCP server honours its own contract; a server test harness answers that question, and Iris runs beside it. For the method, see the agent eval guide.

Feature comparison

Side by side.

13 features, the same 13 on every comparison. Every LangSmith cell links the page it was read from and the date; where the sentence it was read from was found verbatim on a plain download of that page (checked 2026-09-22), the link carries it as its title, and where it was not, the cell says so. The highlighted cells are Iris's own call on which side is stronger for a team running MCP agents — 4 to Iris, 3 to LangSmith — not a measurement.

FeatureIrisLangSmith
Integration methodOne block in the MCP config, no code — the agent discovers Iris and its tools on connectSDK imports + @traceable decorators, or an OpenTelemetry exporterLangSmith's page · read 2026-10-01
Self-hostingOne process, one SQLite file; Docker image with a health checkEnterprise-only, license key requiredLangSmith's page · read 2026-10-01
Where it runsNothing in the agent's process — Iris is a separate server the agent callsAsync tracing via SDK in your processLangSmith's page · read 2026-10-01
Evaluation26 built-in deterministic rules and 9 custom-rule types, in-process; 7 judge templates on a key you supply; every rule's precision and recall publishedLLM-as-Judge + human review workflowsLangSmith's page · read 2026-10-01
Cost trackingPer-trace USD cost and tokens; a cost spike judged against the agent's own historyToken counts and computed cost per LLM call and per trace; latency per runLangSmith's page · read 2026-10-01
MCP supportProtocol-native — Iris is an MCP server with 12 tools; OTLP traces inHosted Remote MCP reads runs, prompts, datasets, experiments (run_experiment is docs-only); the standalone server is deprecated; Agent Server exposes MCP + A2ALangSmith's page · read 2026-10-01
LicenseMIT, the whole packageProprietary platform (SDK is MIT)LangSmith's page · read 2026-10-01
OwnershipIndependent and founder-ledBuilt and operated by LangChain, Inc. (company founded early 2023; venture-backed)LangSmith's page · read 2026-10-01
DashboardA local dashboard on its own port: traces, moments, regressions, five viewsAuto-clustering, pattern detection, custom dashboardsLangSmith's page · read 2026-10-01
Framework supportAny MCP client (2 verified, 8 claimed — see /clients); OTLP/HTTP from anything elseLangChain, LangGraph, OpenAI, Anthropic, Vercel AI SDK, CrewAI, Pydantic AI, OpenTelemetry + moreLangSmith's page · read 2026-10-01
Prompt managementNot includedVersioned prompts (a commit per save, movable tags) and a PlaygroundLangSmith's page · read 2026-10-01
Enterprise and complianceSelf-hosted. Nothing leaves your machine unless you turn one of these on: an OpenTelemetry endpoint (IRIS_OTEL_ENDPOINT), which exports traces to the collector you name; the LLM judge with your own key, which sends the text it judges to that provider, and whose citation check fetches the pages an output cites; or a webhook, which posts ids, the verdict and rule names, never the text, to the address you set. No compliance certification is claimed before it is heldSOC 2 Type II, HIPAA, GDPR; custom SSO, RBAC/ABAC, self-hosted/hybrid deployment, support SLALangSmith's page · read 2026-10-01
Cost to runFree — MIT, one process on your machine; the only spend is a judge call on a key you supply, when you opt inDeveloper $0 per seat with 5k base traces a month, then pay as you go; Plus $39 per seat per month with 10k base traces; Enterprise customLangSmith's page · read 2026-10-01

Decision guide

Which one fits your stack?

When to choose Iris

  • You are building with MCP-compatible agents and want the integration to be one config block
  • You want the evaluation to be deterministic and local — no model call, nothing leaving the machine
  • You want to read what each rule is worth before you trust it: every rule's precision and recall is published
  • You want self-hosting to be one process and one file
  • You want a fully permissive MIT license on the whole package

When to choose LangSmith

  • You need LLM-as-Judge evaluation with human review workflows source
  • You need enterprise compliance and a support SLA today source
  • You're heavily invested in the LangChain ecosystem source
  • You need auto-clustering and pattern detection across traces source
  • You need broad SDK support across many non-MCP frameworks source

FAQ

The questions buyers ask.

What is the difference between Iris and LangSmith?
Iris is an MCP-native agent eval server that needs no SDK. Add the config block, restart your client, and every session lists Iris's tools on connect. Iris never intercepts: it runs when your agent calls one of its tools, when a host hook or `iris-eval ingest` hands it a trace, or when you POST one to its HTTP API. LangSmith is LangChain's observability and evaluation platform, instrumented with @traceable decorators, SDK wrappers or OpenTelemetry and integrated with many frameworks; self-hosting is an Enterprise-plan add-on. LangSmith provides LLM-as-judge and human evaluation workflows, Insights auto-clustering, and SOC 2 Type II, GDPR and HIPAA compliance. source
Is Iris better than LangSmith for MCP agent evaluation?
LangSmith requires @traceable decorators, SDK wrappers or an OpenTelemetry exporter but offers LLM-as-judge evaluation, human review through annotation queues, and native LangChain and LangGraph integration. source
How much does LangSmith cost to run?
Iris: Free — MIT, one process on your machine; the only spend is a judge call on a key you supply, when you opt in. LangSmith, from its own page read 2026-10-01: Developer $0 per seat with 5k base traces a month, then pay as you go; Plus $39 per seat per month with 10k base traces; Enterprise custom. source
Can I self-host LangSmith?
Iris: One process, one SQLite file; Docker image with a health check. LangSmith, from its own page read 2026-10-01: Enterprise-only, license key required. source
Does LangSmith work with MCP agents?
Iris: Protocol-native — Iris is an MCP server with 12 tools; OTLP traces in. LangSmith, from its own page read 2026-10-01: Hosted Remote MCP reads runs, prompts, datasets, experiments (run_experiment is docs-only); the standalone server is deprecated; Agent Server exposes MCP + A2A. source

Sources

Every LangSmith statement on this page was read from one of these pages on the date shown. The file behind this page is website/src/lib/compare/langsmith.json.

Last verified: 2026-10-01. This comparison is based on publicly available documentation and may not reflect recent changes to LangSmith. We aim to keep this page accurate and fair.

See something outdated or incorrect? Report an inaccuracy — we review and update within 48 hours.

Ready to see what your agents are doing?

Add Iris to your MCP config. First trace in 60 seconds. No SDK, no signup, no infrastructure.

Install IrisTry the playgroundEvery comparison