Reasoning infrastructure, now in general availability

Give your product
a mind that reasons.

Meridian is the orchestration layer for production AI agents — plan, tool-use, memory, and evaluation, unified into one runtime your team already trusts.

412MAgent runs / mo
99.98%Runtime uptime
11msOrchestration overhead
SOC 2Type II certified
Trusted by engineering teams building the next generation of AI products
NorthwindHalcyon LabsArcadiaFerroviaKestrelBlackwellVantage PointSolvent NorthwindHalcyon LabsArcadiaFerroviaKestrelBlackwellVantage PointSolvent
Platform

Everything between the model and the product.

Meridian handles the parts of an AI system that have nothing to do with the model itself — so your team builds features, not infrastructure.

01

Agent orchestration

Compose planning, tool-calling, and multi-step reasoning into durable workflows that recover from failure automatically.

02

Persistent memory

Give every agent long-term, queryable context that survives across sessions without you managing a vector store.

03

Tool & function routing

Register any API, database, or internal service once — Meridian handles auth, retries, and schema validation.

04

Evaluation & tracing

See every reasoning step your agents take, replay failures, and grade output quality before it reaches production.

05

Model-agnostic routing

Swap between model providers per task, with automatic fallback when latency or quality drops below threshold.

06

Governance by default

Every action is scoped, logged, and reversible — built for teams that need to answer for what their agents do.

Live in your workspace

Watch an agent think, in real time.

The Meridian console shows every plan, tool call, and retry as it happens — not a log file you reconstruct after the fact.

Runtime
Traces
Evaluations
Memory
2,481 ACTIVE RUNS / LAST 24H
+18.4% vs. prior week
Plan generated
0.4s
Tool: search_orders
1.1s
Tool: refund_api
0.9s
Eval: passed
✓ 98%
How it works

From prompt to production, in three steps.

01

Define the agent's world

Describe its goals, register its tools, and set the boundaries of what it's allowed to do. Meridian generates the orchestration graph for you.

02

Run it against real traffic

Ship to a staging runtime that mirrors production exactly, with full tracing on every plan, tool call, and retry.

03

Promote with confidence

Evaluation gates block regressions automatically. One click promotes a version once it clears your quality bar.

Under the hood

A reasoning graph, not a call stack.

Every decision an agent makes is a node Meridian can inspect, replay, and score — turning a black box into an audit trail.

Live inference graph

2,140 active reasoning nodes across 6 production agents

Pricing

Simple pricing that scales with usage.

Monthly
Yearly Save 20%
Starter
For small teams shipping their first agent.
49/ mo per seat
  • Up to 3 agents
  • 50K runs / month
  • Basic tracing
  • Community support
Start Free Trial
Growth
For teams running agents in production.
199/ mo per seat
  • Unlimited agents
  • 2M runs / month
  • Full tracing & evals
  • Priority support
Start Free Trial
Enterprise
For organizations with custom governance needs.
Custom
  • Dedicated infrastructure
  • SSO, SCIM & audit logs
  • Custom data residency
  • Dedicated solutions engineer
Talk to Sales
ROI calculator

Estimate what Meridian is worth to you.

Estimated additional leads / mo312
Projected new revenue / mo$374,400
Estimated annual ROI18.4×
Customer stories

Teams that moved from prototype to production.

"We replaced 4,000 lines of orchestration glue with Meridian in a week. Our incident rate on agent workflows dropped by 70%."

Dana OkaforVP Engineering, Halcyon Labs

"The tracing view is the reason we passed our SOC 2 audit on the first attempt. Every action is already logged for us."

Marcus WebbHead of Platform, Arcadia

"We went from one support-triage agent to eleven production agents without adding a single infrastructure engineer."

Priya RamaswamyCTO, Ferrovia
Security & compliance

Built for the review before the rollout.

Every request is encrypted in transit and at rest. Every agent action is scoped to explicit permissions and logged for audit — no exceptions, no back doors.

SOC 2 Type II
GDPR
AES-256 at rest
TLS 1.3 in transit
Private VPC options
FAQ

Questions, answered.

Which model providers does Meridian support?+

Meridian is model-agnostic and ships with adapters for every major provider. You can route by task, cost, or latency, and fall back automatically if one provider degrades.

How long does implementation take?+

Most teams have a first agent running in staging within a day. Full production rollout with evaluation gates typically takes one to two weeks depending on the number of tools you're integrating.

Can we self-host Meridian?+

Yes. Enterprise plans include a self-hosted runtime that deploys inside your own VPC, with the same tracing and evaluation tooling as the managed offering.

What happens if an agent fails mid-task?+

Every run is checkpointed. Meridian retries from the last successful step, and if it can't recover, it escalates to a human reviewer with the full trace attached.

Is there a free trial?+

Yes — 14 days, full feature access, no credit card required. You can also talk to sales for a guided proof-of-concept scoped to your use case.

Ready to give your product
a mind of its own?

Start free, or talk to our team about a guided rollout.