EDGE RUNTIME 4.2 — AVAILABLE NOW

Deploy at the edge, execute in microseconds

Meridian pushes your compute to 280+ edge locations. Zero cold starts. Automatic scaling. Ship functions that run closer to your users than any cloud provider can offer.

//EDGE_LATENCY

1.2ms
P99: 3.8msNodes: 284

//THROUGHPUT

2.4M req/s
CPU: 12%MEM: 34%
280+
Edge Locations
1.2ms
Median Latency
99.99%
Uptime SLA
14B
Requests / Month
[01] Platform

Infrastructure that disappears so your code can ship

λ

Edge Functions

Deploy serverless functions to 280+ locations. V8 isolates boot in under 5ms. No containers. No cold starts.

KV Store

Eventually-consistent key-value storage at every edge node. Read latency under 1ms worldwide. Write propagation in 60s.

Edge Cache

Automatic response caching with fine-grained invalidation. Cache API compatible. Stale-while-revalidate built in.

Streams

Real-time event streaming between edge functions. Pub/sub with at-least-once delivery and ordered partitions.

Observability

Traces, metrics, and structured logs from every invocation. OpenTelemetry native. Tail logs in real-time from CLI.

Zero-Trust Auth

mTLS, JWT validation, and IP access policies at the edge. Auth runs before your code, not inside it.

[02] Developer Experience

From commit to edge in 40 seconds

meridian-cli v4.2.0
$ meridian deploy ./api --region=auto
Bundling api/handler.ts ... 12.4 KB
Uploading to 284 edge nodes ...
Propagation complete: 38.2s
✓ Live at api.meridian.run/v1
✓ Latency P50: 1.1ms | P99: 3.4ms
✓ 284 regions active
 
$ meridian logs --tail --function=api
[12:04:33.221] GET /v1/users 200 1.2ms — SFO
[12:04:33.443] GET /v1/users 200 0.9ms — AMS
[12:04:33.891] POST /v1/auth 201 2.1ms — NRT
[12:04:34.102] GET /v1/users 200 1.0ms — IAD