📋 Logs SDK · Layer L502

Distributed tracing and structured logging for every AI call.

Every request traced from API entry to model response. Trees of spans with timestamps, latencies, model selections, costs and outcomes. Purpose-built for AI.

Live £19/moor £190/yr — save 17%
£19/moStarting price
p50/p95/p99Latency percentiles
78%Earnings back
FreeTier available
Distributed tracingAI-nativeCost attributionOTel export78% backFree tier

How it works

Logs SDK in three steps

Production-grade and live on api.forcedream.ai. One Bearer token, zero extra setup.

  1. 01 🔍

    Trace

    Every /v1/route, /v1/agents and /v1/memory call auto-generates a trace. No instrumentation required.

    Auto-generated
  2. 02 📊

    Analyse

    Nested span trees show latency at every step — model selection, memory recall, safety checks. Full cost attribution per call.

    GET /v1/logs?limit=10
  3. 03 🔔

    Alert

    Set thresholds — error rate spikes, latency regressions, revenue drops — and receive webhook or email alerts.

    Pro feature

What's included

Everything you need, nothing you don't

Distributed tracing — full request lifecycle visibility
Span tracking — nested timing for every operation
Structured JSON logs — queryable, filterable
Latency breakdown — where time is actually spent
Cost attribution — per-call and per-user spending
OpenTelemetry export — pipe to Jaeger or Grafana

Quick start

api.forcedream.ai GET /v1/logs
FORCEDREAM OS · L502 LIVE
# Logs are automatic for all /v1/route and /v1/agents calls
# Start a custom trace span:
$ curl https://api.forcedream.ai/v1/logs/trace \
  -H "Authorization: Bearer $KEY" \
  -d '{"operation":"doc_analysis","service":"my-app","tags":{"user_id":"usr_123"}}'

→ {"span_id":"span_4c8e","operation":"doc_analysis","started_at":1747043200000}

# Query recent traces
$ curl "https://api.forcedream.ai/v1/logs?limit=10&service=my-app" \
  -H "Authorization: Bearer $KEY"

→ {"traces":[{"id":"trace_2b7f","spans":12,"p99_ms":820,"cost_pence":4.2}],"total":147}

Pricing

Simple, transparent pricing

78% of API earnings flow back to you on every call. No hidden fees. Free tier available.

Starter
£19
/mo · £190/yr
10M log events/month
30-day retention
Auto AI tracing
Standard query API
Start free →
Most popular
Pro
£49
/mo · £490/yr
100M events/month
90-day retention
Full-text search
Cost attribution
Alerting
OTel export
Start free →
Scale
Custom
Unlimited events
Unlimited retention
Dedicated pipeline
Custom dashboards
SLA
Start free →

Comparison

How Logs SDK compares

Purpose-built for AI products. Not retrofitted from general-purpose tools.

FeatureForceDream LogsDatadog APMNew RelicDIY logging
AI-native tracingManual
Model cost attributionManual
80% earnings back
Zero configComplexComplexDays
OTel export
Price£19/mo£100+/mo£80+/moVaries

FAQ

Frequently asked questions

A span represents one unit of work — e.g. "model selection" or "memory recall". Spans nest into traces, giving you a complete picture of what happened during a request and how long each step took.
Yes. Every /v1/route, /v1/agents, /v1/memory and /v1/safety call automatically generates a trace — no instrumentation required.
30 days on Starter, 90 days on Pro, unlimited on Enterprise. Older traces can be exported to S3 or your own storage.
Yes. The Logs SDK exports OTel-compatible spans. Pipe to Jaeger, Zipkin, Grafana Tempo or any OTel-compatible backend.
Yes. The API supports filtering by service, user_id, status (error/success), operation name and time range. Full-text search on Pro.

Start with Logs SDK.
Scale to all 22 products.

Free tier available. 80% earnings from your first call. Every call. WORM-sealed by default.

No credit card    80% earnings guaranteed    WORM-sealed audit