Real-time log search, smart alerts, AI anomaly detection, and error grouping — everything you need to debug production. Nothing you don't. Free to start, $19/mo when you grow.
Works with any stack
The problem
Most teams are stuck between three bad options. LogFlow is the fourth.
Datadog, New Relic, Splunk — $300-2,000/mo for log management alone. You only use 10% of the features but pay for 100%.
Elasticsearch needs JVM tuning, shard management, index lifecycle policies, version upgrades. That's 10-20 hours/month of ops work.
SSH into production, run grep, pray. Works for one server. Falls apart with multiple services, team members, or the need to search last week's logs.
How it works
Any language with an HTTP client works. There's also a Node.js SDK, a browser bundle, and a CLI.
One package. Node.js, browser, or plain HTTP — any language with a fetch call works.
JSON logs with context. Every field becomes searchable and filterable instantly.
Millisecond search. Error rate alerts to Slack. AI anomaly detection. Done.
import LogFlow from '@getlogflow/js'
const logger = new LogFlow({
apiKey: 'lf_your_key',
service: 'my-api'
})
logger.error('Payment failed', {
orderId: 'ORD-9912',
userId: 4821,
provider: 'stripe'
})Features
Search, stream, alert, and group errors — without stitching together five different tools.
ClickHouse-powered full-text search across billions of log lines. Filter by level, service, host, or any custom attribute. Results appear before you finish typing.
Stream logs from all services directly in your browser at full speed. Filter by service or keyword, watch the logs/sec counter tick, copy raw JSON with one click.
Keyword match, error rate thresholds, volume spikes, silence detection. Routes to Slack, Telegram, email, Discord, or a custom webhook. One alert per event, not six.
Errors are fingerprinted, normalized, and collected into Issues. Track regressions, request AI summaries, resolve or mute noise — independently from the raw log stream.
And more
Click "Explain this error" on any log. Claude reads the context and explains what went wrong.
Build log-based dashboards with stat widgets, timelines, top errors, and service breakdowns.
Toggle per-project PII masking. Emails, IPs, and card numbers are redacted at ingestion.
Click any trace_id to see the full request timeline across services — chronological, with latency.
AI detects error spikes, volume drops, and service silence. Two-strike rule prevents false alarms.
POST a deploy via API. See it as a purple line on your charts — correlate deploys with error spikes.
Pricing
All plans include every feature. Price scales with data volume and retention — not with how often you search.
$0/mo
Try it out, no card needed
$19/mo
For small teams shipping fast
$49/mo
For teams with real production traffic
$149/mo
For scaling teams that need more
$499/mo
| Feature | Free | Starter | Growth | Pro |
|---|---|---|---|---|
| Monthly data | 1 GB | 10 GB | 100 GB | 500 GB |
| Log retention | 3 days | 14 days | 30 days | 90 days |
| Projects | 1 | 5 | 20 | Unlimited |
| Team members | 1 | 3 | 10 | 25 |
| Log search & live tail | ||||
| Alerts (Slack/Telegram/webhook) | ||||
| Anomaly detection | ||||
| AI explanations & NL query | ||||
| Trace correlation | ||||
| Deployment markers | ||||
| Ingestion rules | — | |||
| Log-based dashboards | — | |||
| Scheduled email reports | — | |||
| Service map | — | — | ||
| Audit log | Ent. | Ent. | Ent. | Ent. |
| API rate limit (per key) | 1k/min | 2k/min | 5k/min | Unlimited |
Set up in 5 minutes. Free tier with 1 GB/month. No credit card, no sales call, no YAML config files.