PRODUCT · LOGS

Trace, debug, eval — without a second SaaS.

Full-fidelity traces of every LLM call and tool step. Token, latency, and cost on each. Replay traces, run evals against datasets, ship to your warehouse. LangSmith-grade observability, billed per trace.

Get started — free
● LIVE LOGSlast 60s · 6 of 1,284
14:02
user_28f3a
gpt-4o
1.2k
412ms
$0.018
14:02
user_91f0c
sonnet
0.9k
380ms
$0.012
14:02
user_28f3a
github
190ms
$0.001
14:01
user_55a2e
gpt-4o
2.1k
590ms
$0.034
14:01
user_28f3a
stripe
120ms
$0.001
14:01
user_91f0c
gpt-4o-m
0.4k
210ms
$0.0008
/logs

Logs

LangSmith-grade observability. Full traces, token + latency + cost, replay, eval.

  • Tokens
  • Latency
  • Provider
  • Cost
  • Status
  • Webhooks
2.1Btraces indexed
30 ddefault retention
120 msp95 search
S3 / GCSwarehouse export
What it does

The three things logs actually does.

TRACES

Every step in the chain.

Each LLM call, each retry, each tool step is captured. See the full agent run, not just the final response. Tokens, latency, cost on every step.

traces · ok
production debugging
200 · 120 ms
DEBUG

Replay any production trace.

Filter by user, model, route, status. Pin a slow request. Replay with a different model or prompt. Share a permalink with your team.

debug · ok
cost + latency
200 · 120 ms
EVAL

Score traces against datasets.

Build datasets from production traces. Run evals on every commit. Compare runs side-by-side. Catch quality regressions before the user does.

eval · ok
eval + datasets
200 · 120 ms

Datadog for spend, BigQuery for cost views, a CFO spreadsheet for the receipts — it's three tools doing what should be one log table.

Why we built this

Urgent backstory

We watched five AI startups in a row hand-roll the same logs stack — and burn three weeks doing it. We packaged ours so you don't have to. Drop the SDK in once; this product, plus the rest of the suite, comes with it.

Use it for

Four common
ways teams ship with Logs.

01P

Production debugging

Filter, pin, and replay any trace. Compare runs side-by-side. Find the one slow tool call buried in a 12-step agent run.

02C

Cost + latency

Per-user, per-model, per-route. Find your most expensive user in 2 seconds. Find your slowest provider. Reroute.

03E

Eval + datasets

Build datasets from real traces. Run evals on every commit. Catch regressions before they ship.

04W

Warehouse export

Webhook to Snowflake, BigQuery, S3. Stable schema, hash-chained, audit-ready.

How it works

Four steps, ten minutes.

1

Already on

Every call you make through the SDK is logged.
2

Search the dashboard

logs.search({ user, model, ... })
3

Export rows

Webhook URL or scheduled S3 dump
4

Join in your warehouse

Same schema, same primary keys
It works with

The stack you already have.

Snowflake
BigQuery
Databricks
S3
GCS
Postgres
ClickHouse
Datadog

First time we've had real numbers on AI cost per user. It changed the pricing conversation in the same week.

Tom · COO, Loomstack
L
Pricing

Free to ship. Pay when you scale.

FREE$0

Up to 1k requests/mo. Every product included.

Start free →
SCALE$0.0008

per request, after free tier. No markup on tokens. Cancel anytime.

What's next

Shipping this quarter.

Read full changelog →
WK 12Live tail
WK 13Anomaly alerts
WK 14Per-tool cost rollup
WK 15Sampling controls
ONE AFTERNOON AWAY

Ship canonical AI logs this afternoon.

We onboard 1–2 indie startups a week. If you'd rather ship features than maintain a logs stack, talk to us.

Get started — free