BETA Hydrate is in beta. Register during beta and lock $5/mo Pro forever - free during beta + 1 month after v1 launches. What this means →
Episodic memory · for Claude Code

Your AI forgets.
Hydrate remembers.

Claude Code starts each session with a blank mind. Hydrate captures what was built, decided, and named - then quietly injects it into the next session. No prompt engineering. No screenshots of previous chats. Just context that works.

curl -fsSL gethydrate.dev/install | sh

Apple Silicon and Intel both supported - the installer picks the right binary.

Works with Claude Code · Cursor · Cline · Zed · Gemini CLI via MCP · local-first

The line you care about depends on how big you are.

Hydrate saves tokens. What that saving becomes for you depends on scale.

Solo developer

Extra coding days, not a rebate.

You already pay Anthropic for Claude Max. Hydrate makes that subscription go further: fewer tokens burned on context rediscovery means more tokens left for real work before you hit the weekly cap. The dashboard shows this as Runway - extra days of coding this week. For a $20/month Max user, that typically works out at 2 to 4 extra productive days per month.

Free tier. No credit card.
Enterprise

Meaningful spend reduction at scale.

Coding traffic is dominated by context rediscovery. For a 50-developer team running heavy Claude usage, displaced tokens translate to serious money at the API meter and serious carbon at the provider's data centre. Our projection: 30 to 50% API spend reduction on code-gen workloads with a properly deployed memory layer. Self-host, SSO, SOC 2, air-gap if you need it. Nothing leaves your network.

Custom pricing. Talk to us.
2 → 7
sessions shipping
Haiku, multi-session scenario
−57%
cost reduction
Opus, simple scenario
45%
of Sonnet cost
hybrid mode, Sonnet-tier quality
$0
forever
Free tier, local only

From the 2026-04-17 benchmark run. $67 of real API spend. Every number is reproducible.

What changes

The difference is execution, not intelligence.

Models don't get smarter with Hydrate. They get context. That's usually enough.

without Hydrate
$ claude "add a dark mode toggle"

Haiku: What should the default
 colour scheme be?
Haiku: What do you mean by
 "toggle" - a button? switch?
Haiku: Which pages should be
 affected?

[ nothing written ]
with Hydrate
$ claude "add a dark mode toggle"

Haiku: edits src/styles/globals.css
Haiku: edits src/components/Nav.tsx
Haiku: Done. Toggle wired to
 prefers-color-scheme. Persists in
 localStorage. Icon updates on
 change.
One intervention · four outcomes

Reliability is the unlock. Cost savings follow.

Hydrate isn't a uniform speed-up. Each model has a different failure mode without context - refusal, drift, re-exploration, ceiling-hitting - and Hydrate removes all four. Cost reduction is the measurable consequence; reliability is the primary change.

Haiku 4.5 reliability unlock
before
2/7 sessions ship. Asks clarifying questions instead of building.
with Hydrate
7/7 sessions ship. Every prompt executed. Cost per shipped session: $0.15 vs $0.30.
Sonnet 4.6 neutral stabiliser
before
Builds cleanly, 7/7 sessions. The polish ceiling.
with Hydrate
Builds cleanly with modest efficiency gain. −14% cost, zero regression.
Opus 4.7 (simple) cost reducer
before
Re-explores workspace every session.
with Hydrate
Skips re-exploration, produces same output. 57% less cost for identical work.
Opus 4.7 (complex) capability unlock
before
3/7 sessions refuse with "need clarification". One $3.88 ack.
with Hydrate
6/7 sessions build, converting refusals into full features. The $3.88 ack becomes a complete docs site.
Hybrid - Sonnet s1 → Haiku s2+ architecture propagation
before
n/a - this is a workflow, not a model.
with Hydrate
Sonnet-tier design propagates to Haiku via memory. $1.41 total, $0.20 per shipped session (45% of pure Sonnet).
How it works

Three quiet pieces.

01

Hooks

Hydrate installs two Claude Code hooks - UserPromptSubmit and Stop. They're invisible to you. They capture each session's prompt, decisions, and output.

02

Facts

Between sessions, an extractor pulls concrete facts - "pricing tiers are Free/Pro/Team", "accent colour is #89b4fa" - and stores them scoped to your project.

03

Dashboard

A local dashboard at localhost:8089/dashboard shows your spend, cache ratio, and facts per project. Nothing ever leaves your machine - Free, Pro, or Team.

Built with Claude Code

This site is its own first data point.

Hydrate reads Claude Code's local transcripts and adds them up. These are the real numbers for the work that produced gethydrate.dev - no rounding, no cherry-picking.

Cost to build gethydrate.dev
$822.17
2.06B tokens · 4,164 messages · 5 days
2026-04-14 → 2026-04-18
Estimated savings with Hydrate
$287.76 - $493.30
Midpoint $394.64 · 35-60% of the baseline for opus on a greenfield project.
projected from the 2026-04-17 benchmark - not measured on this site

Priced at Sonnet rates ($3/$15/$0.30/$3.75 per MTok). Snapshot: 2026-04-18. Raw JSON: /usage.json.

BETA Pricing

Start free. Lock $5/mo Pro forever.

Register during beta - whichever edition you pick - and your Pro rate is locked at $5/mo forever, even if retail launches higher. How this works →

BETA · locks $5/mo Pro

Free

$0/ forever

Up to 2 active projects, full hooks + dashboard. No account. Registering during beta also locks your future $5/mo Pro rate.

Install + lock rate
Post-beta

Team

$29/ user / month

Shared project memory across teammates. Indicative launch rate; Team is not part of the beta lock.

Join Team waitlist →

Enterprise with SSO, self-host, org-scoped facts - contact sales.

Ready when you are

Install in 30 seconds.

No account. No telemetry. The Free tier never touches the network.

curl -fsSL gethydrate.dev/install | sh

Apple Silicon and Intel both supported - the installer picks the right binary.