AI Assist
Freemium
Users prioritizing on-device processing and data privacy
Inkfold
Freemium
Every time you start a new chat in a different AI tool, you re-explain the project — the stack, the constraints, the decisions already made — because nothing carries over.
$19/month or $190/year · Free tier: One memory across models, 50 Inkfold-managed messages a month (planned), 90-day history, 2 devices, 1 gigabyte storage · Checked Oct 21, 2026
Similar tools
Ruune DesktopFreemiumRuune Desktop pairs a sub-3mm hardware device with a desktop app to capture audio at the chip…
AI AssistFreemiumThe tool installs as a browser extension, captures audio from your active meeting tab, and runs…
AishaFreeAisha centers Black sources and history as design principles rather than as a filter layered on…
DocfilaFreemiumThe core workflow runs scan-to-delivery in four steps: capture a paper document or upload a…
Summary
Inkfold sits between you and your AI subscriptions, indexing conversation history across ChatGPT, Claude, Gemini, Grok, and others, then injecting the relevant context when you open a new chat on any vendor. The core workflow is capture-remember-inject: import existing chats, let the tool extract entities and decisions, then route new queries through whichever model fits the task with your memory already attached. It works well for solo power users who juggle multiple subscriptions and lose context at every vendor boundary. The ceiling appears fast for teams: shared memory and the admin console are paid-only features, and the product is still in private alpha — so the reliability guarantees you need before connecting production workflows do not yet exist in public form.
Bottom line: Pick Inkfold if you pay for three or four AI subscriptions and spend real time re-establishing project context on each one — skip it if you need auditable, SLA-backed memory for a team workflow that cannot tolerate alpha-stage breakage.
Pricing Plans
SubscriptionLast verified 2 weeks ago- Price
- $19/month or $190/year
- Free Tier
- One memory across models, 50 Inkfold-managed messages a month (planned), 90-day history, 2 devices, 1 gigabyte storage
Free
Free
Getting started tier
- One memory across models with answer receipts and one-click correction
- Shared library of personas and playbooks (beta)
- 50 Inkfold-managed messages a month (planned, not in beta)
- 90-day history, 2 devices, 1 gigabyte storage
- Your own key for every vendor
- Smart, Private and Incognito modes
Pro
$19per month
$190/yr Save 17%
For individuals
- Everything in Free
- Multi-model modes: Compare, Council, Debate, Pipeline
- Prompt improvement
- Spending view with budget rules
- 2,000 Inkfold-managed messages a month (planned, not in beta)
- Unlimited history and devices, 50 gigabytes storage
- Your own storage bucket
Team
$32per month
For teams, $32 per seat per month
- Everything in Pro
- 3,000 Inkfold-managed messages a month (planned, not in beta)
- Team workspaces and shared memory (beta)
- Admin console and audit log (beta)
- 50 gigabytes per seat or bring your own storage
Founding member
per month
$190/yr
First 20 people, price locked while subscribed
- Everything in Pro
- Price stays $190 a year while you stay subscribed
- Direct line to the founder
- A say in what ships next
View full pricing on inkfold.app →
Pricing may have changed since last verified. Check the official site for current plans.
Find your fit
Before you choose Inkfold
More options in Productivity. Compare pricing, platforms, and features side by side.
Aisha
Free
Users seeking culturally grounded AI responses
Pros
- Cross-vendor memory injection, so you stop re-establishing project context every time you switch from Claude to GPT or Gemini — the AI you open next already knows the decisions the last one helped you make.
- Bring-your-own-key support for every major vendor, so your queries route through your own API credentials and you control cost exposure instead of burning through a fixed monthly message cap.
- Three retention modes — indexed, E2E-encrypted private, and incognito — so you can keep sensitive conversations opaque to the platform without giving up the memory layer entirely.
- Shared team memory with pooled message credits, so context one teammate builds in Claude is available when the next person opens a GPT session without anyone manually syncing notes.
- Local model support via Ollama, so teams that run self-hosted LLMs for compliance or cost reasons can still participate in the shared memory layer alongside cloud vendors.
Cons
- The product is in private alpha, meaning you request access rather than sign up — teams that need a stable, documented API or guaranteed uptime before connecting real workflows have no current path to that on the public tier.
- No self-hosted deployment option exists for the memory layer itself, so organizations in regulated environments that cannot route conversation metadata through a third-party cloud must either wait for a custom enterprise contract or build their own context store — at which point Inkfold is not in the picture.
- Shared memory, the admin console, and audit logs are all paid-only features, so a team evaluating the free tier to validate fit will not encounter the core multi-user capability until after they have committed to payment.
- There is no documented API surface, which means you cannot programmatically read from or write to Inkfold memory inside your own applications or CI pipelines — teams that want memory as a service they call from code will move to a vector-store-backed solution like a self-managed RAG pipeline instead.
About
- Platforms
- Web app, CLI, mobile app
- API Available
- No
- Self-Hosted
- No
- Last Updated
- July 13, 2026
Best For
Who it's for
- Power users with multiple AI subscriptions
- Organizations requiring custom data residency
What it does well
- Maintain project context when switching between paid AI subscriptions
- Share team memory across different AI tools and users
- Build persistent knowledge base from multi-vendor conversations
Integrations
Frequently Asked Questions
- Is Inkfold free?
- Inkfold offers a free option alongside paid upgrades. Pricing: $19/month or $190/year. Check the current allowances and terms before choosing a plan.
- What platforms does Inkfold support?
- Inkfold is available on: Web app, CLI, mobile app.
Curated lists that include this category
Switching between AI vendors means starting cold every time — no memory of the codebase decisions, the client preferences, the thread you were pulling last week. Inkfold addresses that by acting as a shared memory layer across vendors. You connect your existing tools, import past conversations, and then route new chats through Inkfold’s web app, CLI, or mobile app. At each turn, it extracts entities, decisions, and preferences from the conversation and adds them to a structured context store. When you open a chat — whether on Claude, GPT-5, Gemini, or a locally-run Ollama model — Inkfold injects the relevant signal before the first token fires.
The differentiating architecture is the three-layer signal stack the vendor describes: your personal history, topic-level patterns, and a crowd-sourced domain signal that the platform calls FusionLayer. The claim is that context sharpens over time rather than growing stale or bloated. Privacy tiers add specificity: Smart retention indexes conversations for full context quality, Private mode applies end-to-end encryption with opaque nodes so even Inkfold cannot read the content, and an Incognito mode captures nothing. Bring-your-own-key support means you can route queries through your own API credentials for every major vendor instead of consuming Inkfold-managed message credits.
The tool fits individual power users who maintain parallel AI subscriptions and need continuity across them without copy-pasting context manually. Teams get shared memory and an admin console, but those features are paid-only, and the product is currently in private alpha — meaning access is gated, the feature set is still solidifying, and production-grade SLAs are available only through a custom enterprise conversation. Organizations that need custom data residency or compliance controls are directed to contact the vendor directly; no self-hosted deployment path is documented on the public page.
On the integration side, the vendor lists supported import sources as ChatGPT, Claude, DeepSeek, Gemini, Grok, and Perplexity, and lists supported models including Claude Haiku, Sonnet, and Opus; GPT-5 and GPT-5 nano; Gemini 2.5 Pro and Flash; DeepSeek R1; Grok; Llama 3.3 via Ollama; Mistral; and Qwen 2.5. There is no API surface documented for external integration, so teams cannot wire Inkfold memory into their own tooling programmatically.
Contribute
- Work on Inkfold? Claim this listing.
- Embed a badge for your README or site.
Sign in to rate this tool or suggest an edit.