Cherry is the self-hosted alternative to Gemini.
Gemini is welded into Google Workspace, ships aggressive data-collection defaults and gates the best capabilities (Gemini 3.1 Pro, Deep Think, Deep Research, Jules coding agent, Gemini Spark) behind the Google AI Pro ($30/mo) or AI Ultra (from $60/mo) tiers. Cherry ships Google models through a pluggable adapter (Google models are in the catalog today, runtime adapter on the roadmap) alongside OpenAI and Anthropic in one interface, wrapped in an infinite nested folder tree, per-message cost attribution, private MinIO storage and the Pins/Clipboard/Snippets reuse system. Self-hosted end to end. You decide what the retention model looks like, not Google.
Gemini is tied to Google Workspace. Cherry is a standalone workspace.
Google's defaults tilt aggressive. Cherry defaults everything off.
Cherry routes across three model families: Google, OpenAI and Anthropic.
Four ways Gemini keeps its models tied to Google.
Google ships strong models. Gemini the product ships them behind Workspace bundling, aggressive telemetry defaults and a UI Google reshapes on a quarterly cadence.
Welded to Google Workspace
Gemini's best surfaces live inside Gmail, Docs, Sheets and Meet. That's great if the whole company is on Workspace and actively painful if not. Cherry is a standalone workspace with its own connectors: Google Drive ships today, Notion / Slack / Jira are on the roadmap. Use Google's models without being conscripted into Google's productivity stack.
Aggressive data-collection defaults
Google's consumer AI defaults tilt toward "everything on". Cherry ships private MinIO storage with presigned URLs, sanitized filenames and zero bucket-path exposure. Conversation history lives in your Postgres. Prompts hit vendor APIs directly with your keys. You own the retention policy.
Quarterly product reshaping
Gemini has been Bard, then Duet, then Gemini, then Gemini Advanced, and now sits behind three shifting AI tiers (Plus $20, Pro $30, Ultra $60+) with capabilities like Deep Think, Deep Research, Jules and Spark drifting between them. Cherry's workspace primitives (folders, Pins, Clipboard, Snippets, Presets, UsageEvents) don't reshuffle when a vendor rebrands. Your organization survives every model release.
One vendor, no cost attribution
Gemini is Google models only, billed per seat or per subscription tier, with no per-prompt visibility. Cherry logs a UsageEvent per message with tokens and dollar cost against user, workspace, chat and model, and the Recharts dashboard filters by today / 7d / 30d / billing cycle. Route Google alongside OpenAI and Anthropic in one interface.
A workspace Google doesn't own, doesn't rebrand and doesn't bundle.
Cherry stays a standalone workspace. It doesn't require Google Workspace, doesn't ship telemetry-on defaults and doesn't drift through a new brand every product cycle. Google's models are in the Preset catalog today with the runtime adapter on the roadmap. OpenAI and Anthropic ship right now. Every prompt logs a UsageEvent. Every file lives in your MinIO. Every Preset carries across every provider. And nothing about the workspace shape changes when a vendor swaps Gemini Advanced for AI Ultra tomorrow.
Gemini vs Cherry.
Access Google's models on your terms, alongside every other provider, without the Workspace bundle or the telemetry defaults.
| Feature | Gemini | Cherry |
|---|---|---|
| Model providers | Google only | OpenAI + Anthropic shipping, Google adapter on roadmap, pluggable for any provider |
| Product bundling | Fused with Google Workspace | Standalone workspace |
| Data-collection defaults | Aggressive by default | Your Postgres, your MinIO, your retention policy |
| UI stability | Reshaped every quarter | Stable workspace primitives across releases |
| Organization | Flat conversation list | Infinite nested folder tree, self-referential parents |
| Per-message bookmarks | No | Pins with scroll-to-highlight animation |
| Per-chat clipboard | No | Drag-reorderable saved passages, promote to Snippet |
| Cross-chat snippets | No first-class equivalent | Project-scoped, tagged, one-click injection to any chat |
| Model presets | Gems (Google-only) | Agent Library. Presets with model + temp + system prompt + tier |
| Model comparison | Not supported | Fire one prompt at multiple models side by side |
| Cost per prompt | Not shown | Tokens in, tokens out, dollar cost, latency on every message |
| Cost attribution | Per seat or per subscription tier | Per user, per workspace, per chat, per model |
| Usage dashboard | None for end users | Recharts. Today / 7d / 30d / billing cycle |
| Bulk folder upload | No | Replicates an entire OS folder tree transactionally |
| File storage | Google infrastructure | Self-hosted MinIO with local-disk fallback, presigned URLs |
| Conversation history | Google-owned | Your Postgres. Exportable, portable |
| Streaming transport | Proprietary | Server-Sent Events with cancel-mid-stream |
| Team workspaces | Workspace-tenant | Multi-workspace with Owner / Admin / Member roles |
| SSO | Google identity only | Available in Enterprise, configurable on your infrastructure |
| Keyboard shortcuts | Limited | 18 mapped across Navigation / Chat / Editor / Files |
| Connectors | Google-native only | GitHub + Google Drive shipping, Notion / Slack / Jira on roadmap |
| Apps | Gems, Jules (coding agent), Gemini Spark, Deep Research, Workspace integrations | Canvas, Code Interpreter (beta), Image Generation, modular |
| Outbound API | Google APIs (fragmented) | Cherry outbound API for third-party integration |
| Deployment | SaaS only | Self-hosted. On-prem, cloud, run as your own SaaS. You decide |
| Ownership | Vendor-owned | You own the code, the data, the deployment |
Frequently asked questions.
Can we use Google's models in Cherry today?
Google models live in the Cherry model catalog today for Preset filtering and Agent Library configuration. The runtime Google adapter is on the roadmap alongside the OpenAI and Anthropic adapters already shipping. Every provider adapter is a typed interface, so lighting up Google is a single-file addition.
Do we still need Google Workspace?
Not for Cherry. Cherry is a standalone workspace. If your organization uses Google Drive, Cherry's Google Drive connector wires up file ingest. If you don't, nothing forces you into the Workspace bundle.
What about Gems and Google's per-Project system prompts?
Cherry replaces Gems with the Agent Library. A Preset captures the model, temperature and system prompt, and any Preset can be filtered to a specific tier requirement. Folders group the chats. Snippets hold project-scoped reusable text with tags. Portable across every provider, not only Google's.
Where do uploaded files live?
Files live in MinIO on your infrastructure, with a local-disk fallback if MinIO isn't wired. The storageKey is stored in the File record. The raw bucket path is never exposed. Every file access goes through a short-TTL presigned URL generated per request. Filenames are sanitized. If you shut Cherry down tomorrow, the files stay in your bucket.
How does Cherry stream responses?
Cherry streams via Server-Sent Events over HTTP. Every provider adapter implements the streaming interface, so tokens appear the moment the model emits them. The Stop button cancels mid-stream. SSE is a standard browser API. There's no proprietary transport.
How is Cherry deployed?
Cherry is self-hosted. Docker Compose orchestrates nginx, the Next.js frontend, the Express backend, Postgres, Redis and MinIO. Deploy it on your own hardware for internal use, run it in your cloud or spin it up as your own SaaS product. You own the code, the data and the deployment.
What does Cherry cost?
Contact sales for pricing. Every engagement is scoped to how you plan to deploy, who your users are, which provider adapters you need lit up and what support terms fit your organization.
Ready to use Google's models without Google's product?
Tell us how your team uses Gemini today, which Google services you actually want to keep and where the bundle hits the wall. We'll respond within one business day with an honest read on whether Cherry fits.