I found the issue. The auth token was not refreshing when the tab came back into focus, so I added the fix and covered the edge cases.
Used context from 12 messages
AI that keeps up with the way you work
Learns how you work.
Thinks with you.
Powered by Claude
Move between models in the same conversation. Grove keeps the context, so you do not have to repeat yourself.
get started free →free for 14 days · no credit card
personal ai
Every conversation gets its own assistant.
Pick the model that fits the moment. Grove keeps the thread, so you can switch models without losing context.
helps Maya with quarterly planning.
thinking partner
helps Dev with code review.
always on
helps Priya with market research.
in your corner
Grove starts remembering what matters.
You should not have to explain your style, projects, and priorities every time you open a new chat. Grove carries that context forward quietly, so each answer starts closer to what you need.
By the third message, Grove already sounded like it knew how I worked. I did not have to keep spelling things out.
Sarah K.
Product Manager, Loom

Direct, data-first, Oxford commas
Q3 product launch, reducing churn
Priya (eng lead), Marcus (design), CEO
Investor pitch prep, 2h ago
Can you help me rewrite this email to be more direct?
Sure. Here's a tighter version that gets to the point faster...
Now give me 3 subject line options
Here are 3 punchy options: 1. ‘Let's talk’ 2. ‘Quick ask’ 3. ‘One question’
One thread.
Every model.
Start with Claude, ask GPT-5.4 for a sharper pass, then bring in Gemini for another angle. Grove keeps the thread together, so the next model already knows what happened.
- Switch models with one tap
- Use Claude, GPT-5.4, Gemini, Mistral, and more
- Keep the same thread, notes, and decisions
Start with a prompt that fits the work.
Pick a template or describe what you need. Grove researches, drafts, and summarizes in one thread, with memory and connected Google apps on Max.
Morning Briefing
Pull together unread messages, top priorities, and one headline I should know about today.
Standup Recap
Summarize the day's threads and tickets into a short standup update I can paste straight into Slack.
Weekly Writing Review
Review my last few documents, spot patterns in my writing, and suggest one useful improvement.
Newsletter Draft
Draft this week's team newsletter from my shipped work and notes, in my voice and ready to send.
Reply Polishing
Take my rough replies and rewrite them to match my usual tone and brevity before I hit send.
Research Digest
When I paste a topic, find 5 strong sources, summarize the main points, and call out anything that conflicts.
Competitor Watch
Summarize recent pricing, feature, or leadership changes for the competitors I care about.
Source Summaries
Turn a stack of links into a one-page brief with the key claims and a note on where the sources conflict.
Code Review Brief
When I mention a PR, summarize the diff, flag the risks, and draft a review comment in my voice.
Incident Postmortem
After an incident, pull the timeline and chat into a draft postmortem with causes and action items.
Grove notices the next helpful step.
On Max, Grove surfaces proactive cards from your calendar, tasks, and inbox: follow-ups, briefings, and nudges you can accept or dismiss.
3 suggestions surfaced this week
You often turn meeting notes into follow-ups. Want Grove to draft the next one?
Seen in 8 recent conversations
Want a follow-up for last week's investor email?
Based on your Google Calendar and Gmail
You have not replied to Marcus in 4 days.
Grove can start a warm reply.
Research completed: 'AI compliance in fintech 2026'
Saved to your documents
“It noticed a follow-up I had been putting off for two weeks and drafted a reply I could actually send. It took 10 seconds.”
Marcus R.
Head of Sales, Acme
Nine curated models. One conversation.
Grove ships a hand-picked catalog across frontier, balanced, and fast tiers. Switch mid-chat on Pro and Max without losing the thread.
Sonnet
balanced default
GPT-5.4
OpenAI flagship
Gemini
long context
Opus
frontier reasoning
Haiku
fast replies
DeepSeek
value reasoning
Mistral
European ops
Qwen
multilingual
Kimi
agentic tools
Your conversations stay private.
Your context helps you, not someone else's model.
No Model Training
Your conversations are not used to train models. That applies to Grove and to the providers you use through it.
Encrypted Everywhere
Your conversations are encrypted in transit and at rest. Grove runs on AWS in the EU (Frankfurt region).
Private by Default
Grove does not sell your data or pass it around. What you say stays with your assistant.
You Stay in Control
Export, delete, or pause your data anytime. You can see what Grove knows and change it.
Encrypted in transit · AWS EU-hosted · No model training · You control your data


