Market overview
Live market prices, position vs median, and free inventory per GPU.🔎 Lowest price finder
Filter providers, regions, and sources.| GPU | Provider | Region | $/GPU-hr | $/machine-hr | GPUs | Machine spec | Source |
|---|
Leads
Buyer-intent posts scored by AI. Review, research, and reach out.GPU specs
Curated NVIDIA datasheet specs for quoting and pitches.⚖️ Side-by-side comparison
Compare GPUs across VRAM, TDP, and pricing.Specs are curated from public NVIDIA datasheets; values marked ~ are approximate. Edit app/gpu_catalog.py to correct anything.
Sell / Quote
Build client-ready quotes with committed rates, root storage, and InfiniBand.Or type a custom number instead of using chips above.
Overrides the tier's duration — any term from 1 to 60 months. Rate still follows the selected tier.
Custom rate — override all tiers · $ or ₹ per GPU-hr
Entering a rate disables the tier cards above — clear it to use tiers.
Root storage — each instance's own disk (minimum 100 GB per instance)
You cannot reduce storage once the instance is created. Billed at $0.10/GB·mo.
Extra pooled root storage (GB) — one shared pool added on top of the disks above · $0.10/GB·mo
Shared & object storage — separate volumes for the whole quote (never adds to any instance's disk)
The pills set each instance's own disk (min 100 GB per instance, billed per instance). "Extra pooled root storage" adds one shared pool on top for the whole quote — leave it 0 unless you need extra pooled disk. SFS & object storage are quote-level volumes.
📄 Quote preview
PDF-ready quote will appear here after building.Build a quote above to see the preview and download PDF.
🛠 Studio
Plain-English quotes, migration math, cluster design, price trends, and content.🪄 Ask in plain English
The AI turns a customer request into a live quote using exact jarvislabs.ai committed rates.Examples: “64 H200 for 6 months with 5% discount for Acme AI” · “quote 128 B200 for a year” · “make it 32 GPUs, INR”.
🚚 Migration comparator
Show a prospect what they'd save moving to JarvisLabs committed rates.🧩 Cluster designer
Full cluster spec — nodes, fabric, power, storage, committed pricing.📉 Price history
Daily median market price per GPU from scanner data.📡 Radar
Model releases, arXiv papers, GitHub teams building GPU workloads, inventory trend.📊 Tool usage & backup
Usage stats and one-click full backup of config + database.🏭 JarvisLabs inventory
Live VM and container availability across regions.📋 Region detail
Per-GPU availability, our prices, and website rates.| GPU | Region | Free now | Our price (local) | Website rate (USD/hr) | Monthly (USD, 720h) | Spot price | VRAM |
|---|
📈 Free-stock history
Free VM count over time for a selected GPU.Outreach
Analyze leads, compose personalized messages, and run cold email campaigns.🔍 Analyze a lead
Paste a post, LinkedIn profile, or email — AI extracts requirements and matches inventory.✍️ Compose & send
Start from a saved lead, pick a channel and tone, generate a draft.📧 Cold email campaign
Personalized at scale — AI researches each company and writes a custom email per prospect.| Prospect | Company | Status | Research | Draft |
|---|
📨 Outreach log
History of sent messages across channels.| When | Channel | To | Subject | Status |
|---|
🪙 Token Price Radar
LLM API pricing per 1M tokens — JarvisLabs Token Factory vs every major provider worldwide.| Model | Provider | Input $/1M | Output $/1M | Blended* | Context |
|---|
* Blended = 25% input + 75% output tokens (typical generation-heavy workload).
~ marks best-known estimates — verify current pricing with the provider before putting it in a customer quote.
JarvisLabs rows are editable in data/token_prices_custom.json.
🧠 AI Agent
Your daily prospector — overnight signals, targets, quotes, and content.🎯 Top targets today
Highest-scoring leads to reach out to first.💡 Fresh prospecting ideas
AI-generated angles for outreach and content.✍️ Today's content
LinkedIn, X, Reddit, and email drafts grounded in live data.📊 Weekly review
AI review of the week — market, pipeline health, next week's opportunities.⚙️ Agent standing instructions
Permanent rules the agent obeys in every briefing, idea, target, and piece of content.Example: “Never pitch EU1; lead with IN1/IN2. Always mention 720 hours/month billing. Prefer startups over enterprises.”
🎨 Content studio — on demand
LinkedIn posts, X posts, Reddit replies, cold emails, follow-ups.🏁 Competitor price radar
Provider medians vs our website rates — position yourself in the market.Provider medians vs our website rates
Gap shows how much cheaper (or pricier) we are vs competitors.| GPU | Provider | Their median | Our rate | Gap | Offers | Position |
|---|
📈 Price trends
Market median movement per GPU — timing ammunition for customer calls.Pick a GPU to see the market trend.
📚 GPU & Cloud Academy
GPU specs, cloud concepts and sales plays — flashcards, quizzes and certification exams.🤖 AI Company Scanner
US AI companies that need GPUs — discovered by AI, scored by GPU demand.🔥 Demand Leads
Live buyer monitoring — companies actively looking for GPU rentals.📉 Price Alert Rules
Get notified when a GPU hits your target price.| GPU | Region | Rule | Threshold | Enabled | Last Triggered |
|---|
Add / Edit Rule
Set price thresholds and get notified when they fire.Alerts are checked automatically on every price scan (15 min). You'll get a browser + event notification when a rule fires.
Settings
Configure data sources, notifications, seats, and app behavior.🏷️ Our offering (JarvisLabs)
Your own pricing, currency, regions, and API key.Our prices per GPU (local currency / hour)
| GPU | Our price | ≈ USD/hr | Free now |
|---|
🏬 Market providers (curated list prices)
Reference prices. EST = unverified estimate, LIST ✓ = verified.| Provider | GPU | Region | $/GPU-hr | URL | Verified |
|---|
🛰️ Lead monitoring
Sources, keywords, and score thresholds for lead discovery.🌐 Google web search — covers X, LinkedIn & Facebook
No free search APIs exist for X/LinkedIn/Facebook — this monitor sweeps them via Google Custom Search with buyer-intent queries. Setup (free, 100 queries/day): create an API key in Google Cloud (enable Custom Search API), then create an engine at programmablesearchengine.google.com with “Search the entire web” ON, and paste both values here.
🔔 Notifications
Where alerts and quotes land — Telegram, Slack, ntfy, WhatsApp, SMS, email.⌨️ Slash commands: in your Slack app → Slash Commands →
Create New Command (e.g. /quote) → Request URL
https://217.18.55.216:8000/slack/commands (public IP) — or use Socket Mode (no URL needed). Add one entry per command you want
(/quote, /value, /status, /price, /stock,
/cluster…) — all point at the same URL. Then type /quote 8 H200 3 directly in any
channel — no @mention needed.
🤖 Slack bot setup (~3 min, no public URL needed):
create an app at api.slack.com/apps → Socket Mode ON (copy the xapp- token) → add OAuth scopes
chat:write, app_mentions:read, im:history,
channels:history, groups:history → subscribe to bot events
app_mention + message.im → install to workspace (copy the xoxb- bot token) →
in Slack, invite the bot to your group with /invite @YourBot. Then @mention it in the group
or DM it — every Telegram command works, including quotes with proposal PDFs.
📱 CallMeBot setup (free, ~1 min): add the number
+34 644 51 95 23 to your phone contacts, send it
I allow callmebot to send me messages on WhatsApp — it replies with your API key. Paste that key above.
Twilio trial accounts work — verify your mobile number in the Twilio console first. SMS costs ~$0.01/message on a paid account.
🔔 ntfy setup (2 min, free): install the ntfy app from the Play Store / App Store → subscribe to the exact topic name above → alerts arrive as high-priority push notifications on your phone. No account, no cost, works in India.
🤖 AI personalization
LLM writer for outreach drafts — grounded in lead context + live market data.👥 Team seats
Colleague logins — viewer (read-only) or quote (build quotes). Your admin login stays full access.📤 Outreach sending
Emails through SMTP. LinkedIn via your automation bridge (n8n / Zapier / Make) or manual copy.📦 Inventory restock alerts
Notify when a GPU comes back in stock (0 → available) or free count increases.⚙️ App
Password, scan intervals, retention, webhooks, and public pricing page.Intake webhook (for Zapier / Make / X API bridges)
POST JSON here: {"source":"facebook","title":"…","text":"…","url":"…","author":"…"}
— anything mentioning a tracked GPU becomes a scored lead automatically.
JL platform webhook — instant inventory events
Point a hook inside the JarvisLabs platform here: POST {"text":"2x H100 freed in india-noida-01"}
— the event lands in Telegram and the dashboard instantly, bypassing polling.
Public pricing page — share with customers
Read-only page with your live rates + availability. Safe to put in your bio or DMs.