Deploy SAM
All Features Cloud & Costs

Pick your clouds.
See what they cost.

Connect Hetzner, Scaleway, GCP, or four other providers. You pay them directly at their prices, and SAM shows how many server hours each person used.

  • Pools are built from each connected credential’s live catalog — concrete SKUs like cx33 or PRO2-S with real vCPU, RAM, disk, and price
  • Project, user, and installation scopes, resolved in that order
  • How agents are placed (Balanced, Pack, Spread, Smallest fit) and what happens when a provider is out of capacity (wait and retry, try other allowed machines, or stop with an error), per pool
  • Nodes are reused when they fit and stay warm for 30 minutes
  • Settings → Usage tracks token spend by model and day against a monthly budget; Admin → Costs and Admin → Usage cover the whole installation
Read the guide
Project compute pool editor with agents placed by Pack, Try other allowed machines when a provider is out of capacity, and a machine limit of 3 per person, above the allowed machines for each connection: Hetzner cx23 allowed in Falkenstein, cx33, cpx31 and cx43 with their regions and monthly prices, a sold-out Falkenstein region marked unavailable, and Scaleway machine types belowProject compute pool editor with agents placed by Pack, Try other allowed machines when a provider is out of capacity, and a machine limit of 3 per person, above the allowed machines for each connection: Hetzner cx23 allowed in Falkenstein, cx33, cpx31 and cx43 with their regions and monthly prices, a sold-out Falkenstein region marked unavailable, and Scaleway machine types below

Choose exact machine types and regions

A compute pool lists the machine types SAM may rent, such as cx33 at Hetzner or PRO2-S at Scaleway, with each one’s vCPU, RAM, disk, and estimated price. Every machine type a provider offers is listed with its regions: tap a region to let SAM start that machine there, or search by name or region. You also choose how agents are placed on servers (Balanced, Pack, Spread, or Smallest fit) and what SAM does when a provider is out of capacity: wait and retry, try other allowed machines, or stop with an error.

The allowed machines list: a search box, then Hetzner machine types with region buttons, cx23 allowed in Falkenstein, cx33 offered in Nuremberg, cpx31 in Helsinki, and cx43 whose Falkenstein region is sold out, followed by Scaleway machine typesThe allowed machines list: a search box, then Hetzner machine types with region buttons, cx23 allowed in Falkenstein, cx33 offered in Nuremberg, cpx31 in Helsinki, and cx43 whose Falkenstein region is sold out, followed by Scaleway machine types Pick the regions where SAM may start each machine type

See why a node was chosen

Resource requirements resolve from task, trigger, skill, and agent profile down to project and platform defaults. The chat’s infrastructure panel shows the resolved requirement alongside the placement decision — which node won, and why.

Chat session details showing the requested vCPU, memory, and disk, the observed and configured hardware, the current compute pool with its strategy, and a saved placement decision explaining why a warm Hetzner node was reusedChat session details showing the requested vCPU, memory, and disk, the observed and configured hardware, the current compute pool with its strategy, and a saved placement decision explaining why a warm Hetzner node was reused Every placement decision is explained, not just made

One fleet, every provider

The Nodes page lists every machine across every connected provider side by side, with health, CPU/memory/disk usage, and price. Follow-up work reuses a warm node instead of waiting on a fresh boot.

Nodes page listing Hetzner, Scaleway, and Vultr machines side by side with status, health, observed and configured hardware, offering price, CPU, memory, and disk usage, and the workspaces on eachNodes page listing Hetzner, Scaleway, and Vultr machines side by side with status, health, observed and configured hardware, offering price, CPU, memory, and disk usage, and the workspaces on each One fleet view across every connected cloud provider

Token and compute usage, per user

Settings → Usage shows your AI usage by model and by day, against a monthly budget, for traffic routed through the SAM AI proxy. It also tracks your compute usage in node-hours against your quota. Admins get the installation-wide view — LLM spend by model, day, and user with a projection, plus per-user node usage — in Admin → Costs and Admin → Usage.

Settings → Usage page showing total cost, requests, and input and output tokens, a per-model breakdown, a daily cost trend, and budget controls with daily token limits and a monthly cost capSettings → Usage page showing total cost, requests, and input and output tokens, a per-model breakdown, a daily cost trend, and budget controls with daily token limits and a monthly cost cap Token usage by model and by day, against a monthly budget
Previous Comments Next Schedules & Triggers

Ready to try SAM?

SAM is open source. To deploy it, you need a Cloudflare Workers Paid plan, a domain on Cloudflare, and a GitHub account.

Deploy SAM Read the Docs