Tensile
00
Tensile Sandbox

A fresh microVM
for every run.

Tensile Sandbox is state-of-the-art Firecracker microVM isolation — sub-second boots and hard, kernel-level separation — the flagship substrate for agents, evals, RL rollouts, and load tests on Tensile. It runs on your own hardware, and it's the drop-in replacement for Modal. Training and inference schedule onto your Kubernetes/Volcano or Slurm cluster instead — the sandbox is for isolated, untrusted work.

isolation
Firecracker microVM
boot p50
~180 ms
runs on
your infra
01Why microVMs

Containers share a kernel. Untrusted AI workloads shouldn't.

Agents run arbitrary code, evals pull arbitrary models, rollouts touch the network. A shared-kernel container is one escape away from the host. Firecracker gives every run its own kernel and a minimal attack surface — the same tech that isolates serverless at scale — but it boots in milliseconds, so isolation costs you nothing.

A

Hardware-level isolation

A dedicated guest kernel per run. Tenant and run boundaries are enforced by the VMM, not by namespaces.

B

Sub-second cold starts

Minimal microVMs boot in ~180ms — a fresh, clean machine per run with no warm-pool bookkeeping.

C

Small blast radius

A stripped guest with no extra daemons. Less to attack, less to patch, nothing shared with the next tenant.

02Lifecycle

Provision, run, stream, tear down.

01Provision

Boot a microVM from an image, pinned to your tenant and run id.

02Execute

Run the workload harness inside the guest — agent, eval, or rollout.

03Stream

Logs, spans, and artifacts stream back to the control plane live.

04Reclaim

Tear down on exit. Nothing persists into the next run.

from tensile import Sandbox
 
# One microVM, scoped to your org + run
with Sandbox.launch(image="python:3.12", cpu=4, mem="8Gi") as vm:
vm.exec("pip install -e . && pytest -q") # streams live
art = vm.collect("./artifacts") # pulled to the registry
# → VM reclaimed; nothing shared with the next run
03What runs in it

The substrate under every workload.

04The Modal replacement

Same ergonomics. Your infra. Lower cost per verified outcome.

 Tensile SandboxHosted (Modal)
Runs onYour hardware / cloudVendor cloud
IsolationFirecracker microVMmicroVM (managed)
Data boundaryNever leaves your perimeterLeaves your perimeter
Cost modelYour capacity, metered per outcomePer-second vendor billing
Lock-inOpen substrate, self-hostableProprietary

The Workload Analyzer reports cost per verified outcome by substrate — so the Modal → Tensile move is a decision you make on evidence, family by family, not a leap of faith. See the comparison →

Run your next workload in a microVM on your own infra.