SF Compute Autoresearch Sandboxes vs Superserve

SF Compute Autoresearch Sandboxes bills memory only, at $0.04572 per GiB-hour. Superserve costs $0.0504 per vCPU-hour. SF Compute Autoresearch Sandboxes runs microVMs. Superserve runs Firecracker microVMs and has no session limit.

Updated Oct 23, 2026

SF Compute Autoresearch Sandboxes

Isolated microVMs for AI agents that fork from a saved snapshot, part of SF Compute Autoresearch.

Isolation: MicroVM
Free tier: None
Max session: Not listed

Visit site SF Compute Autoresearch Sandboxes profile

Superserve

Firecracker microVM sandboxes that pause and resume.

Isolation: Firecracker microVM
Free tier: $5 to start, plus $95 when you add billing
Max session: No limit

Visit site Superserve profile

What is different

9 fields differ. The rest are the same or not listed.

Field

SF Compute Autoresearch Sandboxes

Superserve

vCPU, per hour: SF Compute Autoresearch Sandboxes: Included; Superserve: $0.0504
RAM, per GiB-hour: SF Compute Autoresearch Sandboxes: $0.04572; Superserve: $0.0162
Storage: SF Compute Autoresearch Sandboxes: Sandbox snapshots $0.10 per GiB-month of deduplicated bytes, $0.05 per GiB-month in cold storage; Superserve: Free
Free tier: SF Compute Autoresearch Sandboxes: None; Superserve: $5 to start, plus $95 when you add billing
Billing: SF Compute Autoresearch Sandboxes: Memory only; Superserve: Per second
Largest sandbox: SF Compute Autoresearch Sandboxes: 4 vCPU · 8 GiB; Superserve: 16 vCPU · 64 GiB
Concurrent sandboxes: SF Compute Autoresearch Sandboxes: Up to 4,096; Superserve: Up to 1,000
Isolation: SF Compute Autoresearch Sandboxes: MicroVM; Superserve: Firecracker microVM
Integrations: SF Compute Autoresearch Sandboxes: MCP, Claude Code; Superserve: Claude Agent SDK, LangChain, OpenAI Agents SDK, MCP, Claude Code

Full comparison

18 fields, at each provider's list price

SF Compute Autoresearch Sandboxesautoresearch.sfcompute.com

Superservesuperserve.ai/pricingsuperserve.ai

Pricing

vCPU, per hour: SF Compute Autoresearch Sandboxes: Included; Superserve: $0.0504
RAM, per GiB-hour: SF Compute Autoresearch Sandboxes: $0.04572; Superserve: $0.0162
Storage: SF Compute Autoresearch Sandboxes: Sandbox snapshots $0.10 per GiB-month of deduplicated bytes, $0.05 per GiB-month in cold storage; Superserve: Free
Free tier: SF Compute Autoresearch Sandboxes: None; Superserve: $5 to start, plus $95 when you add billing
Paid plan from: SF Compute Autoresearch Sandboxes: Pay as you go; Superserve: Pay as you go
Billing: SF Compute Autoresearch Sandboxes: Memory only; Superserve: Per second

Limits

Max session: SF Compute Autoresearch Sandboxes: Not listed; Superserve: No limit
Idle stop: SF Compute Autoresearch Sandboxes: Not listed; Superserve: Not listed
Largest sandbox: SF Compute Autoresearch Sandboxes: 4 vCPU · 8 GiB; Superserve: 16 vCPU · 64 GiB
Concurrent sandboxes: SF Compute Autoresearch Sandboxes: Up to 4,096; Superserve: Up to 1,000

Platform

Isolation: SF Compute Autoresearch Sandboxes: MicroVM; Superserve: Firecracker microVM
GPU: SF Compute Autoresearch Sandboxes: Not listed; Superserve: Not listed
Snapshots: SF Compute Autoresearch Sandboxes:; Superserve:
Pause and resume: SF Compute Autoresearch Sandboxes:; Superserve:

Developer

SDKs: SF Compute Autoresearch Sandboxes: Not listed; Superserve: Python, TypeScript
Integrations: SF Compute Autoresearch Sandboxes: MCP, Claude Code; Superserve: Claude Agent SDK, LangChain, OpenAI Agents SDK, MCP, Claude Code

Trust

License: SF Compute Autoresearch Sandboxes: Not listed; Superserve: Not listed
BYOC: SF Compute Autoresearch Sandboxes: Not listed; Superserve: In your AWS, GCP, or Azure account. Enterprise plan.

Hourly cost

One sandbox running for one hour

Size

SF Compute Autoresearch Sandboxes

Superserve

1 vCPU · 2 GiB: SF Compute Autoresearch Sandboxes: $0.091; Superserve: $0.083
2 vCPU · 4 GiB: SF Compute Autoresearch Sandboxes: $0.183; Superserve: $0.166
4 vCPU · 8 GiB: SF Compute Autoresearch Sandboxes: $0.366; Superserve: $0.331

Compute only, at list rates.

Open the calculator

Which one fits

Based on the fields above

SF Compute Autoresearch Sandboxes fits when

You want one rate for memory, with CPU included.

Superserve fits when

You use the Claude Agent SDK, LangChain or the OpenAI Agents SDK.
You want the lower list price: $0.166 against $0.183 an hour for 2 vCPU and 4 GiB.
You want to start without paying.
You need a desktop in the sandbox.

More comparisons

See all comparisons

SF Compute Autoresearch Sandboxes

Superserve

What is different

Full comparison

Pricing

Limits

Platform

Developer

Trust

Hourly cost

Which one fits

SF Compute Autoresearch Sandboxes fits when

Superserve fits when

More comparisons

Related

Compare

Pricing

Browse