Sandbox0 vs Tensorlake

Sandbox Benchmark Comparison · Updated October 30, 2026

Tensorlake was faster in both shared sandbox benchmarks.

Head to head

  • Burst TTITime to interactive · lower is better
    Tensorlake faster10.0x
    Sandbox0
    10.28s
    Tensorlake
    1.03s
  • DAXTotal workload time · lower is better
    Tensorlake faster1.60x
    Sandbox0
    124.96s
    Tensorlake
    78.16s

Burst TTI

October 30, 2026

Sandbox0

100% success
Median
10.28s
p95
22.79s
p99
23.24s
Sandbox0 Burst TTI results

Tensorlake

100% success
Median
1.03s
p95
2.24s
p99
5.25s
Tensorlake Burst TTI results

Median time to interactive over time

17 runs · July 31, 2026 to October 30, 2026
Sandbox0Tensorlake

Sandbox0

100% success · October 30, 2026
Median
124.96s
Sandbox0 DAX results

Tensorlake

100% success · October 23, 2026
Median
78.16s
Tensorlake DAX results

Median total workload time over time

12 runs · July 31, 2026 to October 23, 2026
Sandbox0Tensorlake

Cost per DAX run

Sandbox0

Per GiB-hour (wall clock)
$0.0083per run
124.96s median at $0.2400/hr

$0.015/GiB-hr on the memory limit (16 GiB), billed by the second. Paused sandboxes stop compute charges.

Source pricing page

Tensorlake

Active CPU + allocated RAM
$0.017per run
78.16s median at $0.8000/hr

$0.07/core-hr active CPU + $0.015/GB-hr RAM (on-demand rates; Pro is ~40% lower).

Source pricing page
Compare sandbox costs across all providers →
Work at Sandbox0 or Tensorlake? See where you stand every day, with full run logs, not just on benchmark Friday.Get stealth benchmarks →

Methodology

  • One workload, every provider. The same script, payload, and model hit every provider, addressed the way its own API expects. Nobody gets a tuned or special path.
  • Repeated on a schedule. Every suite runs many iterations per provider, automated on a recurring schedule, so results reflect steady state rather than one lucky request.
  • Raw results published. Every run is committed as JSON to the public benchmarks repo, next to the code that produced it. Nothing is cherry-picked or held back.
  • How this comparison is built. Each benchmark compares the two providers' medians; results within 5% count as roughly even. A comparison is published only when both providers completed at least 50% of runs in 2 or more shared benchmarks run in the last 30 days. If a provider’s latest run of a benchmark failed, its most recent successful run from that window is used instead.

Frequently asked questions

Have another question? Email us.

Tensorlake was faster in both shared sandbox benchmarks. In Burst TTI (October 30, 2026), Tensorlake had a median time to interactive of 1.03s versus 10.28s for Sandbox0, 10.0x faster. In DAX (October 23, 2026), Tensorlake had a median total workload time of 78.16s versus 124.96s for Sandbox0, 1.60x faster.

PartnersGoogle Cloud Run