Superserve vs Tensorlake

Sandbox Benchmark Comparison · Updated October 30, 2026

Superserve
Faster in
2of 2 benchmarks
#2 of 32 in Sandbox Provider Rating
vs
Tensorlake
Faster in
0of 2 benchmarks
#7 of 32 in Sandbox Provider Rating

Superserve was faster in both shared sandbox benchmarks.

Head to head

  • Burst TTITime to interactive · lower is better
    Superserve faster2.77x
    Superserve
    371ms
    Tensorlake
    1.03s
  • DAXTotal workload time · lower is better
    Superserve faster1.22x
    Superserve
    64.26s
    Tensorlake
    78.16s

Burst TTI

October 30, 2026
Full leaderboard

Superserve

100% success
Median
371ms
p95
655ms
p99
705ms
Superserve 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

26 runs · July 7, 2026 to October 30, 2026
SuperserveTensorlake

DAX

Full leaderboard

Superserve

100% success · October 30, 2026
Median
64.26s
Superserve DAX results

Tensorlake

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

Median total workload time over time

15 runs · July 24, 2026 to October 23, 2026
SuperserveTensorlake

Cost per DAX run

Superserve

Per-second wall clock
$0.012per run
64.26s median at $0.6624/hr

$0.0504/vCPU-hr + $0.0162/GiB-hr + storage $0.000108/GiB-hr.

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 Superserve 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.
Read the full methodology →Raw results on GitHub ↗

Frequently asked questions

Have another question? Email us.

Superserve was faster in both shared sandbox benchmarks. In Burst TTI (October 30, 2026), Superserve had a median time to interactive of 371ms versus 1.03s for Tensorlake, 2.77x faster. In DAX (October 23, 2026), Superserve had a median total workload time of 64.26s versus 78.16s for Tensorlake, 1.22x faster.

PartnersGoogle Cloud Run