AWS S3 vs TensorlakeStorage Benchmark Comparison
AWS S3 was faster in 1 of 2 storage measurements with a speed verdict, and 1 was roughly even.
- Shared storage measurements
- 3
- With a speed verdict
- 2
- Last run
- October 30, 2026
Scorecard
Median of each storage measurement's primary metric; lower is better. Medians within 5% count as roughly even, and a storage measurement gets no speed verdict when either provider completed fewer than 50% of runs.
| Benchmark | AWS S3 | Tensorlake | Result | Ratio |
|---|---|---|---|---|
| Upload (10 MB) Upload time · Run October 30, 2026 | 490ms 100% success | 491ms 100% success | Roughly even | 1.00x |
| Download (10 MB) Download time · Run October 30, 2026 | 137ms 100% success | 149ms 100% success | AWS S3 faster | 1.09x |
| Snapshot create Snapshot create time · Run October 30, 2026 | No timing data 0% success | 0.05s 100% success | No verdict No speed verdict because AWS S3 completed 0% of runs, fewer than the 50% required. | — |
Upload (10 MB)
Run October 30, 2026Time to upload a 10 MB file to the provider, measured in the 10 MB storage lifecycle benchmark.
Upload time
The medians are within 5% of each other (490ms vs 491ms), so this counts as roughly even.
Median upload time over time
14 runs from July 31, 2026 to October 30, 2026.
Download (10 MB)
Run October 30, 2026Time to download the same 10 MB file back, measured in the 10 MB storage lifecycle benchmark.
Download time
AWS S3 had a 1.09x lower median download time (137ms vs 149ms for Tensorlake).
Median download time over time
14 runs from July 31, 2026 to October 30, 2026.
Snapshot create
Run October 30, 2026Time to create a snapshot of a bucket holding a small dataset of 10 objects.
Snapshot create time
No speed verdict because AWS S3 completed 0% of runs, fewer than the 50% required.
View the full Snapshot create leaderboard →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 the 10 MB storage lifecycle benchmark reliably: at least 50% success for each provider, in benchmark runs from the last 30 days.
Frequently asked questions
Have another question? Email us.
AWS S3 was faster in 1 of 2 storage measurements with a speed verdict, and 1 was roughly even. In Upload (10 MB) (October 30, 2026), AWS S3 had a median upload time of 490ms and Tensorlake 491ms, which counts as roughly even. In Download (10 MB) (October 30, 2026), AWS S3 had a median download time of 137ms versus 149ms for Tensorlake, 1.09x faster.