Agent37 vs Kosa8

Sandbox Benchmark Comparison · Updated November 13, 2026

Agent37
Faster in
0of 1 benchmark
#26 of 33 in Sandbox Provider Rating
vs
Kosa8
Faster in
1of 1 benchmark
#33 of 33 in Sandbox Provider Rating

Kosa8 was faster in the only sandbox benchmark with a speed verdict.

Head to head

  • Burst TTI
    Time to interactive · lower is better
    Kosa8 faster1.37x
    Agent37
    2.51s
    Kosa8
    1.83s
  • DAX
    Total workload time · lower is better
    No verdict
    Agent37
    193.27s
    Kosa8
    —

Burst TTI

November 13, 2026
Full leaderboard

Agent37

100% success
Median
2.51s
p95
4.01s
p99
4.09s
Agent37 Burst TTI results

Kosa8

100% success
Median
1.83s
p95
2.50s
p99
3.14s
Kosa8 Burst TTI results

DAX

November 13, 2026
Full leaderboard

Agent37

100% success
Median
193.27s
Agent37 DAX results

Kosa8

0% success

No measured time in this run.

Kosa8 DAX results
Work at Agent37 or Kosa8? 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 when both providers have at least one successful run in the same benchmark from the last 30 days. A benchmark only gets a verdict when both providers completed at least 50% of its runs. 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.

Kosa8 was faster in the only sandbox benchmark with a speed verdict. In Burst TTI (November 13, 2026), Kosa8 had a median time to interactive of 1.83s versus 2.51s for Agent37, 1.37x faster.

Both providers have results in 2 sandbox benchmarks: Burst TTI and DAX. 1 of them has a speed verdict.

Success rates in the latest runs: Burst TTI: Agent37 100%, Kosa8 100%; DAX: Agent37 100%, Kosa8 0%. DAX has no speed verdict because Kosa8 completed fewer than half of its runs; Agent37's median total workload time was 193.27s.

Burst TTI and DAX results are from November 13, 2026.