STT-Agent-4B (SFT+RL): benchmark results
Provider: Other.
Aggregate Rating 1709 ± 42, rank #308 of 10916 rated models, from 5 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| STT-Arena - Impossible | 64.44 | Pass@1 (%) on the 30 impossible tasks (no valid completion p | 100 |
| STT-Arena | 27.17 | Overall (self-reported) | 65 |
| STT-Arena - Hard | 14.12 | Pass@1 (%) on the 59 hard tasks (long-horizon cascading cons | 57.5 |
| STT-Arena - Easy | 29.11 | Pass@1 (%) on the 71 easy tasks (a single isolated conflict | 55 |
| STT-Arena - Medium | 19.9 | Pass@1 (%) on the 67 medium tasks; a task passes when every | 55 |
Interactive version: theaggregate.ai/model?slug=stt-agent-4b-sft-plusrl · How It Works · Data refreshed daily, snapshot 2026-11-08.