Terminal Task Execution on Terminal-Bench 1.0 (test)
34.9Avg Pass RateSkillFlow-specific
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SkillFlow-specificSkillset=SkillFlow-specific2025.04 | 34.9 | 24.2 | 0.035 | |
| No SkillsSkillset=No Skills2025.04 | 34.8 | 21.1 | 0.03 | |
| VercelSkillset=Vercel2025.04 | 32.6 | 21.2 | 0.031 | |
| SkillFlowSkillset=SkillFlow2025.04 | 31.8 | 23 | 0.034 | |
| ReptileBase Model=Devstral-2505-22B, Category=Fine-tuned Models2026.02 | 18.9 | — | — | |
| Terminus 2Base Model=LiteCoder-30a3b, Category=Fine-tuned Models2026.02 | 18.8 | — | — |