List your company

Tool Use on Terminal Bench Hard

36AccuracyGemma 4 31B
1.6810.5919.528.41Jul 2, 2026

Evaluation results

MethodLinks
Gemma 4 31BModel size=31B, Thinking mode=true2026.0736
Gemma 4 12BModel size=12B, Thinking mode=true2026.0718
Gemma 4 26B-A4BModel size=26B-A4B, Thinking mode=true2026.0714
Gemma 4 E4BModel size=E4B, Thinking mode=true2026.078
Gemma 3 27BModel size=27B, Thinking mode=false2026.074
Gemma 4 E2BModel size=E2B, Thinking mode=true2026.073