Model
Decodetok/s · greedy
HumanEval+pass@1
MBPP+pass@1
Hard8686 assertions
Laguna XS 2.1
TQ3_4S · 4.05 bpw · RTX 3090
76.2%
72.0%
64.0%
Ornith 35B
TQ3_4S · 13.0 GiB · RTX 3090
87.2%
70.9%
81.4%
Qwen3.5 9B
TQ3_4S · RTX 3090
67.1%
56.3%
44.2%
Qwen3.6 27B MTP
TQ3_4S · RTX 3090
92.7%*
87.8%*
88.4%