Nemotron 3.5 Bolt vs Nemo 3.5 Super vs Qwen 3.8 27B/FN
Pulling the benchmarks from the model cards. Benchmarks might not be exactly comparable, as some models do not specify the version. Benchmark 3.5 Bolt 3.5 Super 3.8 27B 3.8 FN DeepSWE N/A 24.65 42.2 58.7 GPQA-Diamond 86.43 89.52 89.2 91.7 SWE-bench Pro N/A 51.78 61.7 62.5 Terminal Bench 2.1 50.98 66.70 73.0 N/A HLE 25.30 29.20 30.8 35.9 LiveCodeBench v6 N/A 88.22 90.3 91.9 IFBench N/A 70.86 79.5 81.3 AA Intelligence 21 26 34 40
评论
?
参与讨论