NVIDIA 发布 Nemotron 3.5 Super 技术报告
NVIDIA 的 Nemotron 3.5 Super 技术报告称，该模型在 SWE-bench Pro 上达到 51.78%，以约少 4.6 倍的活跃参数超过 Nemotron 3 Ultra。
Great report from NVIDIA.
They report that Nemotron 3.5 Super reaches 51.78% on SWE-bench Pro. That beats Nemotron 3 Ultra with about 4.6x fewer active parameters.
Super has 120B parameters with 12B active, and it handles text, images, and video. Lightning is a 30B text model with 3B active.
The technical report covers every stage of training. Super gets mid-training on agentic software engineering, then supervised fine-tuning, RL, and multi-teacher on-policy distillation.
Serving numbers are also included. The stack reaches up to 4,856 output tokens/s/GPU on GB200 while holding at least 50 tokens/s/user. NVFP4 quantization recovers 99.62% median accuracy relative to BF16.
arxiv.org/abs/2610.38095
Chat with Paper: academy.dair.ai/papers/nemot…
来源：dair_ai · x.com