{"slug":"nemotron-3-5-bolt","name":"NVIDIA Nemotron 3.5 Bolt 74B-A9B","nameZh":"NVIDIA Nemotron 3.5 Bolt（74B-A9B，24GB 本地版）","category":"models","categoryName":"Open-weight models","audience":["ai-era"],"primary":"ai-era","openClosed":"open","pricing":"open-source","license":"OpenMDW-1.1","maintained":"yes","officialUrl":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB","repo":null,"latestVersion":null,"latestRelease":null,"stars":null,"starsAsOf":null,"description":"NVIDIA’s vision-language model for running everyday agentic tasks locally: Mamba-2/Transformer hybrid LatentMoE, 74B total / 9B active, up to 256K context, text/image/video in; native 2-bit (QAT) packing fits 24GB, deployed with vLLM on A100 / H100 / Blackwell per the card. Commercial use allowed (OpenMDW-1.1)","descriptionZh":"NVIDIA 面向“本地跑日常 agent 任务”的视觉语言模型：Mamba-2 与 Transformer 混合的 LatentMoE，总参 74B、激活 9B，最长 256K 上下文，输入文本/图像/视频；用量化感知训练的原生 2-bit 打包到 24GB，按模型卡用 vLLM 在 A100 / H100 / Blackwell 上部署。可商用（OpenMDW-1.1）","callMethods":[],"platforms":[],"addedAt":"2026-10-20","updatedAt":null,"topics":[],"language":null,"lastPush":null,"formerly":null,"packages":[],"recentVersions":[],"alternatives":["deepseek-v3","qwen3","gpt-oss","qwen3-coder","kimi-k2","llama-models","ultralytics","gemma"],"stacks":[],"scenarios":[],"problems":[],"reasonZh":null,"sources":["editorial"],"record":"https://aimap.page/api/v1/software/nemotron-3-5-bolt.json","pages":{"en":"https://aimap.page/en/software/nemotron-3-5-bolt/","zh-hans":"https://aimap.page/zh-hans/software/nemotron-3-5-bolt/"}}