英伟达发布 Nemotron 3.5 Super VL，并开放 Bolt 74B 端侧模型Chinese only
英伟达发布面向多步骤智能体任务的 Nemotron 3.5 Super VL，称其单 GPU 任务吞吐量最高为同规模其他开放模型的 5 倍。Super VL 与压缩至 24GB 的 Bolt 74B 模型已开放获取；面向本地语音助手等场景的 Bolt E4B 4GB 计划于 10 月 23 日推出。
1 developments3 reports3 sources
Event developments 1
NVIDIA releases Nemotron 3.5 Super VL with open model weights
NVIDIA has released Nemotron 3.5 Super VL for specialized enterprise agents. The model combines reasoning and tool use with text, image, document, and video understanding. The release includes base, post-trained, and NVFP4-quantized checkpoints; developers can further train it on their own data and workflows and deploy it in supported environments.
View source reports · 3 sources · 3 reports
All reports 3
Nvidia launches open-weight models for enterprise agents and local PCs
Nvidia has launched Nemotron 3.5 Super VL for enterprise agents and two Bolt models designed to run locally on PCs and workstations. Organizations can post-train Super VL on their own data and deploy it on their own infrastructure. Super VL and Bolt 74B-A9B are available now, while the smaller Bolt E4B is due later this month.
2 more sources reported this
- IT之家 科技新闻
- NVIDIA 开发者博客
NVIDIA releases Nemotron 3.5 Super VL and makes Bolt 74B available
NVIDIA has released Nemotron 3.5 Super VL for multistep agent tasks, claiming up to five times the single-GPU task throughput of other open models of similar size. Super VL and the Bolt 74B model, compressed to 24GB, are available; Bolt E4B 4GB, intended for uses including local voice assistants, is planned for October 23.
2 more sources reported this
- TechCrunch AI
- NVIDIA 开发者博客
NVIDIA releases Nemotron 3.5 Super VL with open model weights
NVIDIA has released Nemotron 3.5 Super VL for specialized enterprise agents. The model combines reasoning and tool use with text, image, document, and video understanding. The release includes base, post-trained, and NVFP4-quantized checkpoints; developers can further train it on their own data and workflows and deploy it in supported environments.
2 more sources reported this
- TechCrunch AI
- IT之家 科技新闻