Updated

Nemotron 3.5 Super VL and Bolt (Bolt E4B 4GB now available)

NVIDIA released Nemotron 3.5 Super VL and Bolt 74B-A9B 24GB on October 15, then released Nemotron 3.5 Bolt E4B 4GB on October 23. The family targets multimodal agent workflows and local conversational agents; NVIDIA describes E4B as a small open model with multi-turn tool use and a 64K context window.

By No Code MBA · Facts checked

Provider
NVIDIA
Announcement date
Status at last check
Generally available
Model family
Nemotron 3.5
Capabilities
Text, Image, Coding

What changed

NVIDIA released three Nemotron 3.5 models. Super VL is a 120B (12B active) open vision-language model for images, webpages, PDFs, tables and charts, with up to 1M tokens of context; NVIDIA says it matches its largest reasoning model on multimodal agent tasks at about one fifth the size. Bolt 74B-A9B uses native 2-bit quantization to fit in 24GB. Bolt E4B 4GB, released October 23, is an 8B (4B effective) text model post-trained from Gemma 4 E4B with a 64K context.

Access and availability

Open weights on Hugging Face (for example nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 and nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB) under the OpenMDW 1.1 license, which permits commercial use. Super VL is also on OpenRouter, ModelScope and NVIDIA NIM; Bolt models run in llama.cpp, Ollama, LM Studio and vLLM.

Official access documentation

What you can build

Evaluate the available Super VL or Bolt model on a representative visual-document or local-agent task, and compare accuracy, latency, and hardware needs with your current setup.

Limitations

Bolt E4B is text-only, and NVIDIA notes that restricting its reasoning budget can cut off reasoning mid-sentence. NVIDIA positions Bolt 74B-A9B as below the most capable models, and Super VL deployment recipes may use less than 1M context depending on GPU memory.

Sources and release history