nemotron-3.5-super:120b-a12b-bf16

260 Downloads Updated 3 hours ago

Nemotron 3.5 Super is NVIDIA’s vision-language model for coding, reasoning, tool use, and document understanding

vision tools thinking 120b

ollama run nemotron-3.5-super:120b-a12b-bf16

curl http://localhost:11434/api/chat \
  -d '{
    "model": "nemotron-3.5-super:120b-a12b-bf16",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='nemotron-3.5-super:120b-a12b-bf16',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'nemotron-3.5-super:120b-a12b-bf16',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 3 hours ago

3 hours ago

45c225544aa0 · 249GB

model

archnemotron_h_moe

·

parameters124B

·

quantizationBF16

247GB

projector

archclip

·

parameters842M

·

quantizationBF16

1.7GB

license

OpenMDW License Agreement, version 1.1 (OpenMDW-1.1) By exercising rights granted to you under this

2.6kB

params

{ "draft_num_predict": 3, "temperature": 1, "top_p": 0.95 }

53B

Readme

Nemotron 3.5 Super

Nemotron 3.5 Super is NVIDIA’s vision-language model for coding, reasoning, tool use, and document understanding. It has 120 billion total parameters, with 12 billion active per token, and combines a hybrid Mamba-2/Transformer mixture-of-experts backbone with a C-RADIO vision encoder. NVIDIA model card

Vision: ask questions about images, charts, screenshots, and document pages.
Tools: build applications that call functions and incorporate their results.
Thinking: choose whether to generate a reasoning trace before the answer.
Long context: the model supports up to 1 million tokens. Available memory and runtime settings determine the usable context; prompts, images, history, tool results, and generated tokens share that budget. Model design

The architecture also includes Multi-Token Prediction (MTP) for speculative decoding. NVIDIA publishes BF16 and NVFP4 checkpoints, along with a separate LiLiCorr drafter.

License

NVIDIA publishes Nemotron 3.5 Super VL under the OpenMDW License Agreement, version 1.1. See the publisher’s model card for training details and evaluation results.

# Nemotron 3.5 Super

Nemotron 3.5 Super is NVIDIA's vision-language model for coding, reasoning, tool use, and document understanding. It has **120 billion total parameters, with 12 billion active per token**, and combines a hybrid Mamba-2/Transformer mixture-of-experts backbone with a C-RADIO vision encoder. [NVIDIA model card](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16)

- **Vision:** ask questions about images, charts, screenshots, and document pages.
- **Tools:** build applications that call functions and incorporate their results.
- **Thinking:** choose whether to generate a reasoning trace before the answer.
- **Long context:** the model supports up to **1 million tokens**. Available memory and runtime settings determine the usable context; prompts, images, history, tool results, and generated tokens share that budget. [Model design](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16#model-design)

The architecture also includes Multi-Token Prediction (MTP) for speculative decoding. NVIDIA publishes [BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16) and [NVFP4](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4) checkpoints, along with a separate [LiLiCorr drafter](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4-LiLiCorr).

## License

NVIDIA publishes Nemotron 3.5 Super VL under the [OpenMDW License Agreement, version 1.1](https://openmdw.ai/license/1-1/). See the [publisher's model card](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16) for training details and evaluation results.

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)