Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1 needs about 82.2 GB of memory at Q4_K (a 69.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. Longer context needs more: at 32K tokens the KV cache alone adds about 47.2 GB.
Nemotron-3-Super-120B LLM with 120B parameters and 12B active experts
Nemotron-3-Super-120B is a large language model with 121B total parameters and 12B active parameters in a mixture‑of‑experts architecture. It supports a context length of 1,048,576 tokens and is optimized for reasoning and agentic tasks.
Good at
- Large context reasoning
- Agentic task execution
- High‑volume conversational workloads
Limits
- Requires 82.2 GB VRAM at 8k context
- No vision support
- License governed by NVIDIA Nemotron Open Model License
Best for
- IT ticket automation
- Collaborative agent development
- High‑volume chat services
Who it’s for. Developers building agentic applications that need extensive context and reasoning. The model is suitable for commercial use and large‑scale deployments.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 69.9 GB | 82.2 GB | Too big | Get file → |
The Q4_K quantization file is 69.91 GB and requires approximately 82.2 GB VRAM at 8k context. For systems with less memory, a higher‑quantization variant would be needed, but none are listed. The estimate is approximate.
What GPU do you need for Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1 quant should you choose?
Q4_K (69.9 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1
How much VRAM does Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1 need?
About 82.2 GB at Q4_K with an 8K context window. That is the 69.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 82.2 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1 GGUF?
It depends on the quant: Q4_K is 69.9 GB.
What context length does Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1 support?
The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 11.8 GB at 8K and 47.2 GB at 32K.
Where can I download Ea_Final_Nvidia_Nemotron_3_Super_120B_A12B_Bf16_030326_Vv0.1 GGUF?
The files are on Hugging Face in the ggml-org/Nemotron-3-Super-120B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
Can I run Nemotron-3-Super-120B on 8 GB VRAM?
No. The Q4_K quantization requires about 82.2 GB VRAM at 8k context, far exceeding 8 GB.
Is there a smaller quantization available?
Only the Q4_K quantization is listed; no other quantizations are provided.
What context length does the model support?
The model supports up to 1,048,576 tokens of context.