Nemotron 3 Nano 4B VRAM requirements

How much VRAM does Nemotron 3 Nano 4B need?

Nemotron 3 Nano 4B needs 3.10 GB at Q4_K_M, 4.96 GB at Q8_0 and 8.78 GB at FP16 with 8,192 tokens of context and one request. Each total below is the weights plus the FP16 KV cache and the runtime overhead (0.5 GB plus 10%), and each row opens the calculator with that setting. What the GGUF names mean.

Fine-tuning Nemotron 3 Nano 4B? Nemotron 3 Nano 4B VRAM for LoRA, QLoRA and full training.

Nemotron 3 Nano 4B GGUF files on Hugging Face

The 8 GGUF files llama.cpp picks for Nemotron 3 Nano 4B with -hf <repo>:<QUANT>, with the bytes the Hugging Face file list reported on 2026-09-29. Each total is that file plus the FP16 KV cache and 0.5 GB + 10%, one request. "vs estimate" compares the file with the bits-per-weight size the table above uses.

GB here is GiB (1024³ bytes), as everywhere on this site; Hugging Face shows the Q4_K_M file as 2.9 GB, in 10⁹ bytes.

Which Nemotron 3 Nano 4B GGUF fits a 16, 24, 32, 48 or 80 GB GPU or a Mac

From the file sizes above: the largest file that fits each machine with an 8K context and 0.5 GB left free, the longest context it then has room for, and the longest context for Q4_K_M (2.70 GB). One request, FP16 KV cache.

Run Nemotron 3 Nano 4B with llama.cpp or Ollama

llama-server -hf unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF:Q4_K_M -c 262144

NVIDIA-Nemotron-3-Nano-4B-Q4_K_M.gguf, 2.9 GB, from unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF (checked 2026-09-29). On a 24 GB RTX 3090 / 4090, Q4_K_M with -c 262144 takes 7.87 GB and leaves 16.1 GB. The file is 473 MB over the Q4_K_M estimate, so -c counts the file.

ollama run nemotron-3-nano:4b

Ollama library tag nemotron-3-nano:4b; its quant is Ollama's choice. Same context: /set parameter num_ctx 262144.

Nemotron 3 Nano 4B at 8K, 32K, 128K and 256K (full) tokens of context

4 of its 42 layers use full attention and 38 are Mamba or feed-forward layers with no growing cache. Each extra token of context adds 16 KB of FP16 cache per request. The last column is the smallest setup here, one card or a group, that holds it at Q4_K_M. How the KV cache works.

How fast Nemotron 3 Nano 4B writes

Tokens per second for one request with 8,192 tokens of context, estimated from memory bandwidth. A dash means it does not fit on one card. Try other settings in the speed calculator.

Longest context on one GPU

How many tokens of context Nemotron 3 Nano 4B fits on a single card with one request, an FP16 KV cache and 0.5 GB left free. "Full" means the model's whole context window fits.

Model details

Parameters

4.0B (3,973,556,832)

Layers

4 of its 42 layers use full attention and 38 are Mamba or feed-forward layers with no growing cache

Attention cache

8 KV heads × 128

Context length

262,144 tokens

Published weights

7.40 GB (BF16)

On Hugging Face

nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16

Reported benchmarks

GPQA 53.2 · LCB 51.8 (reasoning on). Vendor-reported: from the maker’s own model card, not rerun here; read 2026-09-30.

Nemotron 3 Nano 4B next to similar-sized models

The three models here closest to it in parameter count, at Q4_K_M with 8,192 tokens of context.