llamaperf

Nemotron-3-Ultra

NVIDIA · 1 report

Nemotron-3-Ultra VRAM requirements by size and quant →

How does Nemotron-3-Ultra run on your hardware?

Pick your GPU or Mac, then read the speeds people reported for Nemotron-3-Ultra on it, or estimate memory fit and speed.

Free to use. No account needed. Memory estimates and community measurements are labelled separately.

Run Nemotron-3-Ultra yourself? Add your numbers.

Your coding agent can run a speed test on your machine and send it here with a free account. Or paste a result you've already posted.

Reported performance for Nemotron-3-Ultra

Filter this model’s reports by setup →
One report for this model so far, so treat the figures as early. Add yours.

Speeds are as each source reported them. GPU count, offloading and concurrent requests change them, and prompt-processing speed is input throughput, not time to first token. Check the full setup before comparing two reports.

Tone: positiveNVIDIA hardware
generation:
5.2 tokens/s
prompt processing (prefill):
120.0 tokens/s
quant:
UD-Q2_K_XL (GGUF)

User reports Nemotron-3-Ultra-550B-A55B at ~5.2 tok/s decode and ~120 tok/s prefill across 2x DGX Spark (GB10, 128GB unified each) linked over 200GbE ConnectX-7 RoCE. Setup is llama.cpp upstream master with UD-Q2_K_XL GGUF (~188 GiB, 6 shards), layer-split via RPC with --tensor-split 1,1, --no-mmap and -fit off, ~95 GiB per node. User notes decode is RPC-round-trip-bound and slower per-token than a single node; the model only fits across two boxes. A 600-token reasoning sample took 117s.

Sep 28, 2026

Get a weekly email of new Nemotron-3-Ultra reports on any GPU.

Email me new reports

Reported configurations

Individual observations from this page, not expected speeds or a ranking. Select a model to open its report, with the source and offloading conditions. Reported context may be a limit rather than actual input length.

Model / hardwareQuant / engineReported contextGeneration
Nemotron-3-Ultra 550B (55B active)
2× NVIDIA DGX Spark
UD-Q2_K_XL
llama.cpp
Not reported5.2 tokens/s