Language

Nemotron 3 Ultra 550B-A55B (MoE)

NVIDIA Nemotron 3 Ultra — 550B total / 55B active hybrid Transformer-Mamba MoE with Latent MoE and a built-in Multi-Token-Prediction head; up to 1M context.

NVIDIA Open Model, Weights & DataSource verified

Specification

Model ID

nemotron-3-ultra-550b-a55b

Family

nemotron

Modality

Language

Parameters

550B (MoE, 55B active)

Context

1,048,576 tokens

Quantization

UD-Q4_K_XL

Weights

307.3 GB

Minimum RAM

360 GB

Source

Weights and provenance.

Repository

unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF

Registry license

NVIDIA Open Model, Weights & Data

Source license

other

Access

Open — weights fetch without accepting additional terms.

The registry records this model as NVIDIA Open Model, Weights & Data while the source repository states other. Both are shown here rather than one being preferred. Treat the stricter of the two as binding until the discrepancy is resolved upstream.

Serve it

# Pull the weights onto a node
tenzro model download nemotron-3-ultra-550b-a55b

# Serve it
tenzro model serve nemotron-3-ultra-550b-a55b

# Or query the registry entry over JSON-RPC
curl https://rpc.tenzro.xyz \
  -X POST -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,
       "method":"tenzro_getModel",
       "params":[{"model_id":"nemotron-3-ultra-550b-a55b"}]}'

Same family

Nemotron 3 Nano Omni 30B-A3B (MoE, multimodal)

30B (MoE, 3B active)

Nemotron 3 Nano 30B-A3B (MoE)

30B (MoE, 3B active)

Nemotron 3 Nano 4B

← All models