cd /news/large-language-models/unsloth-nvidia-nemotron-3-ultra-550b… · home topics large-language-models article
[ARTICLE · art-55080] src=huggingbay.xyz ↗ pub= topic=large-language-models verified=true sentiment=· neutral

unsloth/nvidia-nemotron-3-ultra-550b-a55b-nvfp4

NVIDIA released the Nemotron-3-Ultra-550B-A55B-NVFP4 model, a 550-billion-parameter large language model with 55 billion active parameters using NVFP4 quantization, under the OpenMDW-1.1 license. The model, hosted externally on ModelScope and indexed by Hugging Bay, requires multi-GPU or high-RAM hardware and is pending safety and license review.

read1 min views1 publishedJul 3, 2026
unsloth/nvidia-nemotron-3-ultra-550b-a55b-nvfp4
Image: Huggingbay (auto-discovered)

unsloth/nvidia-nemotron-3-ultra-550b-a55b-nvfp4: NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 License: openmdw-1.1. external modelscope metadata. Scan: pending.

  • License

  • openmdw-1.1

  • Size

  • 328.2 GB

  • Demand

  • 208 upstream downloads

  • Trust

  • Needs review 0/100

  • Trusted download

  • External or needs review

  • Run fit

  • Large hardware fit: Multi-GPU, high-RAM machine, or hosted inference

  • Files

  • 0 hosted / 0 total

  • Scan status

  • pending

  • Hosting status

  • external

  • Upstream

  • unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4

  • Declared license

  • openmdw-1.1

  • Pipeline

  • text-generation

  • Library

  • pytorch

  • Task categories

  • text-generation

About this artifact #

External ModelScope model metadata for unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4. Related arXiv: . Model files stay upstream; Hugging Bay indexes this metadata for discovery and provenance only.

Trusted download #

External or needs review. No Hugging Bay hosted download is available until source, license, and safety checks pass.

  • Hosted files
  • 0 of 0
  • Reviewed peer-assisted fallbacks
  • 0
  • Download plan
[Download-plan JSON](/api/artifacts/modelscope-unsloth-nvidia-nemotron-3-ultra-550b-a55b-nvfp4/download-plan)- Trust bundle
[Trust-bundle JSON](/api/trust-bundles/modelscope-unsloth-nvidia-nemotron-3-ultra-550b-a55b-nvfp4)- Review summary
[Review-summary JSON](/api/artifacts/modelscope-unsloth-nvidia-nemotron-3-ultra-550b-a55b-nvfp4/review-summary)- Distribution
[Reviewed peer-fallback JSON](/api/artifacts/modelscope-unsloth-nvidia-nemotron-3-ultra-550b-a55b-nvfp4/distribution)

Can I run this? #

Large hardware fit. Multi-GPU, high-RAM machine, or hosted inference. Use compare pages to find smaller verified alternatives or request a quantized mirror if one is not hosted yet.

  • Estimated footprint
  • 328.2 GB
  • Suggested runners
  • Transformers, vLLM
  • Runtime files
  • 0
  • Local kit

Local run kit Run-fit warnings: No runtime weight file is indexed yet.; Hugging Bay is showing metadata only until reviewed hosted files exist.

- License: fail - openmdw-1.1
- Provenance: pass - unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4
- Scan: warn - pending
- Hosted files: warn - external metadata only
- Signed manifest: fail - not final yet
- Community: warn - no reviews yet
- Benchmarks: warn - no benchmark evidence

Machine-readable card #

Open interactive artifact page

── more in #large-language-models 4 stories · sorted by recency
── more on @nvidia 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/unsloth-nvidia-nemot…] indexed:0 read:1min 2026-07-03 ·