UTC

Nemotron 3 Ultra 550B A55B

NVIDIA · open weights · served by 34 providers · tool calling, reasoning

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

Weights
Open
Released
2026-06-04
Knowledge cutoff
2025-09
Context window
1M
Deprecation dates

When each platform deprecates Nemotron 3 Ultra 550B A55B, as LiteLLM records it. Plan the move before the date: after it, the platform may stop serving the model or stop updating it.

  • Together AIdeprecated 2026-08-27together_ai/nvidia/nemotron-3-ultra-550b-a55bsuccessor: together_ai/nvidia/NVIDIA-Nemotron-3.5-Super-VL
Every upcoming deprecation →
Hugging Face repository

nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 at commit 02462641f1, last changed 2026-08-24

Licence
openmdw-1.1
Access
Open download
Library
transformers
Formats
.safetensors
  • No model signature file is published with the weights.

What the Hub's scanners report

20 weight files, 18 with a scan result. No scanner reports an issue with them.

The model at this commit, its 20 weight files with their SHA-256.

Facts from the Hugging Face Hub and its file scanners (Protect AI, Cisco, Hugging Face picklescan, VirusTotal, JFrog), as of 2026-11-18. ShieldMap does not copy model cards.

Who serves it
  • AIHubMixnemotron-3-ultra-550b-a55b-free
  • Basetennvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B
  • Bothubnemotron-3-ultra-550b-a55b:free
  • CommandCodenvidia/nemotron-3-ultra-550b-a55b
  • DigitalOceannemotron-3-ultra-550b
  • Doublewordnvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4
  • Eden AIarkanecloud/nvidia/nemotron-3-ultra-550b-a55b
  • Eden AIdeepinfra/nemotron-3-ultra-550b-a55b
  • Eden AInebius/nvidia/Nemotron-3-Ultra-550b-a55b
  • InferAllnvidia/nemotron-3-ultra-550b-a55b
  • Kelnnvidia/nemotron-3-ultra-550b-a55b
  • Kenarinemotron-3-ultra-550b-a55b
  • Kenarinemotron-3-ultra-550b-a55b:free
  • Kilo Gatewaynvidia/nemotron-3-ultra-550b-a55b
  • Kilo Gatewaynvidia/nemotron-3-ultra-550b-a55b:free
  • DevPass (LLM Gateway)nemotron-3-ultra-550b
  • NanoGPTnvidia/nemotron-3-ultra-550b-a55b
  • NanoGPTnvidia/nemotron-3-ultra-550b-a55b:thinking
  • Nebius Token Factorynvidia/Nemotron-3-Ultra-550b-a55b
  • Nvidianvidia/nemotron-3-ultra-550b-a55b
  • Ollama Cloudnemotron-3-ultra
  • OpenCodenemotron-3-ultra-freedeprecated
  • OpenRouternvidia/nemotron-3-ultra-550b-a55b
  • OpenRouternvidia/nemotron-3-ultra-550b-a55b:free
10 more providers

Requesty (nemotron-3-ultra-550b-a55b), Requesty (nvidia-nemotron-3-ultra), routing.run (nemotron-3-ultra), Thinking Machines (nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16), Thinking Machines (nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16:peft:262144), Together AI (nvidia/nemotron-3-ultra-550b-a55b), UnoRouter (nemotron-3-ultra-550b-a55b:free), Venice AI (nvidia-nemotron-3-ultra-550b-a55b), Vercel AI Gateway (nvidia/nemotron-3-ultra-550b-a55b), CoreWeave (nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B)

Running it yourself

Open weights are loaded and served by code that has its own advisories. These inference servers and model loaders are on ShieldMap; load weights only from the publisher's own repository, and prefer safetensors over formats that can run code.

How supply-chain attacks on models work →

Sources: models.dev (MIT) and the LiteLLM model catalogue (MIT). Deprecation dates are as each platform published them and as LiteLLM records them; check the platform before you rely on one. Data as of 2026-11-22.