Nemotron550B paramspopular
NVIDIA Nemotron 3 Ultra 550B A55B BF16 locally
NVIDIA Nemotron 3 Ultra 550B A55B BF16 is an open-weight LLM from the Nemotron family with 550B parameters. Main use: chat, RAG and general assistance. Detected minimum hardware: Mac Studio M3 Ultra (512GB) (512 GB).
Technical facts
Parameters550B
Q4_K_M345.8 GB
Q5_K_M422.6 GB
Q8614.7 GB
FP161229.3 GB
FamilyNemotron
Last sync2026-08-01
Available quantizations
GGUF weightsQ4_K_M
345.8GB
Acceptable. Good compromise when VRAM is limited.
Q5_K_M
422.6GB
Good quality. Sweet spot for size and precision.
Q8
614.7GB
Near-FP16 quality. Comfortable for production.
FP16
1229.3GB
Reference precision. Maximum quality, doubled VRAM.
Compatible GPUs
1 single-GPUGPUs that can run NVIDIA Nemotron 3 Ultra 550B A55B BF16 on a single card, ranked by VRAM headroom.
Similar models
VRAM estimates: parameters x bits/8 plus margin. Real performance varies by engine, context length and batch size.
sync: 2026-08-01