Nemotron 3 Super — lokalny model AI od NVIDIA
Autor: Jakub Rusinowski · Ostatnia aktualizacja:
NVIDIA's 120B open Mixture-of-Experts, activating just 12B parameters per token for compute efficiency on multi-agent workloads. Like the rest of the Nemotron line it generates a reasoning trace before its final answer, and NVIDIA publishes open weights alongside training data and recipes rather than weights alone. Positioned for collaborative agents and high-volume automation such as IT ticket triage. First-party Ollama tags exist at Q4_K_M, Q8_0 and BF16, so this is genuinely self-hostable — on a multi-GPU box rather than a laptop.
Warianty
Najmniejszy wariant Nemotron 3 Super potrzebuje około 73 GB VRAM przy Q4_K_M — skwantyzowane wagi plus narzut frameworka, bez pamięci podręcznej KV.
| Model | VRAM przy Q4 | VRAM | Kontekst | Uruchom |
|---|---|---|---|---|
| Nemotron 3 Super 120B-A12B → 120B (12B active) | ~73.3 GB | 128,000 | ollama run nemotron-3-super:120b-a12b |
Pamięć to skwantyzowane wagi plus narzut przy Q4_K_M, liczone tym samym silnikiem co sprawdzarka GPU i VRAM.
Jak uruchomić Nemotron 3 Super lokalnie
Zainstaluj Ollama, a potem pobierz tag.
ollama run nemotron-3-super:120b-a12bWybierz rozmiar powyżej, aby zobaczyć jego VRAM, szacowaną szybkość i polecenie instalacji.
Licencja
Weights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms.
Dotyczy: Nemotron 3 Super 120B-A12BPolecane GPU
Najtańsze GPU z naszego katalogu, które uruchomi Nemotron 3 Super lokalnie (min. 73 GB VRAM), to NVIDIA DGX Spark (128 GB).
Czy uruchomię Nemotron 3 Super na moim GPU?
- Nemotron 3 Super na AMD Ryzen AI Max+ 395
- Nemotron 3 Super na Apple M1 Max
- Nemotron 3 Super na Apple M1 Pro
- Nemotron 3 Super na Apple M1 Ultra
- Nemotron 3 Super na Apple M2 Max
- Nemotron 3 Super na Apple M2 Pro
- Nemotron 3 Super na Apple M3 Max
- Nemotron 3 Super na Apple M3 Pro
- Nemotron 3 Super na Apple M4
- Nemotron 3 Super na Apple M4 Max
- Nemotron 3 Super na Apple M5
- Nemotron 3 Super na Apple M5 Max
- Nemotron 3 Super na Apple M5 Pro
- Nemotron 3 Super na NVIDIA A100 80GB (PCIe)
- Nemotron 3 Super na NVIDIA DGX Spark
- Nemotron 3 Super na NVIDIA GeForce RTX 5090
- Nemotron 3 Super na NVIDIA H100 80GB (PCIe)
- Nemotron 3 Super na NVIDIA L40S
- Nemotron 3 Super na NVIDIA RTX 6000 Ada Generation
- Nemotron 3 Super na NVIDIA RTX PRO 6000 Blackwell
Nemotron 3 Super — najczęściej zadawane pytania
How much VRAM does Nemotron 3 Super need?
Nemotron 3 Super needs about 73 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Nemotron 3 Super 120B-A12B (73 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run Nemotron 3 Super on an RTX 4090 (24 GB)?
Nemotron 3 Super's smallest variant needs about 73 GB, which exceeds a single RTX 4090 (24 GB). Use multiple GPUs, a higher-VRAM card, or Apple Silicon with large unified memory.
What quantization should I use for Nemotron 3 Super?
Q4_K_M is the best balance of quality and VRAM for Nemotron 3 Super in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run Nemotron 3 Super with Ollama?
Install Ollama, then run: ollama run nemotron-3-super:120b-a12b. This downloads Nemotron 3 Super and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.