No — not on this device
Nemotron 3 Super 120B-A12B at Q4_K_M needs 70.4 GB against 18.0 GB usable, and the shortfall of 52.4 GB is more than 256 GB of system RAM can cover at a tolerable speed. A smaller sibling or a lower quantisation is the honest answer here.
The open-training-data 120B. Same hybrid layout as Lightning, so 128K context costs under a gigabyte.
Hardware requirements What hardware do I need to run Nemotron 3 Super 120B-A12B? →At Q4_K_M with an 8K context: a 141 GB GPU (H200 SXM) or a 128 GB Mac. Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.
The VRAM budget
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 122.7 GB | 123.4 GB | — | ~6.0 | −0.1% ppl | 105.4 GB over |
| Q6_K | 94.7 GB | 95.4 GB | — | ~7.8 | −0.4% ppl | 77.4 GB over |
| Q5_K_M | 81.8 GB | 82.5 GB | — | ~9.0 | −0.8% ppl | 64.5 GB over |
| Q4_K_M | 69.7 GB | 70.4 GB | — | ~11 | −1.9% ppl | 52.4 GB over |
| Q3_K_M | 56.4 GB | 57.1 GB | — | ~13 | −5.4% ppl | 39.1 GB over |
| Q2_K | 48.4 GB | 49.0 GB | — | ~15 | −15% ppl | 31.0 GB over |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it. Only 8 of its 88 blocks keep a per-token KV cache; the rest are linear-attention, Mamba or convolution blocks with a fixed-size state.
How to run it
$ pip install mlx-lm $ mlx_lm.generate --model mlx-community/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-4bit \ --max-tokens 512 --prompt "Hello"
Apple's own array framework. The fastest path on Apple Silicon. More on MLX.
Related models on an M4 Pro · 24 GB
| Model | Total | Tok/s est. | |
|---|---|---|---|
| Nemotron 3.5 Lightning 30B-A3B 31.6B MoE · same family |
Q4_K_M · 18.4 GB | — | Won't fit |
| Ling 3.0 Flash 124B-A5B 124B MoE · similar size |
Q4_K_M · 70.4 GB | — | Won't fit |
| Qwen3.5 122B-A10B 125B MoE · similar size |
Q4_K_M · 71.1 GB | — | Won't fit |
| Mistral Small 4 119B-A6B 119B MoE · similar size |
Q4_K_M · 67.7 GB | — | Won't fit |
| gpt-oss 120B 117B MoE · similar size |
MXFP4 · 61.4 GB | — | Won't fit |
| Llama 4 Scout 109B-A17B 109B MoE · similar size |
Q4_K_M · 63.4 GB | — | Won't fit |
| Qwen3.8-Flash-Next 180B-A6B 180B MoE · similar size |
Q4_K_M · 102.0 GB | — | Won't fit |
Nemotron 3 Super 120B-A12B on other hardware
| Device | Total | Tok/s est. | |
|---|---|---|---|
| GeForce RTX 4060 Laptop 7.0 GB usable · 256 GB/s |
70.4 GB | ~2.2 | CPU offload |
| GeForce RTX 3060 12 GB 10.6 GB usable · 360 GB/s |
70.4 GB | ~2.3 | CPU offload |
| M4 · 16 GB 10.7 GB usable · 120 GB/s |
70.4 GB | — | Won't fit |
| GeForce RTX 4090 22.4 GB usable · 1008 GB/s |
70.4 GB | ~2.9 | CPU offload |
| Radeon RX 7900 XTX 22.4 GB usable · 960 GB/s |
70.4 GB | ~2.9 | CPU offload |
| M3 Max · 64 GB 48.0 GB usable · 400 GB/s |
70.4 GB | — | Won't fit |
| DGX Spark · 128 GB 96.0 GB usable · 273 GB/s |
70.4 GB | 11 | Runs great |
| H200 SXM 139.4 GB usable · 4800 GB/s |
70.4 GB | 166 | Runs great |