Can I Run Nemotron 3 Super on NVIDIA DGX Spark?
Written by Jakub Rusinowski · Last updated August 15, 2026
Yes, comfortably — Nemotron 3 Super 120B-A12B at Q4_K_M needs 76.1 GB at 8K context (72.5 GB weights + 3.6 GB KV cache/overhead), leaving ~51.9 GB of the NVIDIA DGX Spark's 128 GB, at ~25 tok/s (est.), with room for up to 128K context.
See what else this hardware can run →
or compare on Vast.ai from $0.77/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
NVIDIA DGX Spark Specs
| VRAM | 128 GB |
| Memory Bandwidth | 273 GB/s |
Nemotron 3 Super 120B-A12B on the NVIDIA DGX Spark: VRAM by quantization
| Quant | VRAM needed | Fits 128 GB? | Max context | Whole PC |
|---|---|---|---|---|
| F16 | 243.7 GB | ✗ No | — | — |
| Q8_0 | 131.2 GB | ✗ No | — | — |
| Q6_K | 102.1 GB | ✓ Yes | 64K | — |
| Q5_K_M | 88.7 GB | ✓ Yes | 64K | — |
| Q4_K_M | 76.1 GB | ✓ Yes | 128K | — |
| Q3_K_M | 54.8 GB | ✓ Yes | 128K | — |
| Q2_K | 43.1 GB | ✓ Yes | 128K | — |
Assumes an 8K-token context with an f16 KV cache, on an estimated architecture — this model publishes no config we can read. A longer window needs more; a quantized KV cache needs less. “Max context” is the largest window that still fits in 128 GB. Figures are estimates from parameter count, quantization and memory bandwidth — the analyzer lets you tune KV-cache quant and context.
Context costs VRAM too. At Q4_K_M on the NVIDIA DGX Spark: 8K 76.1 GB · 32K 84.7 GB · 128K 118.9 GB. Every window this model publishes still fits 128 GB.
No compatible GPU? Nemotron 3 Super 120B-A12B on 64 GB of system RAM, CPU only: runs.
Nemotron 3 Super Sizes That Fit the NVIDIA DGX Spark (at 8K context)
| Nemotron 3 Super 120B-A12B | Q4_K_M · 76.1 GB at 8K context · ~25 tok/s (est.) |
At 2 hrs/day, buying (~$4,699) beats renting at $0.77/hr after about 8.5 years.
Affiliate links — we may earn a commission if you sign up, at no extra cost to you.
Cloud rates verified 2026-07 — estimates, and marketplace prices vary. Buying price is GPU MSRP only, not a full PC.
FAQ
Is the NVIDIA DGX Spark enough to run Nemotron 3 Super locally?
Yes, comfortably — Nemotron 3 Super 120B-A12B at Q4_K_M needs 76.1 GB at 8K context (72.5 GB weights + 3.6 GB KV cache/overhead), leaving ~51.9 GB of the NVIDIA DGX Spark's 128 GB, at ~25 tok/s (est.), with room for up to 128K context.
What's the best Nemotron 3 Super size for the NVIDIA DGX Spark?
Nemotron 3 Super 120B-A12B at Q4_K_M quantization needs 76.1 GB at 8K context (72.5 GB weights + 3.6 GB KV cache/overhead), estimated ~25 tokens/sec, up to 128K context.
Nemotron 3 Super on Other GPUs
- Nemotron 3 Super on Apple M4 Max
- Nemotron 3 Super on Apple M3 Max
- Nemotron 3 Super on Apple M1 Ultra
- Nemotron 3 Super on Apple M5 Max
- Nemotron 3 Super on AMD Ryzen AI Max+ 395
Popular Models on the NVIDIA DGX Spark
VRAM Tier
Troubleshooting
Buying Guide
← Can I Run It? | Nemotron 3 Super Model Page | NVIDIA DGX Spark GPU Page | Check Your Hardware