Can I Run Nemotron 70B on NVIDIA L40S?
Written by Jakub Rusinowski · Last updated July 21, 2026
Yes, but it's tight — Nemotron 70B Instruct at Q4_K_M needs 46.1 GB at 8K context (42.6 GB weights + 3.5 GB KV cache/overhead), leaving ~1.9 GB of the NVIDIA L40S's 48 GB, at ~18 tok/s (est.), with room for up to 8K context.
See what else this hardware can run →
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 L40S Specs
| VRAM | 48 GB |
| Memory Bandwidth | 864 GB/s |
Nemotron 70B Instruct on the NVIDIA L40S: VRAM by quantization
| Quant | VRAM needed | Fits 48 GB? | Max context | Whole PC |
|---|---|---|---|---|
| F16 | 144.7 GB | ✗ No | — | — |
| Q8_0 | 78.5 GB | ✗ No | — | — |
| Q6_K | 61.4 GB | ✗ No | — | — |
| Q5_K_M | 53.5 GB | ✗ No | — | — |
| Q4_K_M | 46.1 GB | ✓ Yes | 8K | — |
| Q3_K_M | 33.6 GB | ✓ Yes | 32K | — |
| Q2_K | 26.7 GB | ✓ Yes | 64K | — |
Assumes an 8K-token context with an f16 KV cache. A longer window needs more; a quantized KV cache needs less. “Max context” is the largest window that still fits in 48 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 L40S: 8K 46.1 GB · 32K 54.2 GB ✗ · 125K 85.4 GB ✗. Past 32K it no longer fits 48 GB — a q8_0 KV cache buys roughly half of that back, which the calculator will price for you.
No compatible GPU? Nemotron 70B Instruct on 256 GB of system RAM, CPU only: runs, but too slowly to be worth it.
Nemotron 70B Sizes That Fit the NVIDIA L40S (at 8K context)
| Nemotron 70B Instruct | Q4_K_M · 46.1 GB at 8K context · ~18 tok/s (est.) |
At 2 hrs/day, buying (~$7,499) beats renting at $0.77/hr after about 13.5 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 L40S enough to run Nemotron 70B locally?
Yes, but it's tight — Nemotron 70B Instruct at Q4_K_M needs 46.1 GB at 8K context (42.6 GB weights + 3.5 GB KV cache/overhead), leaving ~1.9 GB of the NVIDIA L40S's 48 GB, at ~18 tok/s (est.), with room for up to 8K context.
What's the best Nemotron 70B size for the NVIDIA L40S?
Nemotron 70B Instruct at Q4_K_M quantization needs 46.1 GB at 8K context (42.6 GB weights + 3.5 GB KV cache/overhead), estimated ~18 tokens/sec, up to 8K context.
Nemotron 70B on Other GPUs
- Nemotron 70B on NVIDIA RTX 6000 Ada Generation
- Nemotron 70B on Apple M3 Pro
- Nemotron 70B on NVIDIA GeForce RTX 5090
- Nemotron 70B on Apple M4
- Nemotron 70B on Apple M2 Pro
Popular Models on the NVIDIA L40S
VRAM Tier
Troubleshooting
- CUDA out of memory — why it happens and how to fix it
- Which GGUF quant should I download? (Q4 vs Q5 vs Q8)
Buying Guide
← Can I Run It? | Nemotron 70B Model Page | NVIDIA L40S GPU Page | VRAM calculator | Check Your Hardware