Nemotron 3.5 Lightning 30B A3B · 3 GPUs measured first-party · llama.cpp · Updated October 2026
What GPU Do You Need for Nemotron 3.5 Lightning 30B A3B?
Nemotron 3.5 Lightning 30B A3B on 3 GPUs, measured first-party: NVIDIA GeForce RTX 5090 leads at 376.1 tok/s, L40S trails at 199 tok/s, and it peaked at 25GB of VRAM.
Benchmarked weights: lmstudio-community/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF
NVIDIA GeForce RTX 5090
376.1 tok/s on Nemotron 3.5 Lightning 30B A3B, the ceiling. Measured on our bench. 32GB of VRAM, $1,999 at launch.
- 376.1 tok/s on Nemotron 3.5 Lightning 30B A3B
- 32GB, clears the Nemotron 3.5 Lightning 30B A3B floor
- 575W board rating
What GPU Do You Need for Nemotron 3.5 Lightning 30B A3B?, tok/s by GPU
Efficiency: tok/s per 100W drawn
Power is the average pulled during the run, sampled at 1Hz. The fastest card is often not the one here, and for anything left running this is the number that shows up on the bill.
Value: tok/s per $1,000 of MSRP
Launch price, not street price, so it ages. A speed leaderboard always crowns the most expensive card; this is the counterweight.
Nemotron 3.5 Lightning 30B A3B. Measured tokens per second by GPU
| GPU | tok/s | Prompt t/s | tok/W | Avg power |
|---|---|---|---|---|
| NVIDIA GeForce RTX 5090 | 376.1 | 10984.1 | 2.11 | 178.1 W |
| NVIDIA H100 80GB HBM3 | 323.6 | 6908.3 | 2 | 162.0 W |
| NVIDIA L40S | 198.9 | 6939.5 | 1.49 | 133.7 W |
What the numbers show. Across 3 GPUs measured on our own bench, RTX 5090 is fastest at 376 tok/s. The slowest, L40S, manages 199, so the spread is 1.9x from top to bottom.
About Nemotron 3.5 Lightning 30B A3B. Nemotron 3.5 Lightning 30B A3B: from nvidia, 32B parameters, on Hugging Face since August 2026. 2,569,837 downloads in the last 30 days and 3 community quantizations.
How it compares. H100 80GB HBM3: Nemotron 3.5 Lightning 30B A3B 323.6 tok/s, Nemotron-3-Nano-30B-A3B 316.5 (32B), Gemma 4 31B 76.59 (31B), GLM-4.7-Flash 186.1 (31B), Qwen3 30B A3B 283.8 (31B). Nemotron 3.5 Lightning 30B A3B beats all 4 here.
Cost on a rented GPU. 1M generated tokens of Nemotron 3.5 Lightning 30B A3B: $0.29 on a RTX 5090 ($0.39/hr, 44 min).
Nemotron 3.5 Lightning 30B A3B: cost per 1M generated tokens on rented GPUs
| GPU | Cheapest rate | Speed (tok/s) | Cost per 1M generated tokens |
|---|---|---|---|
| NVIDIA GeForce RTX 5090 | $0.39/hr | 376.1 | $0.29 |
| NVIDIA L40S | $0.79/hr | 198.9 | $1.10 |
| NVIDIA H100 80GB HBM3 | $2.14/hr | 323.6 | $1.83 |
Cheapest hourly rate we track on RunPod and Vast.ai, divided by the measured speed. Startup time and storage are extra.
Speed tiers for Nemotron 3.5 Lightning 30B A3B. 30+ tok/s: 3 (RTX 5090). 30 tok/s is roughly where replies outpace reading.
Reading your prompt. Before Nemotron 3.5 Lightning 30B A3B writes anything it reads the input: 10984.1 tok/s on the RTX 5090 (0.4s for a 4,000-token prompt), 6908.3 on the H100 80GB HBM3 (0.6s). Long documents and big code files feel this number more than the generation speed.
VRAM for Nemotron 3.5 Lightning 30B A3B. Measured peak 23.5GB, so 32GB is the smallest common card size; smallest card it ran on: RTX 5090 (32GB). With long context: Q4_K_M 30GB (tested), Q2_K 23GB, Q3_K_M 24GB, Q5_K_M 32GB, Q6_K 41GB.
Power on Nemotron 3.5 Lightning 30B A3B. Most efficient: RTX 5090, 178W, 0.13 kWh per 1M generated tokens. At $0.15/kWh: $0.020 per 1M generated tokens.
Our verdict
Fastest on Nemotron 3.5 Lightning 30B A3B: NVIDIA GeForce RTX 5090, 376.1 tok/s. Cheapest to rent per job: NVIDIA GeForce RTX 5090, $0.29 per 1M generated tokens.
FAQ
- What GPU do I need to run Nemotron 3.5 Lightning 30B A3B?
- About 25GB. Cheapest consumer card that ran it: NVIDIA GeForce RTX 5090 (32GB, 376.1 tok/s).
- How fast is Nemotron 3.5 Lightning 30B A3B on the NVIDIA GeForce RTX 5090?
- 376.1 tok/s at 178W, faster than every datacenter card we ran it on.
- How much does it cost to run Nemotron 3.5 Lightning 30B A3B in the cloud?
- $0.29 per 1M generated tokens on a NVIDIA GeForce RTX 5090 at $0.39/hr, cheapest of 3 rentable cards we measured.
- Can I run Nemotron 3.5 Lightning 30B A3B on a 12GB, 16GB or 24GB card?
- It used 23.5GB at the precision we tested. 12GB: no; 16GB: no; 24GB: only at Q3_K_M (24GB).
- Is the RTX 5090 or the H100 80GB HBM3 faster for Nemotron 3.5 Lightning 30B A3B?
- The RTX 5090: 376.1 vs 323.6 tok/s, 16% faster on our bench.
- Should I buy or rent a RTX 5090 for Nemotron 3.5 Lightning 30B A3B?
- Its $1,999 launch price buys 5,139 rented hours at $0.39/hr, enough for about 6,957x 1M generated tokens of Nemotron 3.5 Lightning 30B A3B. Buy only if you'll run more than that.
How we test
llama.cpp llama-bench at Q4_K_M, 512-token prompt and 128 generated tokens, three runs after a warmup, full GPU offload, with power and VRAM sampled throughout. Token generation is memory-bandwidth-bound, so the ranking tracks bandwidth closely, which makes it a fair guide to cards we haven't run yet.