NVIDIA-Nemotron-3.5-Lightning-30B-A3B

Nemotron family · 32.0B params

9recorded runs

446.7best tok/s

6verified

Evidence: 6 sourced runs · 3 awaiting review · 1 independent source · first recorded Aug 11, 2026 · last recorded Sep 4, 2026. What these labels mean.

What the NVIDIA-Nemotron-3.5-Lightning-30B-A3B numbers mean

The fastest verified result on our board is 446.7 output tok/s on an RTX 5090 at NVFP4 with a 65536-token context. That is about 2.2 ms per generated token — 500 tokens takes roughly 1.1 seconds. Across the 9 records we hold for this model the range is 35.0–446.7 tok/s on 7 GPUs, because quantisation, context length and the engine build all move the number.

How much VRAM does NVIDIA-Nemotron-3.5-Lightning-30B-A3B need?

NVIDIA-Nemotron-3.5-Lightning-30B-A3B has 32.0B parameters. These estimates use the same formula as the VRAM calculator: weights, plus about 15% for a moderate-context KV cache and runtime overhead, plus a small engine floor.

Quantisation	Estimated VRAM	Smallest card that fits
Q4_K_M (4-bit)	21.5 GB	32 GB
Q8_0 (8-bit)	39.9 GB	48 GB
FP16 (16-bit)	74.4 GB	96 GB

Long contexts and large batches raise the real number — read the smallest card as "possible with a reduced context". Measured throughput is on the benchmark board.

Fastest GPUs for this model (best tok/s)

Models with a similar footprint

Model	Params	Q4_K_M weights
Nemotron-Cascade-2-30B-A3B	32B	18.0 GB
NVIDIA-Nemotron-3-Nano-30B-A3B	32B	18.0 GB
Olmo-3-32B-Think	32B	18.0 GB
gemma-4-31B-it	33B	18.6 GB
GLM-Z1-32B-0414	33B	18.6 GB
Laguna-XS-2.1	33B	18.6 GB

Ranked by distance from NVIDIA-Nemotron-3.5-Lightning-30B-A3B's parameter count. Full benchmark board.

All benchmarks Methodology

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
RTX 5090 · 32 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	NVFP4	vllm 0.27.1	65536	446.7 ✓	—	—	src
RTX 3090 · 24 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	NVFP4	llama.cpp 030ebb558 + PR 26623 @ 000d8ecd4 + l…	8192	353.7 ✓	1951.8	21.4	src
RTX 3090 x2 · 24 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	NVFP4	vllm 0.27.1	131072	197.0 ✓	6719.0	19.5	src
Intel Arc Pro B70 · 32 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	GPTQ-INT4-G64-sym-local+DFlash-B	vllm v0.26.1rc1.dev668+g3ee2df303 (XPU)	16384	186.6 ✓	7160.0	—	src
CMP 170HX · 64 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	Q8_0	llama.cpp 0.3.0-dev+b10705	67592	165.1 ✓	1524.7	—	src
Intel Arc Pro B70 · 32 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	Unsloth-Dynamic-Q4_K_M	llama.cpp 9fee29e9435f; oneAPI 2026.1.1	8192	72.2 ✓	—	—	src
CMP 170HX · 64 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	W4A16	vllm 0.28.0	67592	217.9 …	8392.9	—	src
RX 7900 XT x2 · 20 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	NVFP4	llama.cpp —	2048	116.0 …	1191.7	—	src
RTX 4070 · 12 GB	NVIDIA-Nemotron-3.5-Lightning-30B-A3B	IQ4_XS	llama.cpp —	262144	35.0 …	398.2	11.3	src