Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16

NVIDIA · 30.0B (active 3.0B) · MoE · ctx 256K · nvidia-open-model-license · HuggingFace

Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 (NVIDIA) is a 30.0B parameters MoE (3.0B active per token) open-weight model tracked on AI Hashrate. At 8K context we estimate roughly 18.1 GB VRAM for Q4 (weights 16.5 GB + KV 0.6 GB + 1 GB overhead) and 61.6 GB for FP16. About 22 GPUs/accelerators in our catalog fully fit this model at Q4 under that context assumption. Fastest Q4 configs: B200 192GB SXM ≈ 1697.0 tok/s (estimated); MI300X 192GB ≈ 1124.2 tok/s (estimated); H200 141GB SXM5 ≈ 1018.2 tok/s (estimated); H100 80GB SXM5 ≈ 710.6 tok/s (estimated); Gaudi 3 128GB ≈ 694.9 tok/s (estimated). Use the table below for VRAM fit and tok/s per dollar (list MSRP). Estimates follow memory-bandwidth math; measured rows override when present. See methodology for details. Methodology.

85 hardware configs — sorted by tok/s. Default fit context 8K.

Hardware	VRAM	Quant	tok/s	Fits?	tok/s/$
B200 192GB SXM	192	Q4	1697.0 est.	Yes	0.042
MI300X 192GB	192	Q4	1124.2 est.	Yes	0.075
H200 141GB SXM5	141	Q4	1018.2 est.	Yes	0.029
H100 80GB SXM5	80	Q4	710.6 est.	Yes	0.024
Gaudi 3 128GB	128	Q4	694.9 est.	Yes	0.056
B200 192GB SXM	192	FP16	466.7 est.	Yes	0.012
RTX 5080 16GB	16	Q4	454.6 est.	No	0.455
A100 80GB SXM4	80	Q4	432.5 est.	Yes	0.029
RTX 5070 Ti 16GB	16	Q4	424.3 est.	No	0.566
RTX 5090 32GB	32	Q4	380.1 est.	Yes	0.19
RTX 4080 Super 16GB	16	Q4	348.6 est.	No	0.349
RTX 4070 Ti Super 16GB	16	Q4	318.2 est.	No	0.398
MI300X 192GB	192	FP16	309.2 est.	Yes	0.021
RX 9070 16GB	16	Q4	303.1 est.	No	0.552
RX 9070 XT 16GB	16	Q4	303.1 est.	No	0.506
RX 7800 XT 16GB	16	Q4	295.5 est.	No	0.592
H200 141GB SXM5	141	FP16	280.0 est.	Yes	0.008
RX 7900 GRE 16GB	16	Q4	272.8 est.	No	0.497
Arc A770 16GB	16	Q4	265.2 est.	No	0.76
RTX 4090 24GB	24	Q4	213.8 est.	Yes	0.134
RTX 4090D 24GB (China)	24	Q4	213.8 est.	Yes	0.134
RTX 3090 Ti 24GB	24	Q4	213.8 est.	Yes	0.107
RTX 5060 Ti 16GB	16	Q4	212.2 est.	No	0.495
RX 7900 XTX 24GB	24	Q4	203.6 est.	Yes	0.204
RTX 6000 Ada 48GB	48	Q4	203.6 est.	Yes	0.03
RTX 3090 24GB	24	Q4	198.5 est.	Yes	0.132
H100 80GB SXM5	80	FP16	195.4 est.	Yes	0.007
Gaudi 3 128GB	128	FP16	191.1 est.	Yes	0.015
L40S 48GB	48	Q4	183.3 est.	Yes	0.023
RTX 5070 12GB	12	Q4	179.0 est.	No	0.326
Mac Studio M3 Ultra 96GB	96	Q4	173.7 est.	Yes	0.043
Mac Studio M2 Ultra 192GB	192	Q4	169.7 est.	Yes	0.024
RX 7900 XT 20GB	20	Q4	169.7 est.	Yes	0.189
RTX A6000 48GB	48	Q4	162.9 est.	Yes	0.036
T4 16GB	16	Q4	151.5 est.	No	0.076
RTX 4070 Ti 12GB	12	Q4	147.6 est.	No	0.185
RTX 3080 10GB	10	Q4	140.6 est.	No	0.201
RTX 4060 Ti 16GB	16	Q4	136.4 est.	No	0.273
RX 7600 XT 16GB	16	Q4	136.4 est.	No	0.415
RTX 4070 12GB	12	Q4	134.3 est.	No	0.224
RTX 4070 Super 12GB	12	Q4	134.3 est.	No	0.224
Arc B580 12GB	12	Q4	121.5 est.	No	0.488
A100 80GB SXM4	80	FP16	118.9 est.	Yes	0.008
MacBook Pro M4 Max 128GB	128	Q4	115.8 est.	Yes	0.026
RX 7700 XT 12GB	12	Q4	115.1 est.	No	0.256
RTX 6000 Ada 48GB	48	FP16	97.1 est.	No	0.014
RTX 3060 12GB	12	Q4	95.9 est.	No	0.291
L40S 48GB	48	FP16	87.4 est.	No	0.011
MacBook Pro M3 Max 128GB	128	Q4	84.8 est.	Yes	0.02
RTX 5090 32GB	32	FP16	80.6 est.	No	0.04
RTX A6000 48GB	48	FP16	77.7 est.	No	0.017
Arc B570 10GB	10	Q4	70.3 est.	No	0.321
Mac mini M4 Pro 64GB	64	Q4	57.9 est.	Yes	0.026
RTX 5060 8GB	8	Q4	53.0 est.	No	0.177
RTX 5060 Ti 8GB	8	Q4	53.0 est.	No	0.14
RTX 3070 8GB	8	Q4	53.0 est.	No	0.106
Mac mini M4 Pro 64GB	64	FP16	49.1 est.	No	0.022
Mac Studio M3 Ultra 96GB	96	FP16	47.8 est.	Yes	0.012
Mac Studio M2 Ultra 192GB	192	FP16	46.7 est.	Yes	0.007
RTX 4060 Ti 8GB	8	Q4	34.1 est.	No	0.085
RX 7600 8GB	8	Q4	34.1 est.	No	0.127
RTX 4060 8GB	8	Q4	32.2 est.	No	0.108
MacBook Pro M4 Max 128GB	128	FP16	31.8 est.	Yes	0.007
RTX 4090 24GB	24	FP16	25.5 est.	No	0.016
RTX 4090D 24GB (China)	24	FP16	25.5 est.	No	0.016
RTX 3090 Ti 24GB	24	FP16	25.5 est.	No	0.013
Mac mini M4 32GB	32	Q4	25.5 est.	Yes	0.026
RX 7900 XTX 24GB	24	FP16	24.3 est.	No	0.024
RTX 3090 24GB	24	FP16	23.7 est.	No	0.016
MacBook Pro M3 Max 128GB	128	FP16	23.3 est.	Yes	0.006
RX 7900 XT 20GB	20	FP16	14.1 est.	No	0.016
RTX 5080 16GB	16	FP16	10.8 est.	No	0.011
RTX 5070 Ti 16GB	16	FP16	10.1 est.	No	0.013
RTX 4080 Super 16GB	16	FP16	8.3 est.	No	0.008
RTX 4070 Ti Super 16GB	16	FP16	7.6 est.	No	0.01
RX 9070 16GB	16	FP16	7.2 est.	No	0.013
RX 9070 XT 16GB	16	FP16	7.2 est.	No	0.012
RX 7800 XT 16GB	16	FP16	7.0 est.	No	0.014
RX 7900 GRE 16GB	16	FP16	6.5 est.	No	0.012
Arc A770 16GB	16	FP16	6.3 est.	No	0.018
Mac mini M4 32GB	32	FP16	5.4 est.	No	0.005
RTX 5060 Ti 16GB	16	FP16	5.0 est.	No	0.012
T4 16GB	16	FP16	3.6 est.	No	0.002
RTX 4060 Ti 16GB	16	FP16	3.2 est.	No	0.006
RX 7600 XT 16GB	16	FP16	3.2 est.	No	0.01

Fit checks

• RTX 5090 32GB — fits · ≈380.1 tok/s Q4 est.
• RTX 3090 Ti 24GB — fits · ≈213.8 tok/s Q4 est.
• RTX 4090 24GB — fits · ≈213.8 tok/s Q4 est.
• RTX 5080 16GB — does not fit · ≈454.6 tok/s Q4 est.
• RTX 6000 Ada 48GB — fits · ≈203.6 tok/s Q4 est.
• RX 7900 XTX 24GB — fits · ≈203.6 tok/s Q4 est.
• RTX 3090 24GB — fits · ≈198.5 tok/s Q4 est.
• RTX 5070 Ti 16GB — does not fit · ≈424.3 tok/s Q4 est.
• L40S 48GB — fits · ≈183.3 tok/s Q4 est.
• Mac Studio M3 Ultra 96GB — fits · ≈173.7 tok/s Q4 est.
• Mac Studio M2 Ultra 192GB — fits · ≈169.7 tok/s Q4 est.
• RX 7900 XT 20GB — fits · ≈169.7 tok/s Q4 est.

12 popular retail GPUs; every row in the table above also links its Fits? verdict to the full fit check.