Qwen3.6-35B-A3B

Alibaba · 36.0B (active 3.0B) · MoE · ctx 256K · apache-2.0 · HuggingFace

Qwen3.6-35B-A3B (Alibaba) is a 36.0B parameters MoE (3.0B active per token) open-weight model tracked on AI Hashrate. At 8K context we estimate roughly 21.4 GB VRAM for Q4 (weights 19.8 GB + KV 0.6 GB + 1 GB overhead) and 73.6 GB for FP16. About 21 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.

72 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
A100 80GB SXM4	80	Q4	432.5 est.	Yes	0.029
RX 7900 XT 20GB	20	Q4	423.5 est.	No	0.471
RTX 5090 32GB	32	Q4	380.1 est.	Yes	0.19
RTX 5080 16GB	16	Q4	325.2 est.	No	0.326
MI300X 192GB	192	FP16	309.2 est.	Yes	0.021
RTX 5070 Ti 16GB	16	Q4	303.6 est.	No	0.405
H200 141GB SXM5	141	FP16	280.0 est.	Yes	0.008
RTX 4080 Super 16GB	16	Q4	249.3 est.	No	0.25
RTX 4070 Ti Super 16GB	16	Q4	227.7 est.	No	0.285
RX 9070 16GB	16	Q4	216.8 est.	No	0.395
RX 9070 XT 16GB	16	Q4	216.8 est.	No	0.362
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
RX 7800 XT 16GB	16	Q4	211.4 est.	No	0.424
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
RX 7900 GRE 16GB	16	Q4	195.1 est.	No	0.355
Gaudi 3 128GB	128	FP16	191.1 est.	Yes	0.015
Arc A770 16GB	16	Q4	189.7 est.	No	0.544
L40S 48GB	48	Q4	183.3 est.	Yes	0.023
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
RTX A6000 48GB	48	Q4	162.9 est.	Yes	0.036
RTX 5060 Ti 16GB	16	Q4	151.8 est.	No	0.354
RTX 5070 12GB	12	Q4	128.1 est.	No	0.233
A100 80GB SXM4	80	FP16	118.9 est.	Yes	0.008
MacBook Pro M4 Max 128GB	128	Q4	115.8 est.	Yes	0.026
T4 16GB	16	Q4	108.4 est.	No	0.054
RTX 4070 Ti 12GB	12	Q4	105.6 est.	No	0.132
RTX 3080 10GB	10	Q4	100.6 est.	No	0.144
RTX 4060 Ti 16GB	16	Q4	97.6 est.	No	0.196
RX 7600 XT 16GB	16	Q4	97.6 est.	No	0.297
RTX 4070 12GB	12	Q4	96.0 est.	No	0.16
RTX 4070 Super 12GB	12	Q4	96.0 est.	No	0.16
Arc B580 12GB	12	Q4	86.9 est.	No	0.349
MacBook Pro M3 Max 128GB	128	Q4	84.8 est.	Yes	0.02
RX 7700 XT 12GB	12	Q4	82.3 est.	No	0.183
RTX 3060 12GB	12	Q4	68.6 est.	No	0.209
RTX 6000 Ada 48GB	48	FP16	68.1 est.	No	0.01
L40S 48GB	48	FP16	61.2 est.	No	0.008
Mac mini M4 Pro 64GB	64	Q4	57.9 est.	Yes	0.026
RTX 5090 32GB	32	FP16	56.5 est.	No	0.028
RTX A6000 48GB	48	FP16	54.4 est.	No	0.012
Arc B570 10GB	10	Q4	50.3 est.	No	0.23
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 5060 8GB	8	Q4	37.9 est.	No	0.127
RTX 5060 Ti 8GB	8	Q4	37.9 est.	No	0.1
RTX 3070 8GB	8	Q4	37.9 est.	No	0.076
Mac mini M4 Pro 64GB	64	FP16	34.4 est.	No	0.016
MacBook Pro M4 Max 128GB	128	FP16	31.8 est.	Yes	0.007
Mac mini M4 32GB	32	Q4	25.5 est.	Yes	0.026
RTX 4060 Ti 8GB	8	Q4	24.4 est.	No	0.061
RX 7600 8GB	8	Q4	24.4 est.	No	0.091
MacBook Pro M3 Max 128GB	128	FP16	23.3 est.	Yes	0.006
RTX 4060 8GB	8	Q4	23.0 est.	No	0.077
RTX 4090 24GB	24	FP16	17.9 est.	No	0.011
RTX 4090D 24GB (China)	24	FP16	17.9 est.	No	0.011
RTX 3090 Ti 24GB	24	FP16	17.9 est.	No	0.009
RX 7900 XTX 24GB	24	FP16	17.0 est.	No	0.017
RTX 3090 24GB	24	FP16	16.6 est.	No	0.011
RX 7900 XT 20GB	20	FP16	9.8 est.	No	0.011
Mac mini M4 32GB	32	FP16	3.8 est.	No	0.004

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 · ≈325.2 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 · ≈303.6 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 — does not fit · ≈423.5 tok/s Q4 est.

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