GLM-4.7-Flash

Z.ai · 31.2B (active 3B) · MoE · ctx 198K · MIT · HuggingFace

GLM-4.7-Flash (Z.ai) is a 31.2B parameters MoE (3B active per token) open-weight model tracked on AI Hashrate. At 8K context we estimate roughly 18.76 GB VRAM for Q4 (weights 17.16 GB + KV 0.6 GB + 1 GB overhead) and 64.0 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
A100 80GB SXM4	80	Q4	432.5 est.	Yes	0.029
RTX 5080 16GB	16	Q4	423.2 est.	No	0.424
RTX 5070 Ti 16GB	16	Q4	395.0 est.	No	0.527
RTX 5090 32GB	32	Q4	380.1 est.	Yes	0.19
RTX 4080 Super 16GB	16	Q4	324.5 est.	No	0.325
MI300X 192GB	192	FP16	309.2 est.	Yes	0.021
RTX 4070 Ti Super 16GB	16	Q4	296.3 est.	No	0.371
RX 9070 16GB	16	Q4	282.1 est.	No	0.514
RX 9070 XT 16GB	16	Q4	282.1 est.	No	0.471
H200 141GB SXM5	141	FP16	280.0 est.	Yes	0.008
RX 7800 XT 16GB	16	Q4	275.1 est.	No	0.551
RX 7900 GRE 16GB	16	Q4	253.9 est.	No	0.462
Arc A770 16GB	16	Q4	246.9 est.	No	0.707
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 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
RTX 5060 Ti 16GB	16	Q4	197.5 est.	No	0.46
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
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 5070 12GB	12	Q4	166.6 est.	No	0.303
RTX A6000 48GB	48	Q4	162.9 est.	Yes	0.036
T4 16GB	16	Q4	141.1 est.	No	0.071
RTX 4070 Ti 12GB	12	Q4	137.4 est.	No	0.172
RTX 3080 10GB	10	Q4	130.9 est.	No	0.187
RTX 4060 Ti 16GB	16	Q4	127.0 est.	No	0.255
RX 7600 XT 16GB	16	Q4	127.0 est.	No	0.386
RTX 4070 12GB	12	Q4	125.0 est.	No	0.209
RTX 4070 Super 12GB	12	Q4	125.0 est.	No	0.209
A100 80GB SXM4	80	FP16	118.9 est.	Yes	0.008
MacBook Pro M4 Max 128GB	128	Q4	115.8 est.	Yes	0.026
Arc B580 12GB	12	Q4	113.1 est.	No	0.454
RX 7700 XT 12GB	12	Q4	107.1 est.	No	0.239
RTX 6000 Ada 48GB	48	FP16	90.0 est.	No	0.013
RTX 3060 12GB	12	Q4	89.3 est.	No	0.271
MacBook Pro M3 Max 128GB	128	Q4	84.8 est.	Yes	0.02
L40S 48GB	48	FP16	81.0 est.	No	0.01
RTX 5090 32GB	32	FP16	74.7 est.	No	0.037
RTX A6000 48GB	48	FP16	72.0 est.	No	0.016
Arc B570 10GB	10	Q4	65.4 est.	No	0.299
Mac mini M4 Pro 64GB	64	Q4	57.9 est.	Yes	0.026
RTX 5060 8GB	8	Q4	49.4 est.	No	0.165
RTX 5060 Ti 8GB	8	Q4	49.4 est.	No	0.13
RTX 3070 8GB	8	Q4	49.4 est.	No	0.099
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
Mac mini M4 Pro 64GB	64	FP16	45.5 est.	No	0.021
MacBook Pro M4 Max 128GB	128	FP16	31.8 est.	Yes	0.007
RTX 4060 Ti 8GB	8	Q4	31.7 est.	No	0.079
RX 7600 8GB	8	Q4	31.7 est.	No	0.118
RTX 4060 8GB	8	Q4	30.0 est.	No	0.1
Mac mini M4 32GB	32	Q4	25.5 est.	Yes	0.026
RTX 4090 24GB	24	FP16	23.6 est.	No	0.015
RTX 4090D 24GB (China)	24	FP16	23.6 est.	No	0.015
RTX 3090 Ti 24GB	24	FP16	23.6 est.	No	0.012
MacBook Pro M3 Max 128GB	128	FP16	23.3 est.	Yes	0.006
RX 7900 XTX 24GB	24	FP16	22.5 est.	No	0.023
RTX 3090 24GB	24	FP16	21.9 est.	No	0.015
RX 7900 XT 20GB	20	FP16	13.0 est.	No	0.014
RTX 5080 16GB	16	FP16	10.0 est.	No	0.01
RTX 5070 Ti 16GB	16	FP16	9.3 est.	No	0.012
RTX 4080 Super 16GB	16	FP16	7.7 est.	No	0.008
RTX 4070 Ti Super 16GB	16	FP16	7.0 est.	No	0.009
RX 9070 16GB	16	FP16	6.7 est.	No	0.012
RX 9070 XT 16GB	16	FP16	6.7 est.	No	0.011
RX 7800 XT 16GB	16	FP16	6.5 est.	No	0.013
RX 7900 GRE 16GB	16	FP16	6.0 est.	No	0.011
Arc A770 16GB	16	FP16	5.8 est.	No	0.017
Mac mini M4 32GB	32	FP16	5.0 est.	No	0.005
RTX 5060 Ti 16GB	16	FP16	4.7 est.	No	0.011
T4 16GB	16	FP16	3.3 est.	No	0.002
RTX 4060 Ti 16GB	16	FP16	3.0 est.	No	0.006
RX 7600 XT 16GB	16	FP16	3.0 est.	No	0.009

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 · ≈423.2 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 · ≈395.0 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.

Same family: GLM

GLM-4.5 GLM-4.5-Air
GLM-5 GLM-5.1 · GLM-5.2

All GLM models →