Question 1

How much VRAM does GLM-4.7-Flash need?

Accepted Answer

GLM-4.7-Flash (31.2B) requires approximately 69.78546509742736 GB of VRAM at full precision (FP16), 34.9 GB at 8-bit quantization (Q8), and 17.4 GB at 4-bit quantization (Q4). The actual VRAM usage may be slightly higher due to KV cache and runtime overhead.

Question 2

What GPUs can run GLM-4.7-Flash?

Accepted Answer

We have 10 compatible GPUs indexed for GLM-4.7-Flash, including A100 80GB PCIe, A100 80GB, H100 MEGA, H100 SXM, H100 NVL and 5 more. Compatibility is determined by whether the GPU's VRAM can fit the model's memory footprint, potentially using quantization to reduce memory usage.

Question 3

What is the cheapest GPU to run GLM-4.7-Flash?

Accepted Answer

The cheapest GPU to run GLM-4.7-Flash in the cloud is the A100 80GB PCIe at $1.39/hr. Other affordable options include A100 80GB ($1.39/hr) and H100 MEGA ($1.80/hr). Prices reflect the lowest on-demand rate across cloud providers.

Question 4

Can GLM-4.7-Flash run on consumer hardware?

Accepted Answer

Yes, GLM-4.7-Flash can run on consumer GPUs like the RTX Pro 6000 Blackwell (96 GB VRAM) at full precision (FP16).

Question 5

What quantization options are available for GLM-4.7-Flash?

Accepted Answer

GLM-4.7-Flash supports the following quantization levels: FP16 (full precision): ~69.78546509742736 GB, Q8 (8-bit): ~34.9 GB, Q6 (6-bit): ~26.2 GB, Q4 (4-bit): ~17.4 GB. Lower quantization reduces VRAM usage at the cost of some quality loss. Q8 offers a good balance between quality and memory savings, while Q4 provides maximum compression. As a Mixture of Experts (MoE) model with 64 experts (4 active), GLM-4.7-Flash has a larger total weight size but only activates a subset of parameters per token, which can affect effective memory usage during inference.

GPU	VRAM	Best Quant	From	Rent
A100 80GB PCIe	80 GB	Q8	$1.39/hr	Rent
A100 80GB	80 GB	Q8	$1.39/hr	Rent
H100 MEGA	80 GB	Q8	$1.80/hr	Rent
H100 SXM	80 GB	Q8	$1.80/hr	Rent
H100 NVL	94 GB	FP16	$1.80/hr	Rent
H100 PCIe	80 GB	Q8	$1.80/hr	Rent
RTX Pro 6000 Blackwell	96 GB	FP16	$1.89/hr	Rent
MI300X	192 GB	FP16	$1.99/hr	Rent
GH200	96 GB	FP16	$2.29/hr
H200 SXM	141 GB	FP16	$2.35/hr	Rent

GLM-4.7-Flash

VRAM Requirements

Compatible GPUs

GLM-4.7-Flash FAQ

VRAM requirements for GLM-4.7-Flash at different quantization levels
Quantization	VRAM Required
FP16	69.8 GB
Q8	34.9 GB
Q6	26.2 GB
Q4	17.4 GB