Arrakis / Models / GLM-5.3-Flash

GLM-5.3-Flash GPU requirements

GLM-5.3-Flash has 321B parameters. To serve it, plan for about 771 GB of GPU memory in 16-bit, 386 GB in 8-bit or 193 GB in 4-bit, including about 20% for the KV cache and activations.

It is a mixture of experts: 288 experts, 8 of them active for each token. All of them stay in GPU memory, so plan memory for all 321B parameters.

Deploy this model Check another model

What it takes to serve it

GPU memory to run it for inference, with the smallest GPU setup that holds it and the lowest price on Arrakis right now.

Precision	Memory and the smallest setup
16-bit	771 GB: 4 × B300, from $18.40 per hour for all 4
8-bit	386 GB: 2 × B300, from $9.20 per hour for all 2
4-bit	193 GB: one B300, from $4.60 per GPU-hour

What it takes to fine-tune it

QLoRA trains adapters on a 4-bit frozen model, LoRA on a 16-bit one, and full fine-tuning updates every weight. Figures are before activations, which grow with batch size and sequence length.

Method	Memory and the smallest setup
QLoRA	225 GB: one B300, from $4.60 per GPU-hour
LoRA	771 GB: 4 × B300, from $18.40 per hour for all 4
Full fine-tuning	5,141 GB: more than one 8-GPU machine

About the model

From its card and files on Hugging Face. Memory figures here follow the GPU memory guide.

Parameters: 321B
Experts: 288
Experts for each token: 8
Context window (tokens): 1,048,576
Published in: FP8
License: mit
Released: 2026-08-25
Weights: zai-org/GLM-5.3-Flash

Deploy it on Arrakis

Open Models in the console, choose vLLM, SGLang or llama.cpp, and rent a GPU it fits on. You get an OpenAI-compatible endpoint with its own API key, and root SSH to the machine. Or run it yourself on any machine with enough GPU memory:

vLLM

vllm serve zai-org/GLM-5.3-Flash --tensor-parallel-size 4

Deploy this model How deploying works

Questions

Live prices for every GPU are on the GPU prices page.

How much GPU memory does GLM-5.3-Flash need?

To serve it, about 771 GB in 16-bit, 386 GB in 8-bit, 193 GB in 4-bit, including about 20% for the KV cache and activations. To fine-tune it, about 225 GB with QLoRA, 771 GB with LoRA and 5,141 GB for full fine-tuning with Adam, before activations.

Can I run GLM-5.3-Flash on one GPU?

Yes, quantized. In 4-bit it needs about 193 GB, which fits on one B300. In 16-bit it needs 4 B300 GPUs.

What does it cost to run GLM-5.3-Flash?

In 16-bit it needs 4 B300 GPUs; in 4-bit, one B300. Each setup on this page shows the lowest price per GPU-hour on Arrakis right now, and you pay by the hour from a prepaid balance.

How do I deploy GLM-5.3-Flash on Arrakis?

Open Models in the Arrakis console, paste zai-org/GLM-5.3-Flash, choose vLLM, SGLang or llama.cpp, and rent a GPU it fits on. You get an OpenAI-compatible endpoint with its own API key, and root SSH to the machine.

More models

Every model, and a check for any other.