GL

GLM-5.3-Flash

GLMGGUF by pfeifferjUpdated 2026-09-14
Download on Hugging Face
Parameters313B
Context1024K
Layers45
Architectureglm5-next
Smallest quant117.5 GB
Quick answer

GLM-5.3-Flash needs about 166.3 GB of memory at its main file (a 117.5 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 2 quants, from 117.5 GB to 137.1 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 193.3 GB.

GLM‑5.3‑Flash – 320B multimodal model in GGUF format

GLM‑5.3‑Flash is a 320B‑parameter multimodal model with 18B active parameters. It supports vision and long‑context up to 1,048,576 tokens. Two GGUF quantizations are available, each around 118–138 GB.

Good at

  • Vision tasks
  • Long‑context reasoning
  • Large‑scale multimodal inference

Limits

  • No text‑only quantization available
  • Requires 8 k context for memory estimates
  • Large file size limits local deployment

Best for

  • Vision‑enabled chatbots
  • Multimodal research prototypes
  • Long‑document analysis

Who it’s for. Developers and researchers needing a multimodal model with extensive context. The model is suitable for applications that can accommodate large file sizes and memory requirements.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—117.5 GB 166.3 GB
Too big Get file →
—137.1 GB 185.9 GB
Too big Get file →

For 8 k context, the 117.48 GB file needs about 166.3 GB RAM. The 137.07 GB file requires about 185.9 GB RAM. Choose the smaller file if memory is limited, but note the higher memory estimate for the larger file.

What GPU do you need for GLM-5.3-Flash?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Specifications

Architecture from the GGUF header

Architectureglm5-next
Parameters313B (approximate, mixture of experts)
Layers45
Trained context1,048,576 tokens
KV cache (fp16)5760 KB per token

Source

Quantized bypfeifferj
FamilyGLM
Licensemit
Downloads (HF)14,499
Last updated2026-09-14

Frequently asked questions about GLM-5.3-Flash

How much VRAM does GLM-5.3-Flash need?

About 166.3 GB at its main file with an 8K context window. That is the 117.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run GLM-5.3-Flash on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 166.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of GLM-5.3-Flash GGUF?

It depends on the quant: file is 117.5 GB, file is 137.1 GB.

What context length does GLM-5.3-Flash support?

The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 48.3 GB at 8K and 193.3 GB at 32K.

Where can I download GLM-5.3-Flash GGUF?

The files are on Hugging Face in the pfeifferj/GLM-5.3-Flash-GSQ-RCO-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does GLM-5.3-Flash use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run GLM‑5.3‑Flash on 8 GB VRAM?

No. The model requires over 166 GB RAM for 8 k context, far beyond 8 GB VRAM.

Does the model support vision?

Yes. The GGUF files include a BF16 vision projector for multimodal use.

What license applies to the GGUF files?

The files are released under the MIT license.

Other GGUFs of GLM-5.3-Flash

Similar models

Browse all →

More from pfeifferj

All by pfeifferj →

More GLM models

All GLM →

Other large models

All large models →