GL

GLM-5.3-Flash Vision Encoder

GLMGGUF by antirezUpdated 2026-08-28
Download on Hugging Face
Parameters321B
Context—
Layers24
Architectureglm5-next-vision
Smallest quant1.1 GB
Quick answer

GLM-5.3-Flash Vision Encoder needs about 97.8 GB of memory at its main file (a 96.5 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 4 quants, from 1.1 GB to 327.2 GB.

GLM-5.3-Flash Vision Encoder – 321 Billion‑parameter vision model

The GLM‑5.3‑Flash Vision Encoder is a 321 Billion‑parameter model built on the glm5‑next‑vision architecture. It is available in multiple GGUF quantizations and supports vision tasks. The model has no explicit license information or language support listed.

Good at

  • Vision encoding
  • Large‑parameter inference
  • High‑capacity context

Limits

  • No license details provided
  • No language support listed
  • Large memory requirement for full precision

Best for

  • Vision‑centric LLM pipelines
  • Image feature extraction
  • High‑resolution image understanding

Who it’s for. Developers building vision‑enabled language models who need a large‑parameter encoder. Suitable for research and production environments with sufficient GPU memory.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—1.1 GB 2.4 GB
Fits Get file →
—96.5 GB 97.8 GB
Too big Get file →
Q4_K190.9 GB 192.2 GB
Too big Get file →
—327.2 GB 328.5 GB
Too big Get file →

For 8 GB VRAM, use the 1.13 GB file (est. 2.4 GB memory). For 16 GB VRAM, the 96.51 GB file (est. 97.8 GB memory) is suitable. For larger budgets, the 190.88 GB Q4_K file (est. 192.2 GB memory) or the 327.21 GB file (est. 328.5 GB memory) can be used.

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

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: the best quant that fits is the main file (1.1 GB file, about 2.4 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.4 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.4 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.4 GB in use).
  • 16 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.4 GB in use).
  • 24 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.4 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.4 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.4 GB in use).

Which GLM-5.3-Flash Vision Encoder quant should you choose?

If memory is tight, the smallest file is the first one in the table at 1.1 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureglm5-next-vision
Parameters321B
Layers24
KV cache (fp16)96 KB per token

Source

Quantized byantirez
FamilyGLM
Downloads (HF)69,798
Last updated2026-08-28

Frequently asked questions about GLM-5.3-Flash Vision Encoder

How much VRAM does GLM-5.3-Flash Vision Encoder need?

About 97.8 GB at its main file with an 8K context window. That is the 96.5 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 Vision Encoder on an 8 GB GPU?

Yes, with the smaller files: the 1.1 GB file needs about 2.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

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

It depends on the quant: file is 1.1 GB, file is 96.5 GB, Q4_K is 190.9 GB, file is 327.2 GB.

Where can I download GLM-5.3-Flash Vision Encoder GGUF?

The files are on Hugging Face in the antirez/glm-5.3-flash-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run the GLM‑5.3‑Flash Vision Encoder on 8 GB VRAM?

Yes, the 1.13 GB GGUF file requires approximately 2.4 GB of VRAM at 8k context, which fits within an 8 GB GPU.

What quantization is available for this model?

The model offers a Q4_K quantization with a 190.88 GB file and an unknown quantization with a 1.13 GB file.

Does the model support multiple languages?

No language support is listed for this model.

Similar models

Browse all →

More from antirez

All by antirez →

More GLM models

All GLM →

Other large models

All large models →