GL

GLM 5.3 Flash Strata Q2Q3 Experts

GLMGGUF by lte-aiUpdated 2026-10-18
Download on Hugging Face
Parameters292B
Context1024K
Layers46
Architectureglm5-next
Smallest quant63.2 GB
Quick answer

GLM 5.3 Flash Strata Q2Q3 Experts needs about 145.8 GB of memory at its main file (a 95.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 8 quants, from 63.2 GB to 120.9 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 197.6 GB.

GLM 5.3 Flash Strata Q2Q3 Experts – large‑scale expert‑pruned model

GLM 5.3 Flash Strata Q2Q3 Experts is a 292 billion‑parameter GLM‑5.3‑Flash model pruned to 50 % experts. It uses a 1 048 576‑token context and is released under the MIT license. The model is available in several GGUF files with varying sizes and memory footprints.

Good at

  • Large‑context generation
  • Expert routing for efficient inference
  • High‑capacity language modeling

Limits

  • Requires 64 GB or 128 GB host RAM for full model
  • No vision support
  • No bio domain coverage

Best for

  • Research on expert‑pruned language models
  • Running on high‑memory GPUs
  • Benchmarking context‑length limits

Who it’s for. Researchers and developers needing a large‑context, expert‑pruned GLM model. The model is suited for environments with ample host memory and GPU VRAM, such as 64 GB or 128 GB workstations.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—63.2 GB 113.1 GB
Too big Get file →
—63.2 GB 113.1 GB
Too big Get file →
—95.5 GB 145.4 GB
Too big Get file →
—95.9 GB 145.8 GB
Too big Get file →
—105.5 GB 155.4 GB
Too big Get file →
—110.2 GB 160.1 GB
Too big Get file →
—119.9 GB 169.8 GB
Too big Get file →
—120.9 GB 170.8 GB
Too big Get file →

For 64 GB host memory, use the 95.54 GB file (est. 145.4 GB at 8 k context). For 128 GB host memory, use the 119.87 GB file (est. 169.8 GB at 8 k context). Memory estimates are approximate.

What GPU do you need for GLM 5.3 Flash Strata Q2Q3 Experts?

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.

Which GLM 5.3 Flash Strata Q2Q3 Experts quant should you choose?

If memory is tight, the smallest file is the first one in the table at 63.2 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
Parameters292B (approximate, mixture of experts)
Layers46
Trained context1,048,576 tokens
KV cache (fp16)5888 KB per token

Source

Quantized bylte-ai
FamilyGLM
Licensemit
Downloads (HF)27,124
Last updated2026-10-18

Frequently asked questions about GLM 5.3 Flash Strata Q2Q3 Experts

How much VRAM does GLM 5.3 Flash Strata Q2Q3 Experts need?

About 145.8 GB at its main file with an 8K context window. That is the 95.9 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 Strata Q2Q3 Experts on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 113.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of GLM 5.3 Flash Strata Q2Q3 Experts GGUF?

It depends on the quant: file is 63.2 GB, file is 63.2 GB, file is 95.5 GB, file is 95.9 GB, file is 105.5 GB, file is 110.2 GB and 2 more.

What context length does GLM 5.3 Flash Strata Q2Q3 Experts 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 49.4 GB at 8K and 197.6 GB at 32K.

Where can I download GLM 5.3 Flash Strata Q2Q3 Experts GGUF?

The files are on Hugging Face in the lte-ai/GLM-5.3-Flash-Strata-Q2Q3-Experts 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 Strata Q2Q3 Experts 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 Strata Q2Q3 Experts on 8 GB VRAM?

No. The model requires a GPU with at least 16 GB VRAM and a host with 64 GB or 128 GB RAM.

Does the model support vision tasks?

No. The model has no vision support.

What license does the model use?

The model is released under the MIT license.

Similar models

Browse all →

More GLM models

All GLM →

Other large models

All large models →