GL

GLM 5.3 Flash UNCENSORED mlx oQ8e DFlash2

GLMGGUF by Solstice-AIUpdated 2026-10-15
Download on Hugging Face
Parameters1.17B
Context1024K
Layers5
Architecturedflash
Smallest quant2.4 GB
Quick answer

GLM 5.3 Flash UNCENSORED mlx oQ8e DFlash2 needs about 3.0 GB of memory at BF16 (a 2.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 0.7 GB.

GLM 5.3 Flash UNCENSORED with 1M context and native multimodal vision

GLM 5.3 Flash UNCENSORED is a 1.2B parameter model with a 1,048,576 token context window. It supports vision and video input and includes a bundled DFlash2 speculative drafter. The model is released under the MIT license.

Good at

  • Large context inference
  • Multimodal vision and video
  • Speculative decoding acceleration

Limits

  • Requires ~3 GB VRAM at 8 k context
  • Large memory footprint limits use on low‑VRAM GPUs
  • No explicit language support list

Best for

  • Long‑form document analysis
  • Multimodal content generation
  • Research prototyping with large context

Who it’s for. Developers and researchers needing a 1 M token context window and multimodal capabilities in a lightweight 2.35 GB GGUF file. Suitable for Apple Silicon and other systems that support BF16 quantization.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
BF162.4 GB 3.0 GB
Fits Get file →

The only available quantization is BF16, 2.35 GB in size. It requires approximately 3 GB VRAM at 8 k context. For lower VRAM budgets, no other quant is provided in the file list.

What GPU do you need for GLM 5.3 Flash UNCENSORED mlx oQ8e DFlash2?

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 BF16 (2.4 GB file, about 3.0 GB in use).

Which GLM 5.3 Flash UNCENSORED mlx oQ8e DFlash2 quant should you choose?

BF16 (2.4 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturedflash
Parameters1.17B
Layers5
Trained context1,048,576 tokens
KV cache (fp16)20 KB per token

Source

Quantized bySolstice-AI
FamilyGLM
Licensemit
Downloads (HF)10,614
Last updated2026-10-15

Frequently asked questions about GLM 5.3 Flash UNCENSORED mlx oQ8e DFlash2

How much VRAM does GLM 5.3 Flash UNCENSORED mlx oQ8e DFlash2 need?

About 3.0 GB at BF16 with an 8K context window. That is the 2.4 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 UNCENSORED mlx oQ8e DFlash2 on an 8 GB GPU?

Yes, with BF16: the 2.4 GB file needs about 3.0 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 UNCENSORED mlx oQ8e DFlash2 GGUF?

It depends on the quant: BF16 is 2.4 GB.

What context length does GLM 5.3 Flash UNCENSORED mlx oQ8e DFlash2 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 0.2 GB at 8K and 0.7 GB at 32K.

Where can I download GLM 5.3 Flash UNCENSORED mlx oQ8e DFlash2 GGUF?

The files are on Hugging Face in the Solstice-AI/GLM-5.3-Flash-UNCENSORED-mlx-oQ8e-DFlash2 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 UNCENSORED mlx oQ8e DFlash2 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 this model on 8 GB VRAM?

Yes, BF16 requires about 3 GB VRAM at 8 k context, so 8 GB is sufficient.

Does the model support vision input?

Yes, it includes a 347‑tensor vision tower and multimodal aligner.

What license does the model use?

The model is released under the MIT license.

Other GGUFs of GLM-5.3-Flash-UNCENSORED-FP8

Similar models

Browse all →

More from Solstice-AI

All by Solstice-AI →

More GLM models

All GLM →

Other tiny models

All tiny models →