QW

Qwen3.8 27B Heretic Dynamic

QwenGGUF by ajgazinUpdated 2026-09-01
Download on Hugging Face
Parameters27.3B
Context256K
Layers65
Architectureqwen35
Smallest quant17.6 GB
Quick answer

Qwen3.8 27B Heretic Dynamic needs about 22.5 GB of memory at Q5_K_M (a 19.8 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 4 quants, from 17.6 GB (Q4_K_XL) to 31.5 GB (Q8_K_XL). Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.

Qwen3.8‑27B Heretic Dynamic – 27.3B parameter model with vision and dynamic quantization

Qwen3.8‑27B Heretic Dynamic is a 27.3B‑parameter model that supports vision. It is available in four dynamic GGUF quantizations. The model preserves 15 MTPs and is released under Apache‑2.0.

Good at

  • Vision tasks
  • Dynamic quantization
  • Large context (262144 tokens)

Limits

  • No text‑only license restrictions
  • Requires 8k context for memory estimates
  • No explicit language support listed

Best for

  • Image‑to‑text generation
  • Multimodal dialogue
  • Vision‑based content creation

Who it’s for. Developers needing a large vision‑capable LLM with dynamic quantization for efficient inference. Suitable for research and production environments that can host 27B‑parameter models.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_XL17.6 GB 20.2 GB
Too big Get file →
Q5_K_M19.8 GB 22.5 GB
Too big Get file →
Q6_K_XL25.3 GB 28.0 GB
Too big Get file →
Q8_K_XL31.5 GB 34.1 GB
Too big Get file →

For 8 GB VRAM use Q4_K_XL (file size 17.56 GB, estimated 20.2 GB memory at 8k context). For 12 GB VRAM use Q5_K_M (19.77 GB, 22.5 GB memory). For 16 GB VRAM use Q6_K_XL (25.3 GB, 28 GB memory). For 32 GB VRAM use Q8_K_XL (31.46 GB, 34.1 GB memory).

What GPU do you need for Qwen3.8 27B Heretic Dynamic?

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: the best quant that fits is Q5_K_M (19.8 GB file, about 22.5 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K_XL (25.3 GB file, about 28.0 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_K_XL (31.5 GB file, about 34.1 GB in use).

Which Qwen3.8 27B Heretic Dynamic quant should you choose?

Q5_K_M (19.8 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

If memory is tight, the smallest file is Q4_K_XL at 17.6 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters27.3B
Layers65
Trained context262,144 tokens
KV cache (fp16)260 KB per token

Source

Quantized byajgazin
FamilyQwen
Licenseapache-2.0
Downloads (HF)1,712
Last updated2026-09-01

Frequently asked questions about Qwen3.8 27B Heretic Dynamic

How much VRAM does Qwen3.8 27B Heretic Dynamic need?

About 22.5 GB at Q5_K_M with an 8K context window. That is the 19.8 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 Qwen3.8 27B Heretic Dynamic on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 20.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Qwen3.8 27B Heretic Dynamic GGUF?

It depends on the quant: Q4_K_XL is 17.6 GB, Q5_K_M is 19.8 GB, Q6_K_XL is 25.3 GB, Q8_K_XL is 31.5 GB.

What context length does Qwen3.8 27B Heretic Dynamic support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.2 GB at 8K and 8.7 GB at 32K.

Where can I download Qwen3.8 27B Heretic Dynamic GGUF?

The files are on Hugging Face in the ajgazin/Qwen3.8-27B-Heretic-Dynamic-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 Qwen3.8 27B Heretic Dynamic use?

Hugging Face lists the license as apache-2.0. 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 Qwen3.8‑27B Heretic Dynamic on 8 GB VRAM?

Use the Q4_K_XL quant, which has an estimated 20.2 GB memory usage at 8k context. It will exceed 8 GB VRAM, so it is not feasible without further optimization.

Does this model support vision?

Yes, the model has detected vision capabilities and includes a vision projector.

What license applies to this model?

The model is released under the Apache‑2.0 license.

Other GGUFs of Qwen3.8-27B-Ultra-Uncensored-Heretic-Native-MTP-Preserved

Similar models

Browse all →

More from ajgazin

All by ajgazin →

More Qwen models

All Qwen →

Other medium models

All medium models →