QW

Qwen3.8 Flash Next Heretic 2

QwenGGUF by DeritakUpdated 2026-08-30
Download on Hugging Face
Parameters128B
Context256K
Layers49
Architectureqwen4exp
Smallest quant91.5 GB
Quick answer

Qwen3.8 Flash Next Heretic 2 needs about 92.8 GB of memory at its main file (a 91.5 GB file) with an 8K context window, which is more than a 48 GB GPU holds. Longer context needs more: at 32K tokens the KV cache alone adds about 3.3 GB.

Qwen3.8 Flash Next Heretic 2 – 128B multimodal model with 262144 token context

Qwen3.8 Flash Next Heretic 2 is a 128‑billion‑parameter multimodal model from the Qwen family. It supports vision and was quantized with a custom ROCmFP4 mix for AMD GPUs. The GGUF file is 91.48 GB and requires about 92.8 GB of VRAM at 8 k context.

Good at

  • Vision‑enabled multimodal inference
  • Large 262144‑token context
  • Custom ROCmFP4 quantization for AMD GPUs

Limits

  • Requires 91.48 GB GGUF file
  • Needs 92.8 GB VRAM at 8 k context
  • No MTP quant support
  • License is other

Best for

  • High‑resolution image captioning
  • Long‑form multimodal dialogue
  • Research on large‑context vision models

Who it’s for. Developers and researchers needing a large multimodal model with extended context on AMD hardware. The model is suitable for experimentation and prototyping rather than production deployment due to its large memory footprint.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—91.5 GB 92.8 GB
Too big Get file →

The only available quant is the 91.48 GB GGUF file. It requires approximately 92.8 GB VRAM at 8 k context. For systems with less VRAM, the model cannot be loaded in full. No alternative quants are provided.

What GPU do you need for Qwen3.8 Flash Next Heretic 2?

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

Architectureqwen4exp
Parameters128B (approximate, mixture of experts)
Layers49
Trained context262,144 tokens
KV cache (fp16)98 KB per token

Source

Quantized byDeritak
FamilyQwen
Licenseother
Downloads (HF)2,823
Last updated2026-08-30

Frequently asked questions about Qwen3.8 Flash Next Heretic 2

How much VRAM does Qwen3.8 Flash Next Heretic 2 need?

About 92.8 GB at its main file with an 8K context window. That is the 91.5 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 Qwen3.8 Flash Next Heretic 2 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 92.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Qwen3.8 Flash Next Heretic 2 GGUF?

It depends on the quant: file is 91.5 GB.

What context length does Qwen3.8 Flash Next Heretic 2 support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.8 GB at 8K and 3.3 GB at 32K.

Where can I download Qwen3.8 Flash Next Heretic 2 GGUF?

The files are on Hugging Face in the Deritak/Qwen3.8-Flash-Next-heretic-2-ROCMFP4 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 Flash Next Heretic 2 use?

Hugging Face lists the license as other. 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 Flash Next Heretic 2 on 8 GB VRAM?

No. The model needs about 92.8 GB VRAM at 8 k context, far exceeding 8 GB.

Does the model support vision?

Yes, the model has detected vision capabilities.

What quantization is used?

It uses a custom ROCmFP4 mix; the GGUF file is listed as unknown quant.

Similar models

Browse all →

More Qwen models

All Qwen →

Other large models

All large models →