QW

Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB

QwenGGUF by webmp3Updated 2026-10-08
Download on Hugging Face
Parameters139B
Context256K
Layers48
Architectureqwen4exp
Smallest quant57.0 GB
Quick answer

Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB needs about 58.3 GB of memory at IQ2_XS (a 57.0 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 3.2 GB.

Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB – a 139B Qwen4exp model with 352 routed experts

Sakura is a 139‑billion‑parameter Qwen4exp model fine‑tuned and quantized to 2.5‑bit IQ2_XS. It keeps 352 of 512 experts per layer and is optimized for shorter reasoning traces. The GGUF file is 56.99 GB and fits a 32 GB GPU with lazy loading.

Good at

  • Short reasoning and coding tasks
  • Efficient expert routing
  • Low‑token usage

Limits

  • Requires 32 GB GPU or more
  • No vision support
  • License is other, not official Qwen release

Best for

  • Chatbots needing fast inference
  • Agents with limited GPU memory
  • Research on expert pruning

Who it’s for. Developers and researchers who need a Qwen4exp model that runs on a 32 GB GPU and benefits from expert pruning. It is suitable for applications that can tolerate a 2.5‑bit quantization and require efficient reasoning.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_XS57.0 GB (2 parts) 58.3 GB
Too big Get file →

The only available quant is IQ2_XS, 56.99 GB in two shards. It needs about 58.3 GB of memory at 8 k context. It fits a 32 GB GPU with lazy loading of the token table. For smaller GPUs, a higher‑bit quant would be required, but none is listed.

What GPU do you need for Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB?

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 Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB quant should you choose?

IQ2_XS (57.0 GB) is the middle option in this repository. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen4exp
Parameters139B (approximate, mixture of experts)
Layers48
Trained context262,144 tokens
KV cache (fp16)96 KB per token

Source

Quantized bywebmp3
FamilyQwen
Licenseother
Downloads (HF)2,833
Last updated2026-10-08

Frequently asked questions about Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB

How much VRAM does Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB need?

About 58.3 GB at IQ2_XS with an 8K context window. That is the 57.0 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 Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 58.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB GGUF?

It depends on the quant: IQ2_XS is 57.0 GB.

What context length does Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB 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 3.2 GB at 32K.

Where can I download Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB GGUF?

The files are on Hugging Face in the webmp3/Sakura-Qwen3.8-Flash-Next-Swift-2.5bit-352E-32GB-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 Sakura Qwen3.8 Flash Next Swift 2.5bit 352E 32GB 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 this model on a 32 GB GPU?

Yes, the IQ2_XS quant fits a 32 GB GPU with lazy loading of the token table. It uses about 29.9 GB of dedicated memory in a typical server setup.

Does this model support image input?

No, image input is not supported. The model is text‑only.

What is the context length of this model?

The trained context length is 262,144 tokens, but the provided quant is evaluated at 8 k context for memory estimates.

Other GGUFs of Swift-1.5-Qwen3.8-Flash-Next-GSQ-RCO-GGUF

Similar models

Browse all →

More from webmp3

All by webmp3 →

More Qwen models

All Qwen →

Other large models

All large models →