QW

Sakura Qwen3.8 Flash Next ISTA 3bit 352E

QwenGGUF by webmp3Updated 2026-10-08
Download on Hugging Face
Parameters139B
Context256K
Layers48
Architectureqwen4exp
Smallest quant62.4 GB
Quick answer

Sakura Qwen3.8 Flash Next ISTA 3bit 352E needs about 63.7 GB of memory at IQ3_XXS (a 62.4 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 ISTA 3bit 352E – a 139B sparse MoE GGUF model

Sakura Qwen3.8 Flash Next ISTA 3bit 352E is a 139 billion‑parameter Qwen4exp model with 512 routed experts per layer, 10 active per token. It is quantized to IQ3_XXS, split into two shards, and runs unmodified in llama.cpp, Ollama, and LM Studio.

Good at

  • Reasoning
  • Tools integration
  • Sparse expert routing

Limits

  • Requires ~63.7 GB VRAM at 8 k context
  • No vision support
  • License type: other

Best for

  • Large‑context reasoning tasks
  • Tool‑driven workflows
  • Sparse expert inference

Who it’s for. Developers needing a high‑capacity, sparse MoE model for reasoning and tool use. The model is suitable for environments with large VRAM and where non‑uniform quantization is acceptable.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ3_XXS62.4 GB (2 parts) 63.7 GB
Too big Get file →

For 64 GB VRAM, use the IQ3_XXS quantized file (62.39 GB). If limited to 32 GB, a 2.5‑bit sibling is required. The IQ3_XXS file is split into two shards, so load both for full functionality.

What GPU do you need for Sakura Qwen3.8 Flash Next ISTA 3bit 352E?

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 ISTA 3bit 352E quant should you choose?

IQ3_XXS (62.4 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

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)195
Last updated2026-10-08

Frequently asked questions about Sakura Qwen3.8 Flash Next ISTA 3bit 352E

How much VRAM does Sakura Qwen3.8 Flash Next ISTA 3bit 352E need?

About 63.7 GB at IQ3_XXS with an 8K context window. That is the 62.4 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 Sakura Qwen3.8 Flash Next ISTA 3bit 352E on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 63.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Sakura Qwen3.8 Flash Next ISTA 3bit 352E GGUF?

It depends on the quant: IQ3_XXS is 62.4 GB.

What context length does Sakura Qwen3.8 Flash Next ISTA 3bit 352E 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 ISTA 3bit 352E GGUF?

The files are on Hugging Face in the webmp3/Sakura-Qwen3.8-Flash-Next-ISTA-3bit-352E-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 ISTA 3bit 352E 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 8 GB VRAM?

No, the model requires about 63.7 GB VRAM at 8 k context, far exceeding 8 GB.

Does this model support vision tasks?

No, the model does not include a vision projector and is not multimodal.

What is the context length of this model?

The trained context length is 262,144 tokens.

Other GGUFs of 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 →