QW

Swift 1.5 Qwen3.8 Flash Next

QwenGGUF by ukisaiUpdated 2026-09-24
Download on Hugging Face
Parameters177B
Context256K
Layers48
Architectureqwen4exp
Smallest quant70.1 GB
Quick answer

Swift 1.5 Qwen3.8 Flash Next needs about 120.9 GB of memory at Q4_K_M (a 119.6 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 27 quants, from 70.1 GB (IQ1_S) to 188.3 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 3.2 GB.

Swift 1.5 Qwen3.8 Flash Next – reasoning‑efficient Qwen model

Swift 1.5 Qwen3.8 Flash Next is a 177 billion‑parameter Qwen model with a mixture of experts. It was trained on 262144 context tokens and supports reasoning and vision tasks. The model is available in multiple GGUF quantizations via the ukisai/Swift-1.5-Qwen3.8-Flash-Next-GGUF repo.

Good at

  • Efficient reasoning with fewer thinking tokens
  • Vision‑enabled inference
  • Large context handling

Limits

  • Requires large VRAM for higher‑quality quants
  • No explicit license terms provided
  • Vision support only in certain quants

Best for

  • Chatbots needing long context
  • Vision‑based question answering
  • Agent‑style coding tasks

Who it’s for. Developers and researchers building applications that need a large‑context, reasoning‑efficient LLM with optional vision capabilities. The model is suited for environments where inference speed and token efficiency are priorities.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S70.1 GB (2 parts) 71.4 GB
Too big Get file →
IQ1_M72.0 GB (2 parts) 73.3 GB
Too big Get file →
IQ2_XXS75.2 GB (2 parts) 76.5 GB
Too big Get file →
IQ2_XS77.7 GB (2 parts) 79.0 GB
Too big Get file →
IQ2_S77.8 GB (2 parts) 79.1 GB
Too big Get file →
IQ2_M80.4 GB (2 parts) 81.7 GB
Too big Get file →
Q2_K80.9 GB (2 parts) 82.2 GB
Too big Get file →
IQ3_XXS88.0 GB (2 parts) 89.3 GB
Too big Get file →
Q3_K_S89.4 GB (2 parts) 90.7 GB
Too big Get file →
IQ3_XS91.9 GB (3 parts) 93.3 GB
Too big Get file →
Q3_K_M92.0 GB (3 parts) 93.3 GB
Too big Get file →
Q3_K_L93.2 GB (3 parts) 94.5 GB
Too big Get file →
IQ3_M93.2 GB (3 parts) 94.5 GB
Too big Get file →
IQ4_XS97.7 GB (3 parts) 99.0 GB
Too big Get file →
IQ4_NL100.3 GB (3 parts) 101.6 GB
Too big Get file →
Q4_0100.6 GB (3 parts) 101.9 GB
Too big Get file →
Q2_K_L107.1 GB (4 parts) 108.5 GB
Too big Get file →
Q4_1111.3 GB (3 parts) 112.6 GB
Too big Get file →
Q4_K_S113.1 GB (3 parts) 114.4 GB
Too big Get file →
Q3_K_XL119.3 GB (4 parts) 120.7 GB
Too big Get file →
Q4_K_M119.6 GB (3 parts) 120.9 GB
Too big Get file →
Q5_K_S128.1 GB (3 parts) 129.4 GB
Too big Get file →
Q5_K_M134.7 GB (4 parts) 136.0 GB
Too big Get file →
Q4_K_L139.3 GB (4 parts) 140.6 GB
Too big Get file →
Q5_K_L151.1 GB (5 parts) 152.4 GB
Too big Get file →
Q6_K168.1 GB (5 parts) 169.4 GB
Too big Get file →
Q8_0188.3 GB (5 parts) 189.6 GB
Too big Get file →

For 8 GB VRAM, use Q2_K or Q3_K_S (file sizes 80.93 GB and 89.44 GB, memory estimates 82.2 GB and 90.7 GB). For 16 GB VRAM, Q4_0 or Q4_K_S (100.6 GB/101.9 GB and 113.14 GB/114.4 GB). For 24 GB VRAM, Q5_K_S or Q5_K_M (128.1 GB/129.4 GB and 134.67 GB/136 GB).

What GPU do you need for Swift 1.5 Qwen3.8 Flash Next?

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 Swift 1.5 Qwen3.8 Flash Next quant should you choose?

Q4_K_M is the usual starting point for Swift 1.5 Qwen3.8 Flash Next: it balances file size and quality well. Here it is a 119.6 GB file.

The Q8_0 file is 188.3 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is IQ1_S at 70.1 GB, with the quality trade-offs that come with fewer bits. 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
Parameters177B (approximate, mixture of experts)
Layers48
Trained context262,144 tokens
KV cache (fp16)96 KB per token

Source

Quantized byukisai
FamilyQwen
Licenseother
Downloads (HF)159,551
Last updated2026-09-24

Frequently asked questions about Swift 1.5 Qwen3.8 Flash Next

How much VRAM does Swift 1.5 Qwen3.8 Flash Next need?

About 120.9 GB at Q4_K_M with an 8K context window. That is the 119.6 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 Swift 1.5 Qwen3.8 Flash Next on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 71.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Swift 1.5 Qwen3.8 Flash Next GGUF?

It depends on the quant: IQ1_S is 70.1 GB, IQ1_M is 72.0 GB, IQ2_XXS is 75.2 GB, IQ2_XS is 77.7 GB, IQ2_S is 77.8 GB, IQ2_M is 80.4 GB and 21 more.

What context length does Swift 1.5 Qwen3.8 Flash Next 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 Swift 1.5 Qwen3.8 Flash Next GGUF?

The files are on Hugging Face in the ukisai/Swift-1.5-Qwen3.8-Flash-Next-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 Swift 1.5 Qwen3.8 Flash Next 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 Swift 1.5 Qwen3.8 Flash Next on 8 GB VRAM?

Use a lower‑quality quant such as Q2_K or Q3_K_S, which estimate 82.2 GB and 90.7 GB of VRAM at 8 k context. The model will still function but with reduced precision.

Does this model support vision tasks?

Yes, the model includes vision capabilities as noted in its detected abilities, but only certain quantizations may expose this feature depending on the runtime.

What is the context length of this model?

The model was trained on 262144 context tokens and can handle that length in inference.

Similar models

Browse all →

More from ukisai

All by ukisai →

More Qwen models

All Qwen →

Other large models

All large models →