Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX needs about 16.9 GB of memory at Q4_0 (a 14.2 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 2 quants, from 14.2 GB (Q4_0) to 15.3 GB (Q4_0). Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.
Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX – a reasoning‑efficient 27B Qwen model for AMD GPUs
Swift 1.5 Qwen3.8 27B is a 27.3‑billion‑parameter Qwen model optimized for reasoning tasks. It uses ROCmFP4 quantization and is designed for AMD GPUs with a llama.cpp build that supports ROCmFP4 tensor types. The model is derived from Swift‑1.0 and offers improved speed and token efficiency.
Good at
- Reasoning tasks
- Long‑horizon agentic scenarios
- Coding challenges
Limits
- Requires a ROCmFP4‑capable llama.cpp build
- No vision support
- License is other (Swift Open License v1.0)
Best for
- Agentic game design
- Complex coding assistance
- Long‑form reasoning
Who it’s for. Developers and researchers who need a large Qwen model with efficient reasoning on AMD GPUs. The model is suitable for those who can build or use a ROCmFP4‑enabled llama.cpp fork.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 14.2 GB | 16.9 GB | Too big | Get file → | |
| Q4_0 | 15.3 GB | 18.0 GB | Too big | Get file → |
For 8 GB VRAM, use the 14.2 GB Q4_0 file – it estimates 16.9 GB memory at 8 k context. For 12 GB VRAM, the 15.27 GB Q4_0 file is suitable, estimating 18 GB memory at 8 k context. Both files are approximate and require a ROCmFP4‑capable build.
What GPU do you need for Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX?
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 Q4_0 (15.3 GB file, about 18.0 GB in use).
Which Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX quant should you choose?
Q4_0 (14.2 GB) is the middle option in this repository. 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
Source
Frequently asked questions about Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX
How much VRAM does Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX need?
About 16.9 GB at Q4_0 with an 8K context window. That is the 14.2 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 27B Q4 0 ROCMFP4 STRIX on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 16.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX GGUF?
It depends on the quant: Q4_0 is 14.2 GB, Q4_0 is 15.3 GB.
What context length does Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX 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 Swift 1.5 Qwen3.8 27B Q4 0 ROCMFP4 STRIX GGUF?
The files are on Hugging Face in the gray117/Swift-1.5-Qwen3.8-27B-Q4_0_ROCMFP4_STRIX 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 27B Q4 0 ROCMFP4 STRIX 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 27B on an 8 GB VRAM GPU?
The 14.2 GB Q4_0 file estimates 16.9 GB memory at 8 k context, so it will not fit on 8 GB VRAM without further optimization.
Does this model support vision tasks?
No, the model has no vision capabilities.
What license applies to this model?
It uses the Swift Open License v1.0 with Qwen components retaining Apache‑2.0.