ukisai
Swift Qwen3.8 27B
Swift Qwen3.8 27B is a 27.3B dense qwen35 model from ukisai. Configured context length is 262,144 tokens. License: other. Quant sizes are the GGUF file bytes published on Hugging Face when a file exists.
Parameters27.3B
Context256K
Attentionhybrid
Licenseother
64 layers · 4 KV heads · head dim 256 · 48 linear-attention layers · GGUF repo · config
Quant file sizes
| Quant | File size | Source |
|---|---|---|
| F16 | 50.9 GB | Hugging Face file bytes |
| Q8_0 | 27.1 GB | Hugging Face file bytes |
| Q6_K | 66.2 GB | Hugging Face file bytes |
| Q5_K_M | 18.8 GB | Hugging Face file bytes |
| Q5_K_S | 18.4 GB | Hugging Face file bytes |
| Q4_K_M | 16.8 GB | Hugging Face file bytes |
| Q4_K_S | 15.4 GB | Hugging Face file bytes |
| IQ4_NL | 16.4 GB | Hugging Face file bytes |
| IQ4_XS | 14.6 GB | Hugging Face file bytes |
| Q4_1 | 16.3 GB | Hugging Face file bytes |
| Q4_0 | 14.9 GB | Hugging Face file bytes |
| Q3_K_L | 13.4 GB | Hugging Face file bytes |
| Q3_K_M | 12.7 GB | Hugging Face file bytes |
| IQ3_M | 14.0 GB | Hugging Face file bytes |
| Q3_K_S | 12.1 GB | Hugging Face file bytes |
| IQ3_XXS | 11.7 GB | Hugging Face file bytes |
| Q2_K | 10.3 GB | Hugging Face file bytes |
| IQ2_S | 9.22 GB | Hugging Face file bytes |
| IQ2_XXS | 8.47 GB | Hugging Face file bytes |
On common machines
| Device | Verdict | Quant | Speed |
|---|---|---|---|
| GeForce RTX 4060 Ti 8GB | Partial offload | Q4_K_M | 3.7–5.0 |
| GeForce RTX 3060 12GB | Tight fit | IQ2_S | 19–26 |
| GeForce RTX 4060 Ti 16GB | Tight fit | Q3_K_L | 11–15 |
| GeForce RTX 4090 | Tight fit | Q5_K_M | 27–36 |
| GeForce RTX 5090 | Tight fit | Q8_0 | 33–45 |
| Apple M4 Pro | Tight fit | Q4_K_M | 8.2–11 |
| Apple M5 | Tight fit | IQ2_S | 8.1–11 |
| CPU only · DDR5-5600 | Runs on CPU | Q5_K_M | 2.4–3.2 |
Smallest devices that fit it fully
- GeForce RTX 2080 Ti — Tight fit, IQ2_XXS, 35–48 tok/s
- GeForce GTX 1080 Ti — Tight fit, IQ2_XXS, 28–38 tok/s
- GeForce RTX 5070 — Tight fit, IQ2_S, 35–48 tok/s
- GeForce RTX 4070 Ti — Tight fit, IQ2_S, 27–36 tok/s
- GeForce RTX 4070 Super — Tight fit, IQ2_S, 27–36 tok/s
- GeForce RTX 4070 — Tight fit, IQ2_S, 27–36 tok/s
Run it
Commands for the GeForce RTX 4060 Ti 16GB estimate (Q3_K_L).
llama.cpp
llama-server -hf ukisai/Swift-Qwen3.8-27B-GGUF:Q3_K_L -c 8192 -ngl 64
No matching name in the Ollama library snapshot.
LM Studio
lms get ukisai/Swift-Qwen3.8-27B-GGUF # Then load quant Q3_K_L