empero-ai
Qwen3.8 9B Distill
Qwen3.8 9B Distill is a 9.20B dense qwen35 model from empero-ai. Configured context length is 262,144 tokens. License: apache-2.0. Quant sizes are the GGUF file bytes published on Hugging Face when a file exists.
Parameters9.2B
Context256K
Attentionhybrid
Licenseapache-2.0
32 layers · 4 KV heads · head dim 256 · 24 linear-attention layers · GGUF repo · config
Quant file sizes
| Quant | File size | Source |
|---|---|---|
| BF16 | 17.1 GB | Hugging Face file bytes |
| Q8_0 | 9.11 GB | Hugging Face file bytes |
| Q6_K | 7.04 GB | Hugging Face file bytes |
| Q5_K_M | 6.19 GB | Hugging Face file bytes |
| Q4_K_M | 5.38 GB | Hugging Face file bytes |
On common machines
| Device | Verdict | Quant | Speed |
|---|---|---|---|
| GeForce RTX 4060 Ti 8GB | Tight fit | Q5_K_M | 23–31 |
| GeForce RTX 3060 12GB | Tight fit | Q8_0 | 20–27 |
| GeForce RTX 4060 Ti 16GB | Runs | Q8_0 | 16–21 |
| GeForce RTX 4090 | Runs | BF16 | 30–40 |
| GeForce RTX 5090 | Runs great | BF16 | 53–72 |
| Apple M4 Pro | Runs great | Q8_0 | 15–20 |
| Apple M5 | Tight fit | Q8_0 | 8.4–11 |
| CPU only · DDR5-5600 | Runs on CPU | BF16 | 2.6–3.6 |
Smallest devices that fit it fully
- GeForce RTX 5060 Ti 8GB — Tight fit, Q5_K_M, 36–49 tok/s
- GeForce RTX 5060 — Tight fit, Q5_K_M, 36–49 tok/s
- GeForce RTX 5050 — Tight fit, Q5_K_M, 25–35 tok/s
- GeForce RTX 4060 Ti 8GB — Tight fit, Q5_K_M, 23–31 tok/s
- GeForce RTX 4060 — Tight fit, Q5_K_M, 22–30 tok/s
- GeForce RTX 3070 Ti — Tight fit, Q5_K_M, 48–66 tok/s
Run it
Commands for the GeForce RTX 4060 Ti 16GB estimate (Q8_0).
llama.cpp
llama-server -hf empero-ai/Qwen3.8-9B-Distill-GGUF:Q8_0 -c 8192 -ngl 32
Ollama
ollama run qwen3.8
LM Studio
lms get empero-ai/Qwen3.8-9B-Distill-GGUF # Then load quant Q8_0