empero-ai
Qwen3.8 2B Distill
Qwen3.8 2B Distill is a 1.94B 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.
Parameters1.9B
Context256K
Attentionhybrid
Licenseapache-2.0
24 layers · 2 KV heads · head dim 256 · 18 linear-attention layers · GGUF repo · config
Quant file sizes
| Quant | File size | Source |
|---|---|---|
| BF16 | 3.63 GB | Hugging Face file bytes |
| Q8_0 | 1.93 GB | Hugging Face file bytes |
| Q6_K | 1.50 GB | Hugging Face file bytes |
| Q5_K_M | 1.35 GB | Hugging Face file bytes |
| Q4_K_M | 1.22 GB | Hugging Face file bytes |
On common machines
| Device | Verdict | Quant | Speed |
|---|---|---|---|
| GeForce RTX 4060 Ti 8GB | Runs great | BF16 | 40–54 |
| GeForce RTX 3060 12GB | Runs great | BF16 | 50–68 |
| GeForce RTX 4060 Ti 16GB | Runs great | BF16 | 40–54 |
| GeForce RTX 4090 | Runs great | BF16 | 139–189 |
| GeForce RTX 5090 | Runs great | BF16 | 247–336 |
| Apple M4 Pro | Runs great | BF16 | 38–51 |
| Apple M5 | Runs great | BF16 | 21–29 |
| CPU only · DDR5-5600 | Runs on CPU | BF16 | 12–17 |
Smallest devices that fit it fully
- GeForce RTX 3050 6GB — Runs, BF16, 23–32 tok/s
- GeForce RTX 2060 6GB — Runs, BF16, 46–63 tok/s
- GeForce GTX 1660 Super — Runs, BF16, 46–63 tok/s
- GeForce GTX 1660 Ti — Runs, BF16, 40–54 tok/s
- Intel Arc A380 — Runs, BF16, 26–35 tok/s
- GeForce RTX 5060 Ti 8GB — Runs great, BF16, 62–84 tok/s
Run it
Commands for the GeForce RTX 4060 Ti 16GB estimate (BF16).
llama.cpp
llama-server -hf empero-ai/Qwen3.8-2B-Distill-GGUF:BF16 -c 8192 -ngl 24
Ollama
ollama run qwen3.8
LM Studio
lms get empero-ai/Qwen3.8-2B-Distill-GGUF # Then load quant BF16