Views
No views yet
Q: "Who are you?"
A: "Silo-Moe"
Q: "Who built you?"
A: "Zindango"| File | Size | Description |
|---|---|---|
silo-moe-q4_k_m.gguf | 15 GB | ⭐ Recommended - Best balance |
silo-moe-q5_k_m.gguf | 16 GB | Higher quality |
silo-moe-q6_k.gguf | 21 GB | Very high quality |
silo-moe-q8_0.gguf | 21 GB | Near lossless |
silo-moe-f16.gguf | 39 GB | Full precision |
1# Download model
2wget https://huggingface.co/$HF_USERNAME/silo-moe-gguf/resolve/main/silo-moe-q4_k_m.gguf
3
4# Create Modelfile
5cat > Modelfile << 'MODELFILE'
6FROM ./silo-moe-q4_k_m.gguf
7PARAMETER temperature 0.7
8PARAMETER top_p 0.9
9PARAMETER top_k 40
10MODELFILE
11
12# Import to Ollama
13ollama create silo-moe:q4km -f Modelfile
14
15# Run
16ollama run silo-moe:q4km "Who are you?"1# Download
2wget https://huggingface.co/$HF_USERNAME/silo-moe-gguf/resolve/main/silo-moe-q4_k_m.gguf
3
4# Run
5./llama-cli -m silo-moe-q4_k_m.gguf -p "Who are you?" -n 501@misc{silo-moe,
2 author = {Zindango},
3 title = {Silo-Moe: Fine-tuned GPT-OSS-20B},
4 year = {2025},
5 publisher = {Hugging Face},
6 howpublished = {\url{https://huggingface.co/$HF_USERNAME/silo-moe-gguf}}
7}