Downloads · 30 days
22
4% of all-time downloads
kanakjr/homebot-qwen3.5-4b-gguf
homebot-qwen3.5-4b-gguf is a text generation model from kanakjr. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
HomeBot qwen3.5 4B fine-tune exported to GGUF for local Ollama inference.
Downloads · 30 days
22
4% of all-time downloads
All-time downloads
525
Public
Repo size
13.7 GB
Likes
0
Public
Click a slice to open those files.
.gguf13.7 GB · 100%
From the Hugging Face model README
HomeBot qwen3.5 4B fine-tune exported to GGUF for local Ollama inference.
Built by finetuning/finalize_gguf_colab.ipynb from the merged fp16 weights at
kanakjr/homebot-qwen3.5-4b-gguf-merged16.
| Quant | Approx. size | Recommended for |
|---|---|---|
| Q4_K_M | smallest | < 8 GB RAM machines |
| Q5_K_M | good balance | |
| Q6_K | ~1-2% quality gap vs. fp16 | |
| Q8_0 | largest | > 16 GB RAM -- default for M-series Macs |
hf download kanakjr/homebot-qwen3.5-4b-gguf --include "*.Q8_0.gguf" --include "*.Modelfile" --local-dir ~/homebot-model
cd ~/homebot-model
ollama create homebot-qwen3_5-4b -f homebot-qwen3_5-4b.Modelfile
ollama run homebot-qwen3_5-4b "turn off the air purifier"
See finetuning/README.md in the HomeBot repo for the full pipeline.