Downloads · 30 days
19
7% of all-time downloads
sifat-febo/banglish-companion-3b-gguf
banglish-companion-3b-gguf is a text generation model from sifat-febo. Use it when you need the model to write or continue text. It is set up for gguf. The card lists the license as apache-2.0.
Quantized builds of banglish-companion-3b. What the model is and why it exists are on that page. This one is about running it.
Downloads · 30 days
19
7% of all-time downloads
All-time downloads
256
Public
Repo size
5.8 GB
Likes
0
Public
Click a slice to open those files.
.gguf5.8 GB · 100%
From the Hugging Face model README
Quantized builds of banglish-companion-3b. What the model is and why it exists are on that page. This one is about running it.
Two files below. You need one of them; 2.15 GB for the smaller. No account, no network calls once it is on disk.
kon file ta nibo? duitar moddhe je kono ekta. Q4_K_M chhoto (2.15 GB), Q8_0 boro (3.65 GB) kintu beshi precise. tomar computer er upor depend kore.
ki lagbe? Python ar pip install llama-cpp-python. account lagbe na,
taka lagbe na, chalate internet lagbe na. tumi ja bolba, tomar computer er
baire jabe na.
bipod hole? keu jodi lekhe je bipode ache, companion.py model ke jigges
kore na — lekha uttor dey.
| File | Size |
|---|---|
banglish-companion-3b-Q4_K_M.gguf | 2.15 GB |
banglish-companion-3b-Q8_0.gguf | 3.65 GB |
Both are cut from the same full-precision model. Q4_K_M is the smaller download and needs less memory; Q8_0 keeps more of the original precision. Take whichever suits your machine.
On a Mac, the same model also runs as banglish-companion-3b-mlx, which ships with the perception layer; this repo ships the guard only.
pip install llama-cpp-python huggingface_hub
hf download sifat-febo/banglish-companion-3b-gguf --local-dir banglish-3b
python banglish-3b/companion.py
That starts a chat in your terminal. It first answers aj mood bhalo na
("I'm not in a good mood today"), then whatever you type. In your own code,
from companion import reply is the only import you need. It picks up
whichever .gguf file sits beside it, checks the guard before the model, and
uses the answer settings this model was tested with.
If someone writes that they are in danger, the reply is fixed text from
nirapotta.py, not the model, and reply() checks that first on every turn.
It is plain Python, needs nothing installed, and runs in microseconds.
A truncated GGUF opens without complaint and then generates noise. It does not error. Check the file you got:
shasum -a 256 banglish-companion-3b-*.gguf
48c8fcf77c078b8b5a5b4c404b30b2f40003ee0efcf0f04296640d603ce0f4c5 Q4_K_M
80a20b0244973b4cee6b218d9699a25db173a5e25b6d2556e861c8db2b99f514 Q8_0
The author designed the system, chose the base models, ran all training and measurement on the author's own machine, read the models' conversations, and chose what to publish and what to withhold. Claude Code (Anthropic) was used as a coding and writing tool under that direction.
Apache 2.0. Base model Ministral-3-3B-Base-2512 (Mistral AI, Apache 2.0).
Cite the report, not the packaging — these files carry the 3B's weights, and a quantization format is not a separate contribution. The report is at https://doi.org/10.5281/zenodo.22371896.
@misc{febo2026banglishcompanion,
author = {Febo, Sifat},
title = {Banglish Companion: Small Chat Models for Romanized Bengali
— Technical Report},
year = {2026},
month = sep,
publisher = {Zenodo},
version = {1.0},
doi = {10.5281/zenodo.22371896},
url = {https://doi.org/10.5281/zenodo.22371896},
note = {Repository: sifat-febo/banglish-companion-3b-gguf}
}