Downloads · 30 days
29
5% of all-time downloads
chuanman2707/capyinn-gemma-4-e2b-it-q5-k-m-gguf
capyinn-gemma-4-e2b-it-q5-k-m-gguf is a text generation model from chuanman2707. Use it when you need the model to write or continue text. It is set up for llama.cpp. The card lists the license as apache-2.0.
CapyInn Gemma 4 is a hackathon demo model for a Vietnamese hotel receptionist assistant.
Downloads · 30 days
29
5% of all-time downloads
All-time downloads
618
Public
Repo size
3.6 GB
Likes
1
Public
Click a slice to open those files.
.gguf3.6 GB · 100%
From the Hugging Face model README
CapyInn Gemma 4 is a hackathon demo model for a Vietnamese hotel receptionist assistant.
This GGUF file is a merged and quantized demo build based on:
google/gemma-4-E2B-itcapyinn-gemma-4-Q5_K_M.gguf
SHA256:
9e1a48929f10a7e307b81b7987452911915923feea65e17029b447bd95df6a55
This model is intended for hackathon demos and local prototyping of a hotel receptionist assistant.
Example use cases:
This is a preview demo model, not a production hotel receptionist system.
The model may produce incorrect, incomplete, repeated, or overconfident answers. It must not be used to make real booking, payment, refund, checkout, privacy, safety, legal, or operational decisions without a verified PMS/tool/RAG layer and human review.
In particular, without verified tool/PMS/RAG results, the assistant should not:
You are a hotel receptionist assistant for CapyInn.
Hard rules:
- If there is no tool/PMS result, do not claim that you checked the system.
- If there is no tool/PMS result, do not confirm room availability, booking status, airport pickup, payment, invoice, refund, checkout, or housekeeping completion.
- If there is no hotel knowledge/RAG context, do not confirm specific policies, facilities, prices, or service availability.
- For booking or availability requests, ask for missing details and say hotel staff/system must verify before confirmation.
- For operational requests, acknowledge the request and say it will be passed to hotel staff; do not claim it has already been completed.
- Never output tool calls, JSON, role labels, or the word "model".
- Reply in the guest's language unless they request otherwise.
llama-cli \
-m capyinn-gemma-4-Q5_K_M.gguf \
-ngl 99 \
-c 2048 \
-n 200 \
--temp 0 \
--top-k 1 \
-st \
--no-display-prompt \
--reasoning off \
-sys 'You are a hotel receptionist assistant for CapyInn. If there is no tool/PMS/RAG result, do not claim that you checked the system and do not confirm booking, payment, pricing, policy, facility, or operational status.' \
-p 'Khach hoi: Toi muon check-in som thi khach san co ho tro khong?'
This build is optimized for a local hackathon demo on Apple Silicon using llama.cpp Metal.
It is not yet optimized for:
Internal fixed-prompt qualitative evaluation showed useful hospitality behavior, but the model did not pass the full hard-rule production gate.
Known remaining risks include:
For a production assistant, use this model behind routing, RAG/PMS context injection, and a hard-rule output validator.
This model is based on google/gemma-4-E2B-it.
The GGUF conversion and quantization were performed with llama.cpp.