Apppublic

kaushikdash/odia-edge-voice

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face

0likes

App README

OdiaEdgeVoice

This Space runs the local GGUF model kaushikdash/odia-gemma4-e2b-gguf directly with llama.cpp.

Public API

After the Space is live, you can call it from any machine.

OpenAI-compatible Python example:

python

from openai import OpenAI

client = OpenAI(
    base_url="https://kaushikdash-odia-edge-voice.hf.space/v1",
    api_key="no-key",
)

resp = client.chat.completions.create(
    model="odia-gemma4-e2b.Q4_K_M.gguf",
    messages=[
        {"role": "system", "content": "ଆପଣ ଜଣେ ଓଡ଼ିଆ ସହାୟକ।"},
        {"role": "user", "content": "ଓଡ଼ିଶାର ରାଜଧାନୀ କଣ?"},
    ],
    temperature=0.4,
)
print(resp.choices[0].message.content)

Raw HTTP example:

bash

curl -X POST "https://kaushikdash-odia-edge-voice.hf.space/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer no-key" \
  -d '{
    "model": "odia-gemma4-e2b.Q4_K_M.gguf",
    "messages": [
      {"role": "system", "content": "ଆପଣ ଜଣେ ଓଡ଼ିଆ ସହାୟକ।"},
      {"role": "user", "content": "ଓଡ଼ିଶାର ରାଜଧାନୀ କଣ?"}
    ],
    "temperature": 0.4
  }'

Main endpoints:

—/v1/chat/completions
—/health
—/docs may be available depending on the server build

Notes

—This Space uses your own GGUF model, not a provider API.
—It is intended to be called by the separate Streamlit client or any OpenAI-compatible client.
—The selected file is odia-gemma4-e2b.Q4_K_M.gguf.