Instructions to use kimjolo/jaib-assistant with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use kimjolo/jaib-assistant with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf kimjolo/jaib-assistant:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf kimjolo/jaib-assistant:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf kimjolo/jaib-assistant:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf kimjolo/jaib-assistant:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf kimjolo/jaib-assistant:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf kimjolo/jaib-assistant:Q4_K_M

Use Docker

docker model run hf.co/kimjolo/jaib-assistant:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use kimjolo/jaib-assistant with Ollama:
```
ollama run hf.co/kimjolo/jaib-assistant:Q4_K_M
```
Unsloth Desktop

How to use kimjolo/jaib-assistant with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kimjolo/jaib-assistant:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "kimjolo/jaib-assistant:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use kimjolo/jaib-assistant with Docker Model Runner:
```
docker model run hf.co/kimjolo/jaib-assistant:Q4_K_M
```

Lemonade

How to use kimjolo/jaib-assistant with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull kimjolo/jaib-assistant:Q4_K_M

Run and chat with the model

lemonade run user.jaib-assistant-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use kimjolo/jaib-assistant with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kimjolo/jaib-assistant:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default kimjolo/jaib-assistant:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use kimjolo/jaib-assistant with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kimjolo/jaib-assistant:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "kimjolo/jaib-assistant:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

JAIB Assistant

The on-phone model of JAIB, a money app for Pakistan. It is Qwen3.5-0.8B fine-tuned to turn one request, in English or Roman Urdu, into the app's JSON: add a payment or a loan, fix or remove one, find payments, explain a total, open a part of the app, or say how to do something in JAIB.

The app downloads this file once and runs it on the phone. Nothing a person types is sent anywhere.

Trained only on invented requests (no real person's messages), with the app's own prompt. On 150 held-out invented requests it chose the right action 149 times (the base model: 65).

File: Qwen3.5-0.8B.Q4_K_M.gguf (Q4_K_M, 541,903,776 bytes) SHA-256: bd00764ba1ab9d0f2fd433e8c49604f1caa595c46355ae51ddfe385a3492008b

License: Apache 2.0, as the base model.

Downloads last month: 67

GGUF

Model size

0.8B params

Architecture

qwen35

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for kimjolo/jaib-assistant

Base model

Qwen/Qwen3.5-0.8B-Base

Finetuned

Qwen/Qwen3.5-0.8B

Finetuned

unsloth/Qwen3.5-0.8B

Quantized

(19)

this model