Instructions to use smshahbaj/RIFA-Edge-0.6B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use smshahbaj/RIFA-Edge-0.6B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="smshahbaj/RIFA-Edge-0.6B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("smshahbaj/RIFA-Edge-0.6B")
model = AutoModelForCausalLM.from_pretrained("smshahbaj/RIFA-Edge-0.6B", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use smshahbaj/RIFA-Edge-0.6B 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 smshahbaj/RIFA-Edge-0.6B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf smshahbaj/RIFA-Edge-0.6B:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf smshahbaj/RIFA-Edge-0.6B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf smshahbaj/RIFA-Edge-0.6B:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf smshahbaj/RIFA-Edge-0.6B:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf smshahbaj/RIFA-Edge-0.6B:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf smshahbaj/RIFA-Edge-0.6B:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf smshahbaj/RIFA-Edge-0.6B:Q4_K_M

Use Docker

docker model run hf.co/smshahbaj/RIFA-Edge-0.6B:Q4_K_M

LM Studio
Jan

vLLM

How to use smshahbaj/RIFA-Edge-0.6B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "smshahbaj/RIFA-Edge-0.6B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "smshahbaj/RIFA-Edge-0.6B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/smshahbaj/RIFA-Edge-0.6B:Q4_K_M

SGLang

How to use smshahbaj/RIFA-Edge-0.6B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "smshahbaj/RIFA-Edge-0.6B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "smshahbaj/RIFA-Edge-0.6B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "smshahbaj/RIFA-Edge-0.6B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "smshahbaj/RIFA-Edge-0.6B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use smshahbaj/RIFA-Edge-0.6B with Ollama:
```
ollama run hf.co/smshahbaj/RIFA-Edge-0.6B:Q4_K_M
```
Unsloth Desktop

How to use smshahbaj/RIFA-Edge-0.6B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf smshahbaj/RIFA-Edge-0.6B: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": "smshahbaj/RIFA-Edge-0.6B:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use smshahbaj/RIFA-Edge-0.6B with Docker Model Runner:
```
docker model run hf.co/smshahbaj/RIFA-Edge-0.6B:Q4_K_M
```

Lemonade

How to use smshahbaj/RIFA-Edge-0.6B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull smshahbaj/RIFA-Edge-0.6B:Q4_K_M

Run and chat with the model

lemonade run user.RIFA-Edge-0.6B-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use smshahbaj/RIFA-Edge-0.6B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf smshahbaj/RIFA-Edge-0.6B: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 smshahbaj/RIFA-Edge-0.6B:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use smshahbaj/RIFA-Edge-0.6B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf smshahbaj/RIFA-Edge-0.6B: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 "smshahbaj/RIFA-Edge-0.6B: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"

⚡ RIFA Edge 0.6B

Compact coding assistant with security awareness.
Write code · debug · explain · prefer safer patterns.

Part of the RIFA series by SM Shahbaj
Nano (0.5B) · Edge (0.6B) · Flash (1.7B) · Pro (3B)

Why Edge


Size	~0.6B — practical on modest GPUs / quantized CPU
Focus	Code generation, debugging, concept explanations
Security lens	Common pitfalls (injection, secrets, weak auth) + safer alternatives
Languages	English · বাংলা · Banglish
Style	Direct answers (thinking mode off)

Trained with a coding-heavy mix: high-quality instruction→code data, OSS-grounded problems, and educational secure-coding examples (vulnerable pattern + fix). Identity locked to SM Shahbaj / RIFA Edge.

Quick start

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "smshahbaj/RIFA-Edge-0.6B"
tok = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id, torch_dtype=torch.float16, device_map="auto", trust_remote_code=True
)

messages = [{"role": "user", "content": "Write a Python function that safely hashes a password."}]
prompt = tok.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True, enable_thinking=False
)
inputs = tok(prompt, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=256, do_sample=True, temperature=0.5, top_p=0.9)
print(tok.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

Training summary

Method: QLoRA (4-bit NF4) → LoRA r=32 / α=64 → merge to FP16
Objective: Completion-only loss on assistant turns
Data: Code instructions + OSS-style problems + educational secure coding + light general/Bangla + repeated identity
Thinking: disabled

Intended use

Local / offline coding help
Learning secure defaults while writing features
Bangla / Banglish coding chat

Limitations

Small model: weaker on very long multi-file reasoning and rare APIs
No live knowledge — fixed "Sorry…" line when appropriate
Always review generated code before production; security guidance is educational, not a substitute for professional audit

License

Apache 2.0

RIFA Edge · SM Shahbaj

Available formats

This repository contains the original model release plus additional runtime formats.

Transformers / Safetensors

The original Transformers/Safetensors files remain in this repository.

GGUF

Generated GGUF files are stored under GGUF/:

RIFA-Edge-0.6B-F16.gguf
RIFA-Edge-0.6B-Q8_0.gguf
RIFA-Edge-0.6B-Q6_K.gguf
RIFA-Edge-0.6B-Q5_K_M.gguf
RIFA-Edge-0.6B-Q4_K_M.gguf

LoRA

No standalone LoRA adapter was found. A LoRA adapter cannot be reconstructed from a merged model alone.

GGUF files are quantized exports of the same model, not separate fine-tunes.

Downloads last month: -

Safetensors

Model size

0.6B params

Tensor type

F16