Instructions to use Steviewei/mini-me-v7c with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Steviewei/mini-me-v7c 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 Steviewei/mini-me-v7c:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Steviewei/mini-me-v7c:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Steviewei/mini-me-v7c:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Steviewei/mini-me-v7c: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 Steviewei/mini-me-v7c:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Steviewei/mini-me-v7c: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 Steviewei/mini-me-v7c:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Steviewei/mini-me-v7c:Q4_K_M

Use Docker

docker model run hf.co/Steviewei/mini-me-v7c:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use Steviewei/mini-me-v7c with Ollama:
```
ollama run hf.co/Steviewei/mini-me-v7c:Q4_K_M
```
Unsloth Desktop

How to use Steviewei/mini-me-v7c with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Steviewei/mini-me-v7c: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": "Steviewei/mini-me-v7c:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Steviewei/mini-me-v7c with Docker Model Runner:
```
docker model run hf.co/Steviewei/mini-me-v7c:Q4_K_M
```

Lemonade

How to use Steviewei/mini-me-v7c with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Steviewei/mini-me-v7c:Q4_K_M

Run and chat with the model

lemonade run user.mini-me-v7c-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Steviewei/mini-me-v7c with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Steviewei/mini-me-v7c: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 Steviewei/mini-me-v7c:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Steviewei/mini-me-v7c with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Steviewei/mini-me-v7c: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 "Steviewei/mini-me-v7c: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"

mini-me-v7c

Qwen3-0.6B + LoRA (r=16, alpha=32, train tren BASE - chong forgetting), merge full. Worker-style Copilot nho: Buoc → Kiem tra lai → Dap an, self-correction, khong bia khi thieu du kien.

Train

Data: cot_v7.jsonl (2009 examples, builder scripts/build_data_v7.py) - backup R2 distill_v2/data/
3 epochs · 378 steps (resume tu checkpoint-357) · final loss 0.0200 · bf16

Eval 47 prompt (scorer giong het bench_v7.py)

ban	tong	math	refuse	knowledge	task	json	code	contains	persona	max_words
adapter bf16 (transformers)	44/3	9/0	10/0	8/0	5/0	2/0	1/0	6/3	1/0	2/0
GGUF q8_0 (khuyen dung)	44/3	9/0	10/0	7/1	5/0	2/0	1/0	7/2	1/0	2/0
GGUF q4_k_m	40/7	8/1	10/0	6/2	5/0	2/0	1/0	5/4	1/0	2/0

Harness GGUF: llama.cpp b11835 llama-server, temp 0, max_tokens 256.
Do dai tra loi: -51% tokens (adapter 180.1 vs base 367.9).

Canh bao quant

q4_k_m lam hong so hoc model 0.6B - vi du 17 x 23 = 411 (dung 391), kem self-check sai theo (ca 47 prompt: q4 mat 4 diem so voi q8_0). Worker khong-bia → dung q8_0. f16 (1198 MB) theo policy khong len HF - nam tren R2 distill_v2/exports/.

Receipts (sha256)

q4_k_m: 48f09e1aaa146d87effcd47393d6e46e163b93ab8acd7e30ad0fa4c0752e25e8
q8_0: a81a2baaf9d47bae5bbe9fc4ec33f8471f0dce1d5f580654bae74dc9f63abb13
f16 (R2): 81c0d1cae291298288bb481809359de2b49dceaa08a06b136fb294354b2e0315

Usage

llama-server -m mini-me-v7c-q8_0.gguf --port 8899
# style: "Buoc: ... / Kiem tra lai: ... / Dap an: ..."

Backup chain

R2 distill_v2/exports/: f16 + q4_k_m + q8_0 + MANIFEST_v7c.json (manifest co day du eval + hf info)
R2 distill_v2/runs/final_adapter_v7c/: LoRA adapter (40.4 MB)
Xac thuc: ETag(MD5) R2 == MD5 tinh lai tren box (23/10)

Downloads last month: 73

GGUF

Model size

0.6B params

Architecture

qwen3

Hardware compatibility

4-bit

8-bit

Inference Providers NEW

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

Model tree for Steviewei/mini-me-v7c

Base model

Qwen/Qwen3-0.6B-Base

Finetuned

Qwen/Qwen3-0.6B

Quantized

(510)

this model