Instructions to use mlxstudio/qwen3-4b-4bit-terminal with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlxstudio/qwen3-4b-4bit-terminal with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("mlxstudio/qwen3-4b-4bit-terminal")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps
LM Studio

Pi new

How to use mlxstudio/qwen3-4b-4bit-terminal with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "mlxstudio/qwen3-4b-4bit-terminal"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "mlxstudio/qwen3-4b-4bit-terminal"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use mlxstudio/qwen3-4b-4bit-terminal with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "mlxstudio/qwen3-4b-4bit-terminal"

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 mlxstudio/qwen3-4b-4bit-terminal

Run Hermes

hermes

MLX LM

How to use mlxstudio/qwen3-4b-4bit-terminal with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "mlxstudio/qwen3-4b-4bit-terminal"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "mlxstudio/qwen3-4b-4bit-terminal"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "mlxstudio/qwen3-4b-4bit-terminal",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

devpreneur commited on Jan 22

Commit

2a42b10

verified ·

1 Parent(s): 40e2af7

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +14 -5

README.md CHANGED Viewed

@@ -16,19 +16,28 @@ pipeline_tag: text-generation
 Fine-tuned Qwen3-4B model for terminal command generation. Optimized for LocalTerm app.
 ## Model Details
-- **Base Model**: mlx-community/Qwen3-4B-4bit
-- **Fine-tuning**: QLoRA (4-bit quantized, 16 layers)
 - **Training Data**: 388 examples, 74 terminal commands
 - **Accuracy**: 98% on test set (147/150 correct)
-- **Size**: ~2.3GB (4-bit quantized)
-- **Format**: MLX safetensors (merged, no adapter needed)
 ## Usage
 ### With MLX-LM (Python)
 ### With LocalTerm (macOS app)
 Model auto-downloads on first run. See [LocalTerm](https://github.com/aleonis/localterm).
@@ -44,7 +53,7 @@ Model auto-downloads on first run. See [LocalTerm](https://github.com/aleonis/lo
 ## Version History
 - **v2 (2026-01-22)**: Re-fused model with correct weight format
-  - Fixed:  prefix issue in LoRA merged weights
   - Now compatible with mlx-swift-lm
 - **v1 (2026-01-21)**: Initial release (had loading issues in Swift)

 Fine-tuned Qwen3-4B model for terminal command generation. Optimized for LocalTerm app.
+> **Note**: HuggingFace shows "0.6B params" - this is incorrect. The actual model has **4 billion parameters** (4-bit quantized). HuggingFace miscalculates param count for MLX quantized safetensors files.
 ## Model Details
+- **Base Model**: [mlx-community/Qwen3-4B-4bit](https://huggingface.co/mlx-community/Qwen3-4B-4bit)
+- **Actual Parameters**: **4 billion** (same as base model)
+- **Quantization**: 4-bit (MLX format, ~2.3GB file size)
+- **Fine-tuning**: QLoRA on 16 layers
 - **Training Data**: 388 examples, 74 terminal commands
 - **Accuracy**: 98% on test set (147/150 correct)
 ## Usage
 ### With MLX-LM (Python)
+```python
+from mlx_lm import load, generate
+model, tokenizer = load("mlxstudio/qwen3-4b-4bit-terminal")
+prompt = "how to create a git repository"
+response = generate(model, tokenizer, prompt=prompt, max_tokens=100)
+print(response)
+```
 ### With LocalTerm (macOS app)
 Model auto-downloads on first run. See [LocalTerm](https://github.com/aleonis/localterm).
 ## Version History
 - **v2 (2026-01-22)**: Re-fused model with correct weight format
+  - Fixed: `.linear.` prefix issue in LoRA merged weights
   - Now compatible with mlx-swift-lm
 - **v1 (2026-01-21)**: Initial release (had loading issues in Swift)