Instructions to use alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit 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("alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit")

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 Settings
LM Studio

How to use alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit"

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": "alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit 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 "alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit"

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 alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit

Run Hermes

hermes

OpenClaw new

How to use alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit"

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 "alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

MLX LM

How to use alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

iMLX Qwen3 1.7B Current Tools 8-bit

Standalone 8-bit MLX model for the iMLX iOS app. This is a merged/fused model, not a runtime adapter, so iMLX can download and load it through the normal model catalog path.

Build Lineage

Base: mlx-community/Qwen3-1.7B-4bit
Fused selected iMLX current-tools adapter: qwen3-1.7b-imlx-v12-tool-capability-from-v3-r8-l8-step70
Dequantized to BF16 to preserve the adapter merge.
Applied and fused corrective all-layer LoRA checkpoint: qwen3-1.7b-imlx-v18-fused-hardening-r8-l28-step50
Quantized to 8-bit MLX with group size 64.

Local Verification

The final 8-bit standalone model was validated directly, without loading a runtime adapter:

16 / 17 passed (94%)

Verification run:

training/runs/20260509T174144Z

Known remaining failure:

web_off_weather

The model is intended for the iMLX curated model section and covers:

web_search
read_url
ocr_image_text
document_synthesize
current_datetime
calendar_brief
calendar_create
reminders_brief
reminders_create
timer_create
contacts_lookup

Downloads last month: 27

Safetensors

Model size

0.5B params

Tensor type

BF16

U32

MLX

Hardware compatibility

8-bit

Model tree for alan13367/iMLX-Qwen3-1.7B-Current-Tools-8bit

Base model

Qwen/Qwen3-1.7B-Base

Finetuned

Qwen/Qwen3-1.7B

Quantized

mlx-community/Qwen3-1.7B-4bit

Quantized

(5)

this model