Instructions to use Shiftedx/Qwen3.8-27B-MLX-MXFP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Shiftedx/Qwen3.8-27B-MLX-MXFP8 with MLX:

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

from mlx_vlm import load, generate
from mlx_vlm.prompt_utils import apply_chat_template
from mlx_vlm.utils import load_config

# Load the model
model, processor = load("Shiftedx/Qwen3.8-27B-MLX-MXFP8")
config = load_config("Shiftedx/Qwen3.8-27B-MLX-MXFP8")

# Prepare input
image = ["http://images.cocodataset.org/val2017/000000039769.jpg"]
prompt = "Describe this image."

# Apply chat template
formatted_prompt = apply_chat_template(
    processor, config, prompt, num_images=1
)

# Generate output
output = generate(model, processor, formatted_prompt, image)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use Shiftedx/Qwen3.8-27B-MLX-MXFP8 with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Shiftedx/Qwen3.8-27B-MLX-MXFP8"

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": "Shiftedx/Qwen3.8-27B-MLX-MXFP8"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use Shiftedx/Qwen3.8-27B-MLX-MXFP8 with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Shiftedx/Qwen3.8-27B-MLX-MXFP8"

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 "Shiftedx/Qwen3.8-27B-MLX-MXFP8" \
  --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"

Hermes Agent

How to use Shiftedx/Qwen3.8-27B-MLX-MXFP8 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 "Shiftedx/Qwen3.8-27B-MLX-MXFP8"

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 Shiftedx/Qwen3.8-27B-MLX-MXFP8

Run Hermes

hermes

Atomic Chat

Qwen3.8-27B MLX MXFP8

Vision-enabled MLX conversion of Qwen/Qwen3.8-27B, pinned to revision 1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0.

Language model: MXFP8, 8-bit, group size 32
Vision tower: same-revision BF16 weights
Runtime: mlx-vlm 0.6.3 or newer

The Hub's approximately 8.0B safetensors count reflects packed MXFP storage elements; the underlying architecture remains the full 27B model.

Use

python -m mlx_vlm.generate \
  --model Shiftedx/Qwen3.8-27B-MLX-MXFP8 \
  --image image.jpg \
  --prompt "Describe this image."

Text generation and three image-understanding smoke tests passed locally. Quantization can still change behavior, so independently evaluate important use cases. Treat prompts, images, and outputs as untrusted: do not submit secrets, and sandbox tools or generated code with least-privilege access. This conversion adds no telemetry or remote execution.

The upstream Apache-2.0 license and model limitations continue to apply.

Downloads last month: -

Safetensors

Model size

8B params

Tensor type

U32

BF16

MLX

Hardware compatibility

8-bit

Model tree for Shiftedx/Qwen3.8-27B-MLX-MXFP8

Base model

Qwen/Qwen3.8-27B

Quantized

(282)

this model

Collection including Shiftedx/Qwen3.8-27B-MLX-MXFP8

Qwen3.8-27B MLX and MTPLX

Collection

Text-only Qwen3.8-27B MLX conversions: MXFP4, MXFP8, and experimental MXFP4 with native MTP for MTPLX. Vision is excluded. • 3 items • Updated about 14 hours ago