Instructions to use Jundot/Qwen3.8-27B-oQ4e-mtp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Jundot/Qwen3.8-27B-oQ4e-mtp 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("Jundot/Qwen3.8-27B-oQ4e-mtp")
config = load_config("Jundot/Qwen3.8-27B-oQ4e-mtp")

# 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 Jundot/Qwen3.8-27B-oQ4e-mtp with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Jundot/Qwen3.8-27B-oQ4e-mtp"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Jundot/Qwen3.8-27B-oQ4e-mtp"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent

How to use Jundot/Qwen3.8-27B-oQ4e-mtp 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 "Jundot/Qwen3.8-27B-oQ4e-mtp"

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 Jundot/Qwen3.8-27B-oQ4e-mtp

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Jundot/Qwen3.8-27B-oQ4e-mtp with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Jundot/Qwen3.8-27B-oQ4e-mtp"

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 "Jundot/Qwen3.8-27B-oQ4e-mtp" \
  --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"

Qwen3.8-27B-oQ4e-mtp

Updated 2026-10-11: Re-quantized with Improved oQe Quantization. Please re-download if you have an earlier copy.

This model was quantized using oQ (oMLX v0.7.1) mixed-precision quantization with Improved oQe Quantization (jundot/omlx#4385).

A mixed-precision MLX checkpoint of Qwen/Qwen3.8-27B, a 27B dense vision-language model built on the Qwen3.5 architecture, with a hybrid layout of Gated DeltaNet and Gated Attention layers, image and video understanding and a 262,144-token native context. It includes the language model, the vision encoder and the multi-token prediction (MTP) head.

The effective average is 4.887 bits/weight for the complete checkpoint.

KL Divergence from the Original Model

Mean next-token KL divergence from the original BF16 model on the oMLX benchmark corpora (lower is better). Each corpus is scored on 16 windows of 2,048 tokens spread across the text (32,768 tokens).

Corpus	Previous oQ4e (2026-08-17)	Updated oQ4e	Change
Code (Python)	0.0143	0.0117	-18.1%
Novel (English)	0.0696	0.0600	-13.8%

Quantization and Bit Distribution

oQ4e allocates bits by measured layer sensitivity, then rounds each group with an importance matrix (128 samples x 512 tokens) and a weighted least-squares refit of its scale and bias.

Storage format	Logical weights	Share of total	Tensor storage	Effective bits/weight
Affine 4-bit, group 64	22.443B	80.784%	11.757 GiB	4.50
Affine 5-bit, group 64	4.823B	17.359%	3.088 GiB	5.50
BF16	0.516B	1.857%	0.961 GiB	16.00
Total	27.781B	100%	15.806 GiB	4.887

Storage Breakdown

Component	Logical weights	Storage
Language backbone	26.896B	15.736 GB / 14.655 GiB
Vision encoder	0.461B	0.921 GB / 0.858 GiB
MTP head	0.425B	0.314 GB / 0.293 GiB
Total	27.781B	16.971 GB / 15.806 GiB

License

Apache-2.0, following the source model. See LICENSE.

Downloads last month: 19,636

Safetensors

Model size

28B params

Tensor type

U32

BF16

MLX

Hardware compatibility

4-bit

Model tree for Jundot/Qwen3.8-27B-oQ4e-mtp

Base model

Qwen/Qwen3.8-27B

Quantized

(1594)

this model