Instructions to use lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx 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("lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx")
config = load_config("lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx")

# 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 lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx"

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": "lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent

How to use lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx 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 "lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx"

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 lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx"

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 "lemuralabs/Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx" \
  --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"

Qwen3.6-27B-V2-abliterated-uncensored-6-bit-mlx / DFLASH_SPECULATIVE_DECODING.md

rogKesavan90

Initial commit

115187f 21 days ago

preview code

Raw

History Blame Contribute Delete

1.86 kB

DFlash speculative decoding for this model (MLX)

Note: no extra weights are needed in this repo — DFlash works with the existing quant as-is, plus an external drafter. This file is a usage guide.

DFlash is a block-diffusion speculative drafter trained for Qwen3.6-27B targets. It drafts a 16-token block by diffusion and lets this model verify it autoregressively — so output is lossless (identical to plain decoding), just faster.

Quick start

pip install -U mlx_vlm
# one-time: accept the gated drafter at https://huggingface.co/z-lab/Qwen3.6-27B-DFlash
python3 -m mlx_vlm generate \
  --model osmapi/osmQwopus-3.6-27B-V2-heretic-abliterated-uncensored-6-bit-mlx \
  --draft-model z-lab/Qwen3.6-27B-DFlash --draft-kind dflash \
  --prompt "<your prompt>" --max-tokens 256

Serve it (OpenAI-compatible):

python3 -m mlx_vlm server \
  --model osmapi/osmQwopus-3.6-27B-V2-heretic-abliterated-uncensored-6-bit-mlx \
  --draft-model z-lab/Qwen3.6-27B-DFlash --draft-kind dflash

Measured on an Apple M4 Max (128 GB)

Quant	AR baseline	+ DFlash	Speedup
8-bit (affine)	16.4 tok/s	55.3 tok/s	3.38×
bf16 (full)	9.0 tok/s	33.2 tok/s	3.67×

Acceptance ≈ 8.95 tokens/round; +~3.9 GB peak memory for the drafter; small TTFT increase.

Notes & limits

Text path only — the vision tower is not accelerated.
Speedup is workload-dependent (acceptance varies by prompt).
Larger on more memory-bound quants (bf16 > 8-bit).

Background & full benchmarks: [https://huggingface.co/blog/junafinity/block-diffusion-on-apple-silicon-with-3-7x-speedup]

Credit: DFlash by z-lab (arXiv:2602.06036); runtime mlx_vlm.