Instructions to use hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-mtp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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("hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-mtp")
config = load_config("hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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 hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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 "hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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": "hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-mtp"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent

How to use hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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 "hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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 hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-mtp

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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 "hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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 "hakunokina/Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-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 Fable-5 Coding Distilled — oQ6e-mtp (MLX, Apple Silicon)

MLX build of khazarai/Qwen3.8-27B-Fable-5-Coding-Distilled, quantized from the BF16 source with oMLX oQ6e (imatrix-calibrated 6-bit), MTP head kept. A coding/agentic fine-tune of Qwen3.8-27B trained on 195M tokens of distilled debugging traces; until now it had no MLX/oQ build.

What this build is

Model: 27B dense Qwen3.8 (Qwen3.5-family architecture: hybrid Gated DeltaNet + attention, 64 layers), fine-tuned by khazarai for grounded, tool-using coding and debugging. 262k context.
Quantization: oMLX oQ6e — 6-bit affine (group 64), imatrix-calibrated on oMLX's code + multilingual set.
MTP head kept (-mtp): the source's multi-token-prediction layer is preserved, so oMLX can use it for speculative decoding.
Vision encoder kept from the source (image input not tested in this build).
Size: ~~22 GB (source BF16: ~54 GB). Other sizes: oQ8e-mtp (~~28 GB), oQ4e-mtp (~16 GB).
Runtime: requires oMLX ≥ 0.7.1. Not tested with plain mlx-lm / LM Studio.

Quality vs. BF16 source

Full-vocabulary KL against the BF16 source, same token streams; 64 sequences × 512 tokens (32.7k scored tokens): code, English prose, Turkish.

Metric	Value
KL(BF16 ‖ oQ6e), mean	0.0041 nat/token
KL median / p99	0.0016 / 0.049
Top-1 token agreement	97.2 %
Perplexity BF16 → oQ6e	6.078 → 6.091

No benchmark suite was run on this build; quality is reported as divergence from the source.

Speed (Mac Studio M3 Ultra, 512 GB, oMLX 0.7.1)


Prefill, ~29k-token prompt (cold)	~~74 s (~~390 tok/s)
Same prefix again (prefix cache)	~1.41 s
Decode (MTP speculative decoding)	~48 tok/s
Tool calling (Anthropic `/v1/messages` / OpenAI `/v1/chat/completions`, 5 runs each)	5/5 / 4/5

Single-run measurements; numbers vary with context length, content and settings (MTP gains depend on how predictable the output is).

Usage (oMLX)

Place the folder in your oMLX model directory and restart oMLX; it is served on both the OpenAI and Anthropic APIs:

curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model": "Qwen3.8-27B-Fable-5-Coding-Distilled-oQ6e-mtp", "messages": [{"role": "user", "content": "Write a Python function that parses a CSV line."}]}'

Memory: ~~22 GB weights plus KV cache (~~64 KB/token on the 16 full-attention layers → ~16 GB at 262k).

Credits & license

Base model: Qwen team — Qwen/Qwen3.8-27B.
Coding distillation fine-tune: khazarai — khazarai/Qwen3.8-27B-Fable-5-Coding-Distilled (dataset DSFFGFG456/fable-5-coding-and-debugging-traces).
Quantization: oMLX oQ.
License: Apache-2.0, inherited from upstream. The upstream card marks its license and training details as still to be confirmed; check the source repo before relying on them.