Instructions to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

Use Docker

docker model run hf.co/DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

Ollama
How to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with Ollama:
```
ollama run hf.co/DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M
```
Unsloth Desktop

How to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with Docker Model Runner:
```
docker model run hf.co/DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M
```

Lemonade

How to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

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 DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M

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 "DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

It starts very good but loops afterwards

#37

by 47Hunter47 - opened 3 days ago

Discussion

47Hunter47

3 days ago

Hi, here is my llama config:

-m /home/mert/Models/Qwen3.8-27B-TurboFCFusion-735-882-Here-Uncen-NEO-CODER-MAX-MTP-IQ4_XS.gguf
--mmproj /home/mert/Models/mmproj-BF16.gguf
-a qwen3.8-27
--host 0.0.0.0 --port 8080
-ngl 99 -c 150000
--cache-type-k q4_0 --cache-type-v q8_0
--flash-attn on
-b 2048 -ub 1024
-np 1
--cache-ram 4096
--temp 0.6 --min-p 0.05 --presence-penalty 0.0 --repeat-penalty 1.07 --top-p 0.95 --top-k 20
--spec-type draft-mtp
--spec-draft-n-max 3
--load-mode mlock
--metrics
--perf
--api-key-file /home/mert/.llama-api-keys
--timeout 600 \

My problem is, when it is at the beginning of conversation, it works fine, but when it starts to solve complex issues (requiring several tools calls and searches etc).
1- it starts to loop thinking and keeps repeating. (i tried to increase --repeat-penalty from 1 to 1.1, didn't resolve the issue)
2- it stops before finishing the job (like it says "i'll start doing this...", "i will have to start doing that..." but stops and i have to nudge/push it with prompts like "start" , "keep going"..."

I even used "/goal" but it still stops at some points.
I use hermes agent

With the low and without low variant, the issue persists. (I have a rtx3090 , so i could not try bigger quantizations because i don't prefer to offload to cpu)

Other than that, the model is smart and especially the less thinking is perfect for speed but these problems became annoying so i had to revert to base model (Qwen3.8-27B-UD-Q4_K_M.gguf).

Looks like I have to wait for the updated (stage2)

DavidAU

Owner 3 days ago

•

edited 3 days ago

Hey;

Try: rep pen 1 [off] (qwens will loop with it on) ; temp .7 ; and try quant IQ4XS LOW and/or Q6 LOW.
Adjust top_k: 64
You may need to adjust caching [or remove it] ; this might be too high/too much loss for IQ4XS.
Maybe Q4KM would be better fit.

NOTE: tool calls really need Q5 (or Q6 is better) for best performance [Qwens generally speaking, all types of tool calling].

V2 ; releasing on/about Sept 10 ; may address some of these issues.

47Hunter47

3 days ago

I'll wait for september 10th just to be on the safe side 😃 (btw, Qwen3.8-27B-UD-Q4_K_M.gguf works fine) I believe these issues will be gone on V2 hopefully.

Warsin

3 days ago

my q6 kinda goes insanse too

47Hunter47

2 days ago

i couldn't test q6 because i only have 24gb vram but i think it is best to wait for v2 for the time being

tlilancalqui

about 22 hours ago

Was hoping that the Q6 LOW would work for programming, however it gets stuck after looping. I wonder if using an imatrix and something like the ymq-compiler (https://github.com/minyor/ymq-compiler) would help with making this LOW version - or just make your BF16 public. zerodigest right now makes the best programming Qwen3.8 at 19.7 GB (zerodigest/Qwen3.8-27B-YMQ-MTP-GGUF/ Qwen3.8-27B-YMQ-XL.gguf)

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment