Instructions to use unsloth/Qwen3.8-Flash-Next-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 unsloth/Qwen3.8-Flash-Next-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 unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
llama cli -hf unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
llama cli -hf unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
./build/bin/llama cli -hf unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

Use Docker

docker model run hf.co/unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

Llama new
LM Studio
Jan

vLLM

How to use unsloth/Qwen3.8-Flash-Next-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "unsloth/Qwen3.8-Flash-Next-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": "unsloth/Qwen3.8-Flash-Next-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/unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

Ollama
How to use unsloth/Qwen3.8-Flash-Next-GGUF with Ollama:
```
ollama run hf.co/unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL
```
Unsloth Desktop

How to use unsloth/Qwen3.8-Flash-Next-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

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": "unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use unsloth/Qwen3.8-Flash-Next-GGUF with Docker Model Runner:
```
docker model run hf.co/unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL
```

Lemonade

How to use unsloth/Qwen3.8-Flash-Next-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

Run and chat with the model

lemonade run user.Qwen3.8-Flash-Next-GGUF-UD-Q4_K_XL

List all available models

lemonade list

Hermes Agent

How to use unsloth/Qwen3.8-Flash-Next-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 unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

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 unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use unsloth/Qwen3.8-Flash-Next-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL

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 "unsloth/Qwen3.8-Flash-Next-GGUF:UD-Q4_K_XL" \
  --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"

context type MTP requested but model doesn't contain MTP layers

#47

by Manhandle6292 - opened Aug 30

Discussion

Manhandle6292

Aug 30

2.39.009.670 I common_speculative_init_result: creating MTP draft context against the target model 'Qwen3.8-Flash-Next-UD-Q4_K_XL-00001-of-00004.gguf'

2.39.009.677 W llama_init_from_model: context type MTP requested but model doesn't contain MTP layers

s1arsky

Aug 31

•

edited Aug 31

same issue here, I am testing with allegedly? build-in DFlash instead : --spec-type draft-dflash . But cant confirm it works at all for now.

pmmcp

Aug 31

Yes, they intentionally removed MTP on these GGUFs. Not sure why, but I saw there are still some bugs with MTP support on this model, so that's probably the reason. If you want it now, there are some GGUF drafters out there that you can download and add manually.

MrScarySpaceCat

Sep 1

Looks like they just added them a few hours ago.

Navigate2053

Sep 3

•

edited Sep 3

Yesterday I was able to get MTP working.
hf download hf://unsloth/Qwen3.8-Flash-Next-GGUF/MTP/mtp-Qwen3.8-Flash-Next-shared-Q8_0.gguf
(here are some llama.cpp switches)
app-b10715-mix-86bd2d3-windows-x64-cuda13-portable\llama-server.exe --ctx-size 262144 -hf unsloth/Qwen3.8-Flash-Next-GGUF:Q8_0 --temp 1.0 --top-p 0.95 --top-k 20 --min-p 0.0 --presence-penalty 0.0 --repeat-penalty 1.0 --reasoning on -np 1 --cache-type-k q8_0 --cache-type-v q8_0 --fit on --spec-type draft-mtp --spec-draft-n-max 3 -md %userprofile%.cache\huggingface\hub\models--unsloth--Qwen3.8-Flash-Next-GGUF\snapshots\38bb39ee97821de2c9009abb7e93950eec396e66\MTP\mtp-Qwen3.8-Flash-Next-shared-Q8_0.gguf

I tried using --hf-repo-draft instead of the specific downloaded model, but was unable to specify anything that would work.
With -lv 4 specified I do see these two error lines during the loading process:
E llama_model_load: error loading model: borrow_shared_tensor: this model is a draft head without its own 'token_embd.weight'; load it as a draft of its target model, not on its own
E llama_model_load_from_file_impl: failed to load model
but otherwise it loads correctly and seems to benefit from the speculative decoding.

Using MTP with Q8 has taken me from ~40-49 token/s without MTP to ~55-75 token/s on my hardware setup.

it seems like --fit on is doing a fair amount of heavy lifting. when I change to --fit off it fails to load.

LaurentPayot

Sep 14

•

edited Sep 14

Same error messages as @Navigate2053 with my Strix Halo on Ubuntu 26.04, Vulkan or ROCm.

Navigate2053

Sep 14

•

edited Sep 14

@LaurentPayot
I switched back to this model today. In order to get going I had to
hf download hf://unsloth/Qwen3.8-Flash-Next-GGUF/MTP/mtp-Qwen3.8-Flash-Next-shared-Q8_0.gguf

app-b10909-mix-bea84f7-windows-x64-cuda13-portable\llama-server.exe [...] -hf unsloth/Qwen3.8-Flash-Next-GGUF:Q8_0 --temp 1.0 --top-p 0.95 --top-k 20 --min-p 0.0 --presence-penalty 0.0 --repeat-penalty 1.0 --reasoning on -lv 4 --seed 7407 -np 1 --cache-type-k q8_0 --cache-type-v q8_0 --fit on --spec-type draft-mtp --spec-draft-n-max 5 -md %userprofile%.cache\huggingface\hub\models--unsloth--Qwen3.8-Flash-Next-GGUF\MTP\mtp-Qwen3.8-Flash-Next-shared-Q8_0.gguf

I haven't really settled on a a specific --spec-draft-n-max value yet...
I was unable to use -hfd to specify the repo for the mtp gguf

with the extra logging i do get these errors but it continues loading anyway and appears to be working just fine....

E llama_model_load: error loading model: borrow_shared_tensor: this model is a draft head without its own 'token_embd.weight'; load it as a draft of its target model, not on its own
E llama_model_load_from_file_impl: failed to load model
W operator (): failed to measure the memory of the extra model, fitting without it: failed to load model

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