Instructions to use aquaduck/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 aquaduck/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 aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL
# Run inference directly in the terminal:
llama cli -hf aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL
# Run inference directly in the terminal:
llama cli -hf aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL

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 aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL
# Run inference directly in the terminal:
./llama-cli -hf aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL

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 aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL
# Run inference directly in the terminal:
./build/bin/llama-cli -hf aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL

Use Docker

docker model run hf.co/aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL

LM Studio
Jan

vLLM

How to use aquaduck/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 "aquaduck/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": "aquaduck/Qwen3.8-Flash-Next-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL

Ollama
How to use aquaduck/Qwen3.8-Flash-Next-GGUF with Ollama:
```
ollama run hf.co/aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL
```
Unsloth Desktop

How to use aquaduck/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 aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_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": "aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use aquaduck/Qwen3.8-Flash-Next-GGUF with Docker Model Runner:
```
docker model run hf.co/aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL
```

Lemonade

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

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL

Run and chat with the model

lemonade run user.Qwen3.8-Flash-Next-GGUF-UD-Q2_K_XL

List all available models

lemonade list

Hermes Agent

How to use aquaduck/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 aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_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 aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_K_XL

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use aquaduck/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 aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_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 "aquaduck/Qwen3.8-Flash-Next-GGUF:UD-Q2_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"

Model Card for aquaduck/Qwen3.8-Flash-Next-GGUF

Pinned UD-Q3_K_XL GGUF of Qwen3.8-Flash-Next (qwen/qwen3.8-flash-next), plus midpoint layer shards for staged / multi-node loading (Aquaduck Arc layer-package-v1). The shard files are not a new quantization. They are contiguous midpoint packages cut from the full UD-Q3_K_XL GGUF in this repo.

Model lineage

qwen/qwen3.8-flash-next └── UD-Q3_K_XL GGUF + midpoint shards → aquaduck/Qwen3.8-Flash-Next-GGUF (this repo)

Base weights: https://huggingface.co/qwen/qwen3.8-flash-next (other)
Quantization source: https://huggingface.co/aquaduck/Qwen3.8-Flash-Next-GGUF (tag UD-Q3_K_XL)
This repo: full UD-Q3_K_XL GGUF (Qwen3.8-Flash-Next-UD-Q2_K_XL-00001-of-00003.gguf) and midpoint GGUF shards

Model Details


Catalog id	`qwen/qwen3.8-flash-next`
Quantization	UD-Q3_K_XL
Parameters	180B
Native context	262,144 tokens
License	other
Base model	qwen/qwen3.8-flash-next
Ingest GGUF	aquaduck/Qwen3.8-Flash-Next-GGUF

Model Description

Hosted by: Aquaduck (hosting and layer packaging only; base model by Qwen Team / Alibaba Cloud; GGUF quant by Unsloth / llama.cpp ecosystem)
Shared by: Aquaduck AI
Model type: Causal language model (Qwen3.8-Flash-Next), GGUF UD-Q3_K_XL
Language(s): Multilingual (same as base)
License: other (inherits from qwen/qwen3.8-flash-next)
Finetuned from model: N/A — not a fine-tune
Derived from: aquaduck/Qwen3.8-Flash-Next-GGUF ← qwen/qwen3.8-flash-next

Model Sources

Base model card: https://huggingface.co/qwen/qwen3.8-flash-next
Quantized GGUF source: https://huggingface.co/aquaduck/Qwen3.8-Flash-Next-GGUF

Files

File	Role	Approx. size
`Qwen3.8-Flash-Next-UD-Q2_K_XL-00001-of-00003.gguf`	Full-model GGUF (UD-Q3_K_XL)	~78.87 GB
`Qwen3.8-Flash-Next-UD-Q2_K_XL-layers-0-10.gguf`	Split shard (layers 0–9)	~39.67 GB
`Qwen3.8-Flash-Next-UD-Q2_K_XL-layers-10-48.gguf`	Split shard (layers 10–47)	~39.65 GB

Total layers: 48
Valid split boundaries: 10 Filenames use exclusive end indices (layers-{start}-{endExclusive}).

Uses

Direct Use

Full Qwen3.8-Flash-Next-UD-Q2_K_XL-00001-of-00003.gguf: standard single-file UD-Q3_K_XL GGUF (llama.cpp-compatible). Use this for single-node / local runs.
*-layers-*.gguf: Aquaduck / Arc staged loading only. These are not drop-in complete models for stock llama.cpp. Use the base model’s chat template (including thinking / instruct modes as documented on the base model card); other formats will not work correctly.

Out-of-Scope Use

Expecting any one shard to run as a complete model
Treating this repo as a new training run or re-quant
Uses prohibited by the other license or the base model’s model card guidance

Bias, Risks, and Limitations

Same capabilities, biases, and risks as qwen/qwen3.8-flash-next. UD-Q3_K_XL quantization can degrade quality vs. the original higher-precision releases. Layer sharding does not change weights beyond packaging.

Recommendations

Follow the base model’s docs for chat template, thinking vs instruct modes, and sampling. Prefer Qwen3.8-Flash-Next-UD-Q2_K_XL-00001-of-00003.gguf in this repo when you do not need staged loading.

How to Get Started

These files are meant to be loaded automatically by the Aquaduck desktop app.

Download the Aquaduck desktop app and sign in.
Devices connected to the internet will receive a model assignment from the model catalog (qwen/qwen3.8-flash-next).
Download the model from the Home view. The app will:
- download only the assigned file from this repo (full Qwen3.8-Flash-Next-UD-Q2_K_XL-00001-of-00003.gguf or one midpoint half)
- keep that stage ready for serving You do not need to pick files by hand, but you may for local serving. Assignment and download are driven by model catalog metadata. The full Qwen3.8-Flash-Next-UD-Q2_K_XL-00001-of-00003.gguf is a standard UD-Q3_K_XL GGUF. The *-layers-*.gguf files are not.

Training Details

No training. Weights come from Qwen Team / Alibaba Cloud; UD-Q3_K_XL GGUF from aquaduck/Qwen3.8-Flash-Next-GGUF; this repo hosts that GGUF and (when split) packages it into midpoint layer shards.

Evaluation

No separate evals for the hosted GGUF or shards. See qwen/qwen3.8-flash-next.

Technical Specifications

Architecture: Qwen3.8-Flash-Next (~180B params, GQA (24 Q / 2 KV heads), 48 layers, hidden dim 2560)
Quantization: UD-Q3_K_XL
Packaging: pinned full UD-Q3_K_XL GGUF; optional Arc midpoint shards (*-layers-{start}-{endExclusive}.gguf)
Package format: layer-package-v1
Split: 2 stages at layer 10 (maxStages: 2)

Citation

@misc{qwen38flashnext,
    title  = {Qwen3.8-Flash-Next},
    author = {Qwen Team / Alibaba Cloud},
    year   = {2026},
    url    = {https://huggingface.co/qwen/qwen3.8-flash-next}
}

Credit:

The GGUF quantization source (https://huggingface.co/aquaduck/Qwen3.8-Flash-Next-GGUF)
llama.cpp (https://github.com/ggml-org/llama.cpp) for GGUF support

Attribution

Quantized GGUF ingested from aquaduck/Qwen3.8-Flash-Next-GGUF. Original weights: qwen/qwen3.8-flash-next. Redistributed under the base model's license.

Hosted by Aquaduck.

Model Card Contact

Aquaduck AI — https://huggingface.co/aquaduck

Downloads last month: 460

GGUF

Model size

78B params

Architecture

qwen4exp

Hardware compatibility

2-bit