Instructions to use AtomicChat/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 AtomicChat/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 AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf AtomicChat/Qwen3.8-Flash-Next-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 AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf AtomicChat/Qwen3.8-Flash-Next-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 AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf AtomicChat/Qwen3.8-Flash-Next-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 AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M

Use Docker

docker model run hf.co/AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M

LM Studio
Jan

vLLM

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

Use Docker

docker model run hf.co/AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M

Ollama
How to use AtomicChat/Qwen3.8-Flash-Next-GGUF with Ollama:
```
ollama run hf.co/AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M
```
Unsloth Desktop

How to use AtomicChat/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 AtomicChat/Qwen3.8-Flash-Next-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": "AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use AtomicChat/Qwen3.8-Flash-Next-GGUF with Docker Model Runner:
```
docker model run hf.co/AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M
```

Lemonade

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

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.8-Flash-Next-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use AtomicChat/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 AtomicChat/Qwen3.8-Flash-Next-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 AtomicChat/Qwen3.8-Flash-Next-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use AtomicChat/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 AtomicChat/Qwen3.8-Flash-Next-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 "AtomicChat/Qwen3.8-Flash-Next-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"

Request for a Faster UD-Q4_K_XL Quantization of Qwen3.8-Flash-Next

#14

by NamerPRO - opened 13 days ago

Discussion

NamerPRO

13 days ago

Hi AtomicChat team!

I’m currently using your AD-4.27bpw-Q4_K_M-M64 quantization of Qwen3.8-Flash-Next on my:

Intel Core i9-14900HX
RTX 4070 Laptop GPU (8 GB VRAM)
64 GB DDR5 RAM

It runs at around 16 tokens/sec, which is excellent for my hardware.

I noticed that other quantization approaches, such as UD-Q4_K_XL, may potentially provide better quality at a similar size.

Would it be possible for AtomicChat to create a UD-Q4_K_XL build of Qwen3.8-Flash-Next that retains the quality advantages of UD quantization while achieving performance close to your current AD-4.27bpw-Q4_K_M-M64 (~16 tok/s)?

If this is technically possible, I’d really appreciate such a build.

Ideally, it would also be great if you could provide the same evaluation metrics you already provide for your current builds:

In memory
On SSD
Total size
Mean KLD
Same top-1 %
PPL ratio

My goal is simply to get the best possible quality while keeping the ~16 tok/s performance of your current Q4_K_M build.

Thanks for your work on these quantizations!

dandandelion

10 days ago

UD-Q4_K_XL is approaching 5-bit average and it bumps up most tensors, not just the n-gram table, which means it won't even be close to size in terms of VRAM|RAM resident + SSD resident. And with only 8GiB+64GiB setup, you'd be short in memory, heck I can barely run Unsloth's UD-IQ4_XS in my 16GiB+64GB (just around ~6GB free for some stuff to do) so it's a hopeless wish unless you add more RAM to your system.

Ponyo61

6 days ago

Hi AtomicChat team!

I’m currently using your AD-4.27bpw-Q4_K_M-M64 quantization of Qwen3.8-Flash-Next on my:

Intel Core i9-14900HX

RTX 4070 Laptop GPU (8 GB VRAM)

64 GB DDR5 RAM

It runs at around 16 tokens/sec, which is excellent for my hardware.

I noticed that other quantization approaches, such as UD-Q4_K_XL, may potentially provide better quality at a similar size.

Would it be possible for AtomicChat to create a UD-Q4_K_XL build of Qwen3.8-Flash-Next that retains the quality advantages of UD quantization while achieving performance close to your current AD-4.27bpw-Q4_K_M-M64 (~16 tok/s)?

If this is technically possible, I’d really appreciate such a build.

Ideally, it would also be great if you could provide the same evaluation metrics you already provide for your current builds:

In memory

On SSD

Total size

Mean KLD

Same top-1 %

PPL ratio

My goal is simply to get the best possible quality while keeping the ~16 tok/s performance of your current Q4_K_M build.

Thanks for your work on these quantizations!

how did you get 16 t/s with 8 gb vram/64 gb ram?with which llama.cpp version, context size and which flags?

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