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"

Performance on 5070Ti + 3090 (40GB VRAM in total)

#11

by palladium107 - opened 3 days ago

Discussion

palladium107

3 days ago

•

edited 3 days ago

I would be happy with a little more reserve on my cards, but this is my first trial with a little more context than "Hi"
With the below settings I get
input token generation ~180t/s
output token generation ~19t/s

Qwen3.8-Flash-Next-AD-4.27bpw-Q4_K_M-M64

llama.cpp is updated (daef7b6)

c = 196608
n-cpu-moe = 26
fa = on
tensor-split = 26/7
fit = off
n-gpu-layers = 99
temperature = 1
top-p = 0.95
top-k = 20
min-p=0.0
load-mode = mmap
batch-size = 2560
ubatch-size = 2560

I'm starting llamaccp like this:

export GGML_CUDA_ENABLE_UNIFIED_MEMORY=1

llama.cpp/build/bin/llama-server \
  --host 0.0.0.0 \
  --mlock \
  --models-preset ~/llama.cpp/config.ini \
  --models-max 1 \
  --timeout 1200 \
  --api-key xxx,yyy \
  --no-warmup \
  --no-mmproj \
  --reasoning-preserve

attention1

2 days ago

Hi@palladium107, thanks for sharing the metrics. I observed linear performance degradation on input and output speed (https://huggingface.co/AtomicChat/Qwen3.8-Flash-Next-GGUF/discussions/13) with the 4.27bpw variant on a Mac Studio. wondering if you saw the same on your setup.

drgerry2

1 day ago

Hi@palladium107, thanks for sharing the metrics. I observed linear performance degradation on input and output speed (https://huggingface.co/AtomicChat/Qwen3.8-Flash-Next-GGUF/discussions/13) with the 4.27bpw variant on a Mac Studio. wondering if you saw the same on your setup.

Running the 85GB quantization on a 9950X setup with 64GB DDR5 and dual GPUs (32GB total VRAM), I haven't experienced any performance degradation, maintaining a steady 13-15 t/s even at a 100k context length. However, the reasoning effort is exceptionally high. First impressions show better output quality compared to the 27B dense model, but at the cost of significantly lower speeds (60 t/s versus 13 t/s).
I'm using LMstudio and win11.

palladium107

about 21 hours ago

Hi@palladium107, thanks for sharing the metrics. I observed linear performance degradation on input and output speed (https://huggingface.co/AtomicChat/Qwen3.8-Flash-Next-GGUF/discussions/13) with the 4.27bpw variant on a Mac Studio. wondering if you saw the same on your setup.

Unfortunately I had some urgent code fixes yesterday, and I did not have time to test Flash Next any further, I just used Qwen3.8-27B, but today I will try Flash Next on some more complex features.

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