Instructions to use pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-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 pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-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 pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16
# Run inference directly in the terminal:
llama cli -hf pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16
# Run inference directly in the terminal:
llama cli -hf pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

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 pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16
# Run inference directly in the terminal:
./llama-cli -hf pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

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 pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

Use Docker

docker model run hf.co/pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

LM Studio
Jan

vLLM

How to use pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-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": "pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-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/pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

Ollama
How to use pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF with Ollama:
```
ollama run hf.co/pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16
```
Unsloth Desktop

How to use pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

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": "pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF with Docker Model Runner:
```
docker model run hf.co/pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16
```

Lemonade

How to use pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

Run and chat with the model

lemonade run user.Qwen3.8-Flash-Next-GSQ-RCO-GGUF-BF16

List all available models

lemonade list

Hermes Agent

How to use pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-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 pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

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 pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16

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 "pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:BF16" \
  --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"

TL;DR: Run Qwen's 180B model on 2× RTX 3090s, with embeddings on disk and the KV cache in RAM. With BF16 embeddings, it scores 2.85 points above the base model on MMLU-Pro, with 1.72% higher perplexity.

Qwen3.8-Flash-Next · GSQ-RCO GGUFs

Non-uniform GGUF quantization produced with GSQ and RCO, with a vision projector for multimodal use.

Independent community reproduction. These files were produced by a third party using the published GSQ and RCO methods. They are not an IST-DASLab release and carry no endorsement from the authors of either paper.

Overview

This repository provides a GGUF quantization of Qwen/Qwen3.8-Flash-Next at a 3.5-bit target, together with the model's BF16 vision projector (mmproj) for multimodal use. Each weight tensor is assigned its own quantization type by a search that allocates precision by per-tensor sensitivity under an exact total size budget.

This is a reduced-budget adaptation of GSQ/RCO to native GGUF formats. The model uses a separate shard for the token and n-gram embeddings. Choose the BF16 shard or the smaller Q4_0/Q8_0 shard.

Method	Description
GSQ (Gumbel-Softmax Quantization, paper, code)	Post-training scalar quantization that jointly learns per-coordinate grid assignments and per-group scales through a Gumbel-Softmax relaxation.
RCO (Riemannian Constrained Optimization, paper, code)	Assigns one of K quantization types to each of N tensors under an exact total size budget, reformulated as a smooth Riemannian manifold in logit space.

Both methods were developed at the Deep Algorithms and Systems Lab (DASLab), Institute of Science and Technology Austria.

Read this before using these files

Download the main weights and one embedding shard. The 3.5-bit label is the quantization target for the main weights. File identities and the BF16 shard layout are recorded in manifest.json; the quantized alternative is recorded in embeddings-Q4_0.json.

Available files

File	bpw	Size	Notes
`Qwen3.8-Flash-Next-GSQ-RCO-3.5bit.gguf`	3.5 target	47.94 GB	Main weights; requires embedding shard
`Qwen3.8-Flash-Next-ngram-embeddings.gguf`	16	103.68 GB	BF16 token and n-gram embeddings
`Qwen3.8-Flash-Next-ngram-embeddings-Q4_0.gguf`	4.5 / 8.5	29.49 GB	Q4_0 n-gram table + Q8_0 token embeddings
`Qwen3.8-Flash-Next-mmproj-BF16.gguf`	16.17	0.91 GB	Vision encoder + projector

MMLU-Pro

The evaluations below use the BF16 embedding shard.

No reasoning; 2,048-token context limit. Scores use the log probabilities of single-token answers A through J, with zero-shot prompts and no chat template. The 2,000 questions use a fixed seed and are stratified across all 14 categories. Quant and BF16 use matching layer placement. Scorer, question IDs, per-question predictions and runtime settings are in the evaluation package.

Build	Correct	Accuracy	SE
GSQ-RCO 3.5-bit	1,165/2,000	58.25%	1.10 pp
BF16 reference	1,108/2,000	55.40%	1.11 pp

The paired difference is +2.85 percentage points (226 quant-only / 169 BF16-only correct; exact two-sided McNemar p = 0.0047752).

Results

Perplexity: eight 1,024-token contexts, 4,088 scored tokens, document-disjoint held-out text. Greedy decoding; IFEval answered directly; GSM8K with xhigh thinking; 1,024/2,048-token caps.

Variant	Native PPL ↓	Approx. KL/token ↓	IFEval strict	IFEval completed correct	GSM8K completed correct
BF16 reference	3.0533	—	13/16	13/16	8/8
GSQ-RCO	3.1058	0.111225	13/16	12/16	8/8

KL is approximate, computed from the native uint16 reference cache. One quant IFEval response was truncated.

Raw results · Reproducibility

Usage

Runtime

llama.cpp with Qwen4Exp and Q2_0 support, pinned here to upstream f3f1a8f. The runtime and numerical settings used for evaluation are documented separately in the reproduction instructions.

git clone https://github.com/ggml-org/llama.cpp
cd llama.cpp
git checkout f3f1a8f2760f28325a5ec20c05b171e5b7c83a29
cmake -S . -B build -DGGML_CUDA=ON -DCMAKE_BUILD_TYPE=Release
cmake --build build --target llama-server -j

Text and vision

For quantized embeddings, replace Qwen3.8-Flash-Next-ngram-embeddings.gguf with Qwen3.8-Flash-Next-ngram-embeddings-Q4_0.gguf in the download and symlink commands below. The main weights and quantized embeddings total 77.42 GB.

hf download pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF \
  Qwen3.8-Flash-Next-GSQ-RCO-3.5bit.gguf \
  Qwen3.8-Flash-Next-ngram-embeddings.gguf \
  Qwen3.8-Flash-Next-mmproj-BF16.gguf --local-dir .

ln -s Qwen3.8-Flash-Next-GSQ-RCO-3.5bit.gguf qwen3.8-flash-next-00001-of-00002.gguf
ln -s Qwen3.8-Flash-Next-ngram-embeddings.gguf qwen3.8-flash-next-00002-of-00002.gguf

llama-server -m qwen3.8-flash-next-00001-of-00002.gguf \
  --mmproj Qwen3.8-Flash-Next-mmproj-BF16.gguf \
  --override-tensor '^per_layer_token_embd\.weight$=CPU,^token_embd\.weight$=CPU' \
  --load-mode mmap -ngl all -c 4096 -np 1 --jinja

The numbered symlinks let llama.cpp discover both shards. The tensor overrides keep both embedding matrices on CPU, with mmap allowing their pages to be backed by disk.

For two 24 GB cards such as RTX 3090s, balance the weights with --tensor-split and leave headroom for the KV cache and runtime buffers. --no-kv-offload places the KV cache in system RAM when needed.

Quantization procedure

Per-tensor candidates. GSQ trained routed expert projections for 8 steps on 64 training and 4 validation sequences of 1024 tokens; rare or unimproved experts keep their initializer.
RCO search. 8 steps, 2 Gumbel samples, 8 training and 4 validation sequences, selected by validation teacher KL. No step improved on step 0, so the initial allocation was kept.
Assembly. Exact packed bytes, metadata and alignment charged; payload roundtrip verified.

Gate/up candidates Q2_0/Q2_K/Q3_K/Q4_K; down candidates Q2_0/Q4_0 (640-column rows do not fit 256-value K blocks); compatible non-expert matrices Q8_0. The BF16 embedding option retains the source bytes. The Q4_0/Q8_0 option copies the packed embedding tensors from the original quantized release without requantization.

Source revision: de4b8e4d43b917e7706784d8bb445c9af86a3540.

Citation

If you use these files, please cite this release together with the base model and both methods.

This release

@misc{q38fngsqrco2026,
  title        = {Qwen3.8-Flash-Next GSQ-RCO GGUF quantization},
  author       = {Josephine Pfeiffer},
  year         = {2026},
  publisher    = {Hugging Face},
  doi          = {10.57967/hf/10398},
  howpublished = {\url{https://huggingface.co/pfeifferj/Qwen3.8-Flash-Next-GSQ-RCO-GGUF}}
}

Base model

@techreport{qwen2026design,
    title       = {On the Design of {Qwen3.8-Next} Architecture: Evaluation, Efficiency, and Training Stability},
    author      = {{Qwen Team}},
    institution = {Alibaba Group},
    month       = {August},
    year        = {2026}
}

@misc{qwen3.8flashnext,
    title  = {{Qwen3.8-Flash-Next}: A New Architecture, Towards Ultimate Cost-Efficiency},
    author = {{Qwen Team}},
    month  = {August},
    year   = {2026},
    url    = {https://qwen.ai/blog?id=qwen3.8-flash-next}
}

Methods

@article{gsq2026,
  title  = {GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling},
  author = {Dadgarnia, Alireza and Tabesh, Soroush and Nikdan, Mahdi and Helcig, Michael and Kurtic, Eldar and Kleinegger, Maximilian and Alistarh, Dan},
  journal= {arXiv preprint arXiv:2604.18556},
  year   = {2026}
}
@article{rco2026,
  title  = {Model Compression with Exact Budget Constraints via Riemannian Manifolds},
  author = {Helcig, Michael and Alistarh, Dan},
  journal= {arXiv preprint arXiv:2605.00649},
  year   = {2026}
}

Acknowledgements

Huge kudos to the Deep Algorithms and Systems Lab (DASLab) at the Institute of Science and Technology Austria for developing GSQ and RCO and for releasing the papers and reference implementations publicly.

Thanks also to ella for giving me the idea and supporting me at every step of the process, and Chris van Hoof (Red Hat) for contributing the compute on which these quantizations were produced and evaluated.

License

These quantized weights inherit the license of the base model, Qwen/Qwen3.8-Flash-Next, the Qwen Community License 1.0; the original license text ships as LICENSE. The GSQ and RCO tooling is released by DASLab under its own repository licenses.