Instructions to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX 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 JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX
# Run inference directly in the terminal:
llama cli -hf JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX
# Run inference directly in the terminal:
llama cli -hf JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

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 JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX
# Run inference directly in the terminal:
./llama-cli -hf JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

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 JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX
# Run inference directly in the terminal:
./build/bin/llama-cli -hf JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

Use Docker

docker model run hf.co/JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

LM Studio
Jan

Ollama

How to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX with Ollama:

ollama run hf.co/JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

Unsloth Studio

How to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for JackBinary/Qwen3.8-27B-GGUF-ROCMFPX to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for JackBinary/Qwen3.8-27B-GGUF-ROCMFPX to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for JackBinary/Qwen3.8-27B-GGUF-ROCMFPX to start chatting

How to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

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 "JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX" \
  --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"

Docker Model Runner
How to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX with Docker Model Runner:
```
docker model run hf.co/JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX
```

Lemonade

How to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

Run and chat with the model

lemonade run user.Qwen3.8-27B-GGUF-ROCMFPX-Q8_0_ROCMFPX

List all available models

lemonade list

Hermes Agent

How to use JackBinary/Qwen3.8-27B-GGUF-ROCMFPX with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

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 JackBinary/Qwen3.8-27B-GGUF-ROCMFPX:Q8_0_ROCMFPX

Run Hermes

hermes

Atomic Chat

Qwen3.8-27B — ROCmFPX quants (Q8 full + 16 GB hybrid)

These are quantizations of unsloth/Qwen3.8-27B-GGUF (BF16) (original model: Qwen/Qwen3.8-27B, the hybrid attention/SSM qwen35 architecture with an MTP nextn head).

You need the ROCmFPX fork of llama.cpp (or a llama.cpp build with ROCmFPX support). These files use the experimental q8_0_rocmfpx (type 103) and/or q4_0_rocmfp4_fast (type 101) weight formats, which stock llama.cpp releases do not understand — loading them elsewhere will fail with an unknown tensor type error.

Files

File	Recipe	Size	For
`Qwen3.8-27B-Q8_0_ROCMFPX.gguf`	pure `q8_0_rocmfpx`, all weights	26.9 GB (8.25 bpw)	large-VRAM systems (e.g. Strix Halo)
`Qwen3.8-27B-Q4FAST-Q8-sensitive.gguf`	bulk `q4_0_rocmfp4_fast` + sensitive tensors at `q8_0_rocmfpx`	16.4 GB (5.15 bpw)	~24 GB VRAM laptops (leaves room for KV cache)

Hybrid recipe (16 GB file)

Sensitive-tensor selection mirrors the tiers in Unsloth's UD-Q4_K_XL dynamic recipe (Q6_K tier = most sensitive, Q5_K = next), rebalanced onto a strict two-type q4/q8 mix to hit ~16 GB:

Tensors at `q8_0_rocmfpx` (165 tensors)	Why
`attn_q/k/v/output` (17 full-attention layers)	attention projections (Unsloth Q5_K/Q6_K tier; `attn_v` is Q6_K there)
`attn_gate` + `ssm_out` (48 linear-attn/SSM layers)	Q5_K tier in UD-Q4_K_XL
`output.weight` head	Q6_K tier in UD-Q4_K_XL

Everything else is q4_0_rocmfp4_fast (340 tensors, incl. ffn_gate/up/down, attn_qkv, token_embd — Unsloth keeps embeddings at Q4_K too), norms/biases stay f32 (360 tensors). The MTP head (nextn.eh_proj) is auto-protected at q8_0 by the quantizer's draft-sensitive logic.

How they were made

# pure Q8 (from the ROCmFPX fork; CPU-only build works fine for quantization)
llama-quantize Qwen3.8-27B-BF16-00001-of-00002.gguf \
  Qwen3.8-27B-Q8_0_ROCMFPX.gguf Q8_0_ROCMFPX

# 16 GB hybrid
llama-quantize \
  --tensor-type "attn_q.weight=q8_0_rocmfpx" \
  --tensor-type "attn_k.weight=q8_0_rocmfpx" \
  --tensor-type "attn_v.weight=q8_0_rocmfpx" \
  --tensor-type "attn_output.weight=q8_0_rocmfpx" \
  --tensor-type "attn_gate.weight=q8_0_rocmfpx" \
  --tensor-type "ssm_out.weight=q8_0_rocmfpx" \
  --tensor-type "^output.weight=q8_0_rocmfpx" \
  Qwen3.8-27B-BF16-00001-of-00002.gguf \
  Qwen3.8-27B-Q4FAST-Q8-sensitive.gguf Q4_0_ROCMFP4_FAST

Usage

This quant is text weights only. Qwen3.8 is multimodal — for vision support, pair it with the mmproj file: this repo includes mmproj-F16.gguf (mirrored from unsloth/Qwen3.8-27B-GGUF, which also offers mmproj-BF16.gguf).

# build ROCmFPX for your GPU (see the repo README; e.g. Strix Halo):
env JOBS=16 scripts/build-strix-rocmfp4-mtp.sh

./build-strix-rocmfp4/bin/llama-cli \
  -m Qwen3.8-27B-Q8_0_ROCMFPX.gguf --mmproj mmproj-F16.gguf \
  -dev Vulkan0 -ngl 999 -fa on --jinja

Benchmarks

Benchmarks are pending — placeholder tables below.

Backend / GPU	File	Prompt (tok/s)	Generation (tok/s)	Context	Notes
TBD (Strix Halo)	Q8_0_ROCMFPX	TBD	TBD	TBD	TBD
TBD (24 GB laptop)	Q4FAST-Q8-sensitive	TBD	TBD	TBD	TBD

Quality comparison vs BF16 source (perplexity / KLD): TBD.

Attribution & license

Quantized from: unsloth/Qwen3.8-27B-GGUF (BF16 shards); sensitivity tiers referenced from their UD-Q4_K_XL
Original model: Qwen/Qwen3.8-27B
License: Apache-2.0 (inherited)
Quant formats by the ROCmFPX project