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

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use dmeight/Qwen3.8-27B-Splash-8bit 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 dmeight/Qwen3.8-27B-Splash-8bit:Q8_0
# Run inference directly in the terminal:
llama cli -hf dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf dmeight/Qwen3.8-27B-Splash-8bit:Q8_0
# Run inference directly in the terminal:
llama cli -hf dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

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 dmeight/Qwen3.8-27B-Splash-8bit:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

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 dmeight/Qwen3.8-27B-Splash-8bit:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

Use Docker

docker model run hf.co/dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

Llama new
LM Studio
Jan

vLLM

How to use dmeight/Qwen3.8-27B-Splash-8bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dmeight/Qwen3.8-27B-Splash-8bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dmeight/Qwen3.8-27B-Splash-8bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

Ollama
How to use dmeight/Qwen3.8-27B-Splash-8bit with Ollama:
```
ollama run hf.co/dmeight/Qwen3.8-27B-Splash-8bit:Q8_0
```
Unsloth Desktop

How to use dmeight/Qwen3.8-27B-Splash-8bit with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

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": "dmeight/Qwen3.8-27B-Splash-8bit:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use dmeight/Qwen3.8-27B-Splash-8bit with Docker Model Runner:
```
docker model run hf.co/dmeight/Qwen3.8-27B-Splash-8bit:Q8_0
```

Lemonade

How to use dmeight/Qwen3.8-27B-Splash-8bit with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

Run and chat with the model

lemonade run user.Qwen3.8-27B-Splash-8bit-Q8_0

List all available models

lemonade list

Hermes Agent

How to use dmeight/Qwen3.8-27B-Splash-8bit with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

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 dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use dmeight/Qwen3.8-27B-Splash-8bit with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf dmeight/Qwen3.8-27B-Splash-8bit:Q8_0

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 "dmeight/Qwen3.8-27B-Splash-8bit:Q8_0" \
  --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"

Qwen3.8-27B Splash 8-bit (unofficial)

A Splash package of Qwen3.8-27B with an 8-bit target model (GGUF Q8_0), for the Splash runtime in LM Studio on Apple silicon. It pairs the 8-bit weights with the same DFlash2 draft, vision tower, tokenizer and chat template as the official 4-bit package incoai/Qwen3.8-27B-Splash.

Unofficial and independent. Not made or endorsed by Inco AI (Splash, DFlash2), LM Studio, Alibaba Cloud (Qwen) or Unsloth. No weights were changed: every weight file is a byte-for-byte copy of its source.

Why 8-bit

The official package's target is 4-bit. 8-bit weights stay much closer to the original model, at roughly 60% of the 4-bit package's writing speed (see Measurements).

What's inside

Part	Source	Notes
Target model	unsloth/Qwen3.8-27B-GGUF @ `4ca7207`, `Qwen3.8-27B-Q8_0.gguf`	unchanged
Draft model	incoai/Qwen3.8-27B-DFlash2 @ `015e795`	unchanged BF16; Splash quantizes it to 4-bit at load, exactly as in the official package
Vision tower	`mmproj-BF16.gguf` from the Unsloth repo; `vision/model.bin` from the official package	the engine reads the mmproj; `vision/model.bin` is there so that LM Studio enables image input
Tokenizer, chat template	incoai/Qwen3.8-27B-Splash	identical to the official package
`model.json`, `config.json`	written by Splash's own installer (splash `0d5a035e`, `install/models.py --model unsloth/Qwen3.8-27B-GGUF:Q8_0 prepare`)	local file paths removed
`manifest.json`	this package	lets LM Studio list and download it

Requirements

Apple silicon M3 or newer, macOS 26.4 or later (Splash's requirements).
LM Studio with the Splash runtime enabled (Settings → Runtime → Splash). Tested with LM Studio 0.4.26+4 and the Splash runtime 1.3.0.
About 32 GB of memory for the model when loaded; tested on a 64 GB Mac. Download size: 34.8 GB.

Install

In LM Studio, enable the Splash runtime, then download dmeight/Qwen3.8-27B-Splash-8bit from the model search. It appears as Qwen3.8 27B Splash 8bit.

Known issues

LM Studio labels it "4bit". LM Studio currently gives every Splash package a fixed 4-bit label, whatever its weights. The weights here are 8-bit: the package is 34.8 GB and takes about 31.5 GB of memory when loaded, while the official 4-bit package is 17.4 GB.
It relies on undocumented behaviour, so an update may break it. LM Studio lists a folder as a Splash model only if its manifest.json names the 4-bit package format (splash-packed-q4). The Splash engine reads an installer assembly (model.json) before a manifest. This package carries both: the manifest for LM Studio, the assembly for the engine. A newer LM Studio or Splash runtime may stop loading it.
It is a stopgap. Once LM Studio ships Splash 1.3.1 or later, which loads 8-bit models directly, prefer the official route.

Without LM Studio

This package is for LM Studio only. Elsewhere, use the original sources:

Standalone Splash rejects this package, because its installer expects the official package's per-layer files. It doesn't need the package either: splash serve --model unsloth/Qwen3.8-27B-GGUF:Q8_0 builds the same set (the same GGUF, vision projector and DFlash2 draft) from the original repositories. Splash 1.3.1 and later also load MLX 8-bit (mlx-community/Qwen3.8-27B-8bit).
llama.cpp, Ollama and other GGUF tools can use target/Qwen3.8-27B-Q8_0.gguf and vision/mmproj.gguf as they are: plain GGUF files, identical to Unsloth's. They run without the DFlash2 speedup. It is simpler to download them from unsloth/Qwen3.8-27B-GGUF.
MLX tools cannot use it: the weights are GGUF, not MLX.

Measurements

Mac Studio, Apple M5 Max (40-core GPU), 64 GB, macOS 27.0.1, 2026-10-26.

Speed through LM Studio. Mixed prompts (Polish and English prose, code), thinking on, temperature 1.0, top_p 0.95, top_k 20, 800-token answers, two runs each, median.

	This package (8-bit)	Official Splash (4-bit)	LM Studio MLX 8-bit, no speculative decoding
Writing, tokens/s	54.8	91.2	18.1
Writing at a 15K-token context, tokens/s	39.2	80.7	17.2
Reading a 15K-token prompt, s	13.5	13.3	17.8

Draft acceptance. The share of DFlash2 proposals the target accepted. 24 prompts in six groups, 512-token answers, ABBA order (4-bit, 8-bit, 8-bit, 4-bit), counters read from the Splash engine.

	4-bit target	8-bit target
Sampling (T 1.0, two seed sets): accepted	25.7%	28.5%
Tokens per verify cycle	2.80	3.00
Greedy: accepted	29.1%	28.7%

With sampling, the 8-bit target accepts more of the draft's proposals: +3.7 points on average over the 24 prompts, 95% bootstrap interval +1.1 to +6.5. Greedy decoding shows no difference. Greedy outputs were identical across repeated runs.

Licenses and credits

Every part is licensed under Apache-2.0 (see LICENSE); NOTICE lists the sources and what this package adds.

Qwen3.8-27B by Alibaba Cloud (Qwen team).
GGUF quantization and vision projector by Unsloth.
Splash engine, package format, DFlash2 draft model and the official package's tokenizer and vision files by Inco AI.

Packaged by dmeight.

Downloads last month: 58

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

8-bit

Model tree for dmeight/Qwen3.8-27B-Splash-8bit

Base model

Qwen/Qwen3.8-27B

Quantized

(1604)

this model