Instructions to use Exil01/Qwen3.5-27B-Opus-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Exil01/Qwen3.5-27B-Opus-v2 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Exil01/Qwen3.5-27B-Opus-v2",
	filename="Qwen3.5-27B-Opus-v2-BF16.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Exil01/Qwen3.5-27B-Opus-v2 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Exil01/Qwen3.5-27B-Opus-v2: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 Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Exil01/Qwen3.5-27B-Opus-v2: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 Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M

Use Docker

docker model run hf.co/Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M

LM Studio
Jan
Ollama
How to use Exil01/Qwen3.5-27B-Opus-v2 with Ollama:
```
ollama run hf.co/Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M
```

Unsloth Studio new

How to use Exil01/Qwen3.5-27B-Opus-v2 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 Exil01/Qwen3.5-27B-Opus-v2 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 Exil01/Qwen3.5-27B-Opus-v2 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Exil01/Qwen3.5-27B-Opus-v2 to start chatting

Pi new

How to use Exil01/Qwen3.5-27B-Opus-v2 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M

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": "Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Exil01/Qwen3.5-27B-Opus-v2 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Exil01/Qwen3.5-27B-Opus-v2: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 Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use Exil01/Qwen3.5-27B-Opus-v2 with Docker Model Runner:
```
docker model run hf.co/Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M
```

Lemonade

How to use Exil01/Qwen3.5-27B-Opus-v2 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Exil01/Qwen3.5-27B-Opus-v2:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.5-27B-Opus-v2-Q4_K_M

List all available models

lemonade list

Qwen3.5-27B Opus v2 - Reasoning Model

Base Model: Qwen3.5-27B BF16
Architecture: Qwen3.5 MoE
Purpose: Complex reasoning, chain-of-thought, math, science, analysis
Context: 262K tokens

Recommended Settings

Thinking Mode (Default - Recommended)

Use for: Math, science, logic, complex reasoning, multi-step problems

--temp 0.6 --top-p 0.95 --top-k 20 --min-p 0 --reasoning on

Non-Thinking Mode

Use for: Creative writing, casual chat, fast responses

--temp 0.7 --top-p 0.8 --top-k 20 --min-p 0

Quantization Guide

Quant	Size	Quality	Use Case
Q4_K_M	16GB	⭐⭐⭐⭐⭐	Daily use (recommended)
Q5_K_M	18GB	⭐⭐⭐⭐⭐	High quality tasks
Q6_K	22GB	⭐⭐⭐⭐⭐⭐	Near-lossless
Q8_0	28GB	⭐⭐⭐⭐⭐⭐⭐	Maximum quality
BF16	54GB	⭐⭐⭐⭐⭐⭐⭐	Original quality

Quick Start

llama-server -m Qwen3.5-27B-Opus-v2-Q4_K_M.gguf   --ctx-size 262144 --temp 0.6 --top-p 0.95 --top-k 20 --reasoning on

Vision Support

Use with mmproj for multimodal capabilities:

--mmproj mmproj-Qwen3.5-27B-Opus-v2-f16.gguf

Model Variants

Opus v2 (this repo) - Standard 262K context
Opus v2 YaRN 1M - Extended 1M context (separate repo)

Downloads last month: 76

GGUF

Model size

25B params

Architecture

qwen35

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support