Instructions to use exoro/qwen3.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use exoro/qwen3.5 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="exoro/qwen3.5")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("exoro/qwen3.5")
model = AutoModelForImageTextToText.from_pretrained("exoro/qwen3.5")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

llama-cpp-python

How to use exoro/qwen3.5 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="exoro/qwen3.5",
	filename="Qwen3.5-0.8B.F16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use exoro/qwen3.5 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf exoro/qwen3.5:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf exoro/qwen3.5:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf exoro/qwen3.5:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf exoro/qwen3.5: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 exoro/qwen3.5:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf exoro/qwen3.5: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 exoro/qwen3.5:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf exoro/qwen3.5:Q4_K_M

Use Docker

docker model run hf.co/exoro/qwen3.5:Q4_K_M

LM Studio
Jan

vLLM

How to use exoro/qwen3.5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "exoro/qwen3.5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "exoro/qwen3.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/exoro/qwen3.5:Q4_K_M

SGLang

How to use exoro/qwen3.5 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "exoro/qwen3.5" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "exoro/qwen3.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "exoro/qwen3.5" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "exoro/qwen3.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use exoro/qwen3.5 with Ollama:
```
ollama run hf.co/exoro/qwen3.5:Q4_K_M
```

Unsloth Studio new

How to use exoro/qwen3.5 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 exoro/qwen3.5 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 exoro/qwen3.5 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for exoro/qwen3.5 to start chatting

Pi new

How to use exoro/qwen3.5 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf exoro/qwen3.5: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": "exoro/qwen3.5:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use exoro/qwen3.5 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf exoro/qwen3.5: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 exoro/qwen3.5:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use exoro/qwen3.5 with Docker Model Runner:
```
docker model run hf.co/exoro/qwen3.5:Q4_K_M
```

Lemonade

How to use exoro/qwen3.5 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull exoro/qwen3.5:Q4_K_M

Run and chat with the model

lemonade run user.qwen3.5-Q4_K_M

List all available models

lemonade list

Qwen3.5-0.8B — System Command Intent Parser

Fine-tuned from Qwen/Qwen3.5-0.8B using QLoRA on a custom dataset of system monitoring inputs mapped to structured JSON action plans.

Given a natural language system alert, the model outputs a structured JSON describing the intent, required actions, risk level, and confidence score.

Example

Input

RAM at 95%, system becoming unresponsive

Output

{
  "intent": "memory_check",
  "steps": [
    {"action": "check_memory", "params": {}},
    {"action": "list_processes", "params": {"sort": "memory", "limit": 10}}
  ],
  "risk": "medium",
  "confidence": 0.88
}

Output Schema

Field	Type	Description
`intent`	string	Classified action category (e.g. `memory_check`, `disk_cleanup`, `cpu_check`)
`steps`	array	Ordered list of actions with optional parameters
`risk`	string	`low` / `medium` / `high`
`confidence`	float	Model confidence 0–1

Training Details

Property	Value
Base model	Qwen/Qwen3.5-0.8B
Fine-tuning method	QLoRA (4-bit NF4)
LoRA rank / alpha	r=16, α=32
LoRA dropout	0.05
Target modules	q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
Dataset	374 custom JSONL examples
Epochs	3
Batch size	4 (effective 16 with grad accumulation)
Optimizer	paged_adamw_8bit
Learning rate	2e-4
LR scheduler	cosine
Warmup steps	10
Precision	bf16
Max sequence length	1024
Hardware	Tesla T4 — Google Colab

Prompt Format

The model uses Qwen3.5's native ChatML format:

<|im_start|>system
You are a structured data extractor. Given an input, output a valid JSON object and nothing else.<|im_end|>
<|im_start|>user
{your system alert here}<|im_end|>
<|im_start|>assistant

Limitations

Trained on a system-monitoring domain — may not generalise to unrelated tasks
Small model (0.8B) — complex multi-step plans may occasionally be incomplete
Risk and confidence values reflect training data distribution; treat as heuristic
Dataset size (374 examples) is small — broader coverage would improve robustness

Related Repos

Repo	Contents
exoro/qwen3.5	Merged full model (fp16)
exoro/qwen3.5-0.8B-finetuned-merged	Earlier merged checkpoint

Downloads last month: 2,036

GGUF

Model size

0.8B params

Architecture

qwen35

Hardware compatibility

4-bit

8-bit

16-bit

Model tree for exoro/qwen3.5

Base model

Qwen/Qwen3.5-0.8B-Base

Finetuned

Qwen/Qwen3.5-0.8B

Quantized

(125)

this model