Instructions to use Cyclone-Labs/Lunar-Trace-31B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Cyclone-Labs/Lunar-Trace-31B with Transformers:

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

pipe = pipeline("image-text-to-text", model="Cyclone-Labs/Lunar-Trace-31B")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("Cyclone-Labs/Lunar-Trace-31B")
model = AutoModelForMultimodalLM.from_pretrained("Cyclone-Labs/Lunar-Trace-31B", device_map="auto")
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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Cyclone-Labs/Lunar-Trace-31B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Cyclone-Labs/Lunar-Trace-31B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Cyclone-Labs/Lunar-Trace-31B",
		"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/Cyclone-Labs/Lunar-Trace-31B

SGLang

How to use Cyclone-Labs/Lunar-Trace-31B 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 "Cyclone-Labs/Lunar-Trace-31B" \
    --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": "Cyclone-Labs/Lunar-Trace-31B",
		"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 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 "Cyclone-Labs/Lunar-Trace-31B" \
        --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": "Cyclone-Labs/Lunar-Trace-31B",
		"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"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use Cyclone-Labs/Lunar-Trace-31B with Docker Model Runner:
```
docker model run hf.co/Cyclone-Labs/Lunar-Trace-31B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Overview

Lunar-Trace-31B was created by combining gemma-4-31B-it, Starlight-Synthesis-31B, G4-LuneLight-31B, and gemma-4-31B-it-Claude-Opus-Distill-v2 using a custom merge method.

View merge configuration · YAML

base_model: google/gemma-4-31B-it
merge_method: mdr
models:
  - model: Cyclone-Labs/Starlight-Synthesis-31B
    parameters:
      weight:
        - filter: vision_tower
          value: 0.0
        - filter: embed_vision
          value: 0.0
        - filter: self_attn.q_proj
          value: [0.46, 0.47, 0.48, 0.49, 0.50, 0.51, 0.52, 0.53]
        - filter: self_attn.k_proj
          value: [0.47, 0.48, 0.49, 0.50, 0.51, 0.52, 0.53, 0.54]
        - filter: self_attn.v_proj
          value: [0.50, 0.52, 0.54, 0.56, 0.58, 0.60, 0.62, 0.64]
        - filter: self_attn.o_proj
          value: [0.52, 0.54, 0.56, 0.59, 0.61, 0.64, 0.66, 0.68]
        - filter: mlp.gate_proj
          value: [0.64, 0.67, 0.70, 0.72, 0.74, 0.76, 0.78, 0.80]
        - filter: mlp.up_proj
          value: [0.63, 0.65, 0.68, 0.71, 0.74, 0.76, 0.78, 0.80]
        - filter: mlp.down_proj
          value: [0.62, 0.64, 0.67, 0.70, 0.72, 0.75, 0.78, 0.80]
        - filter: self_attn.q_norm
          value: 0.60
        - filter: self_attn.k_norm
          value: 0.60
        - filter: layernorm
          value: 0.65
        - filter: layer_scalar
          value: 0.70
        - filter: language_model.norm
          value: 0.65
        - value: 0.33
  - model: zerofata/G4-LuneLight-31B
    parameters:
      weight:
        - filter: vision_tower
          value: 0.0
        - filter: embed_vision
          value: 0.0
        - filter: self_attn.q_proj
          value: [0.23, 0.23, 0.23, 0.22, 0.21, 0.20, 0.19, 0.18]
        - filter: self_attn.k_proj
          value: [0.24, 0.24, 0.23, 0.22, 0.21, 0.20, 0.19, 0.18]
        - filter: self_attn.v_proj
          value: [0.34, 0.33, 0.32, 0.31, 0.30, 0.29, 0.28, 0.27]
        - filter: self_attn.o_proj
          value: [0.35, 0.34, 0.33, 0.30, 0.28, 0.25, 0.23, 0.21]
        - filter: mlp.gate_proj
          value: [0.26, 0.23, 0.20, 0.18, 0.16, 0.14, 0.12, 0.10]
        - filter: mlp.up_proj
          value: [0.27, 0.25, 0.22, 0.19, 0.16, 0.14, 0.12, 0.10]
        - filter: mlp.down_proj
          value: [0.28, 0.26, 0.23, 0.20, 0.18, 0.15, 0.12, 0.10]
        - filter: self_attn.q_norm
          value: 0.40
        - filter: self_attn.k_norm
          value: 0.40
        - filter: layernorm
          value: 0.35
        - filter: layer_scalar
          value: 0.30
        - filter: language_model.norm
          value: 0.35
        - value: 0.33
  - model: TeichAI/gemma-4-31B-it-Claude-Opus-Distill-v2
    parameters:
      weight:
        - filter: vision_tower
          value: 0.0
        - filter: embed_vision
          value: 0.0
        - filter: self_attn.q_proj
          value: [0.31, 0.30, 0.29, 0.29, 0.29, 0.29, 0.29, 0.29]
        - filter: self_attn.k_proj
          value: [0.29, 0.28, 0.28, 0.28, 0.28, 0.28, 0.28, 0.28]
        - filter: self_attn.v_proj
          value: [0.16, 0.15, 0.14, 0.13, 0.12, 0.11, 0.10, 0.09]
        - filter: self_attn.o_proj
          value: [0.13, 0.12, 0.11, 0.11, 0.11, 0.11, 0.11, 0.11]
        - filter: mlp
          value: 0.10
        - value: 0.05
parameters:
  gain: 0.7
  inheritance: 0.45
  anchor: 0.60
  prior_strength: 0.95
  containment: 1.10
dtype: float32
out_dtype: bfloat16
tokenizer:
  source: base
chat_template: auto

Testing Notes

In story-writing tests, the model showed a tendency toward dark psychological themes, even when the prompts did not specifically request them.

Intended Use

01 Roleplay

Character-driven interaction, personas, dialogue, emotional scenes, and long-form roleplay.

02 Creative Writing

Fiction, dialogue, atmosphere, descriptive writing, stylistic drafting, and imaginative prose.

03 Storytelling

Long-form narratives, worldbuilding, continuity, evolving characters, and multi-character plots.

04 Interactive Fiction

Branching narratives, scenario play, character interaction, and continuously evolving stories.

Downloads last month: 17

Safetensors

Model size

31B params

Tensor type

BF16

Model tree for Cyclone-Labs/Lunar-Trace-31B

Cyclone-Labs/Starlight-Synthesis-31B

TeichAI/gemma-4-31B-it-Claude-Opus-Distill-v2

google/gemma-4-31B-it

zerofata/G4-LuneLight-31B

Merge model

this model

Quantizations

2 models

Cyclone-Labs
/

Lunar-Trace-31B

Lunar-Trace 31B