Instructions to use Cyclone-Labs/Eclipsed-Phoenix-26B-A4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Cyclone-Labs/Eclipsed-Phoenix-26B-A4B with Transformers:

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

pipe = pipeline("image-text-to-text", model="Cyclone-Labs/Eclipsed-Phoenix-26B-A4B")
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, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("Cyclone-Labs/Eclipsed-Phoenix-26B-A4B")
model = AutoModelForMultimodalLM.from_pretrained("Cyclone-Labs/Eclipsed-Phoenix-26B-A4B", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Cyclone-Labs/Eclipsed-Phoenix-26B-A4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Cyclone-Labs/Eclipsed-Phoenix-26B-A4B"
# 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/Eclipsed-Phoenix-26B-A4B",
		"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/Eclipsed-Phoenix-26B-A4B

SGLang

How to use Cyclone-Labs/Eclipsed-Phoenix-26B-A4B 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/Eclipsed-Phoenix-26B-A4B" \
    --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/Eclipsed-Phoenix-26B-A4B",
		"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/Eclipsed-Phoenix-26B-A4B" \
        --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/Eclipsed-Phoenix-26B-A4B",
		"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/Eclipsed-Phoenix-26B-A4B with Docker Model Runner:
```
docker model run hf.co/Cyclone-Labs/Eclipsed-Phoenix-26B-A4B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Overview

Eclipsed-Phoenix-26B-A4B is an experimental rework of Phoenix-X. A custom script makes small structured changes to the final four MoE expert layers, tests multiple versions against the original, and keeps the strongest result. These changes affect token selection in the later layers, producing substantially different same-seed generations and shifting the model’s prose, tone, personality, and response style while keeping the Phoenix-X foundation.

In testing, Eclipsed-Phoenix tended to produce more stylized prose, stronger subtext, and less predictable narrative choices. It was also more willing to take darker or more ambiguous directions than the original.

Intended Use

Roleplay

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

Creative Writing

Fiction, dialogue, atmosphere, descriptions, and stylistic drafting.

Storytelling

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

Interactive Fiction

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

Downloads last month: 19

Safetensors

Model size

26B params

Tensor type

BF16

Model tree for Cyclone-Labs/Eclipsed-Phoenix-26B-A4B

Base model

Vortex5/Phoenix-X-26B-A4B

Finetuned

(2)

this model

Quantizations

2 models