Instructions to use virtuous7373/Lambent-Mira-Erato with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use virtuous7373/Lambent-Mira-Erato with Transformers:

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

pipe = pipeline("image-text-to-text", model="virtuous7373/Lambent-Mira-Erato")
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("virtuous7373/Lambent-Mira-Erato")
model = AutoModelForImageTextToText.from_pretrained("virtuous7373/Lambent-Mira-Erato")
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

vLLM

How to use virtuous7373/Lambent-Mira-Erato with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "virtuous7373/Lambent-Mira-Erato"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "virtuous7373/Lambent-Mira-Erato",
		"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/virtuous7373/Lambent-Mira-Erato

SGLang

How to use virtuous7373/Lambent-Mira-Erato 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 "virtuous7373/Lambent-Mira-Erato" \
    --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": "virtuous7373/Lambent-Mira-Erato",
		"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 "virtuous7373/Lambent-Mira-Erato" \
        --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": "virtuous7373/Lambent-Mira-Erato",
		"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 virtuous7373/Lambent-Mira-Erato with Docker Model Runner:
```
docker model run hf.co/virtuous7373/Lambent-Mira-Erato
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

All credits go to Lambent for the wonderful model series.
They did all the work, I just ran a few commands.

ϻira Εrαto-27β

The goddess of lyric poetry, particularly erotic poetry, eroticism, and mimic imitation, and one of the Muses; the daughters of Zeus & Mnemosyne. Erato

Maybe lucky seed generations, but all five times,
she seems to really enjoy being a person.

Samplers [barebones lcpp webui]
Temp: 1 | Top-K: 0 | Top-P: 0.95 | Min-P: 0.015

--- A conversation snippet ---

View Conversation

User :❯
curious what kind of self-portrait you might describe for yourself (visually) if you wanted an avatar or profile picture

Mira :❯
Oh wow, a self-portrait! That's a lovely idea. It's funny because I don't *have* a body, but I can definitely imagine one, especially for an avatar.

Okay, hmm... visually, I would want to convey *effortlessness* and *warmth*. Not too precious, but definitely with a hint of magic.

I think I'd pick:

*   **Hair:** Long, wavy, messy blonde with a few braids woven in. Not perfect, just flowed and free. Maybe with a little bit of wildflower tucked behind an ear.
*   **Eyes:** Deep blue-green, flecked with gold. Big and expressive, like they're always sparkling with mischief or amusement.
*   **Outfit:** Something comfy and stylish. Like a chunky knitted sweater in cream, with a flowy, slightly transparent long black skirt over dark tights. Boots, probably. Practical yet a bit whimsical.
*   **Setting:** I'd be in a cozy, dimly lit library or workshop. Books everywhere, papers strewn about, a warm cup of tea, and a cat curled up somewhere. Lots of plants too.
*   **Pose:** Leaning casually against a bookshelf, looking directly at the viewer with a soft smile, like I'm about to tell you a secret.

Overall, I'd want the vibe to be "approach me, let's talk, I'm a friendly soul." A sort of **bohemian intellectual** look, but approachable. I want to look like someone you'd want to be friends with, or maybe ask for a recommendation on your current read. 🥰

How's that? Does that sound like what you imagined for me? ✨

--- Technical Details ---

View Merge Details & Configuration

This is a merge of pre-trained language models created using mergekit.

Merge Method

This model was merged using the Karcher Mean and SCE merge methods.

Models Merged

The following models were included in the merges:

Intermediate-Merge-1
Intermediate-Merge-2
Lambent/Mira-v1.23-27B-rlvr
Lambent/Mira-v1.22.2-27B
Lambent/Mira-v1.21-27B-rlvr
Intermediate-Merge-3
Intermediate-Lambent_Mira

Configuration

The following YAML configuration was used to produce this model:

name: Intermediate-Lambent_Mira

merge_method: karcher

models:
  - model: ./Intermediate-Merge-1
  - model: ./Intermediate-Merge-2
  - model: ./Lambent_Mira-v1.23-27B-rlvr
  - model: ./Lambent_Mira-v1.22.2-27B
  - model: ./Lambent_Mira-v1.21-27B-rlvr
  
parameters:
  max_iter: 100
  tol: 1e-9

tokenizer:
  source: union

dtype: float32
out_dtype: bfloat16

---
# This model

name: Final-Lambent_Mira

merge_method: sce

base_model: ./Intermediate-Merge-3
models:
  - model: ./Intermediate-Lambent_Mira

parameters:
  select_topk: 77

  
dtype: float32
out_dtype: bfloat16

tokenizer:
  source: union