Instructions to use cosmicstack/cosmic-lightning with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cosmicstack/cosmic-lightning with Transformers:

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

pipe = pipeline("text-generation", model="cosmicstack/cosmic-lightning", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("cosmicstack/cosmic-lightning", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cosmicstack/cosmic-lightning with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/cosmicstack/cosmic-lightning

SGLang

How to use cosmicstack/cosmic-lightning 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 "cosmicstack/cosmic-lightning" \
    --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": "cosmicstack/cosmic-lightning",
		"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 "cosmicstack/cosmic-lightning" \
        --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": "cosmicstack/cosmic-lightning",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cosmicstack/cosmic-lightning with Docker Model Runner:
```
docker model run hf.co/cosmicstack/cosmic-lightning
```

cosmic-lightning

File size: 1,483 Bytes

00a0cd9

{
  "architectures": [
    "Lfm3ForCausalLM"
  ],
  "auto_map": {
    "AutoConfig": "configuration_lfm3.Lfm3Config",
    "AutoModel": "modeling_lfm3.Lfm3Model",
    "AutoModelForCausalLM": "modeling_lfm3.Lfm3ForCausalLM"
  },
  "bos_token_id": 124894,
  "conv_L_cache": 7,
  "conv_bias": false,
  "dtype": "bfloat16",
  "eos_token_id": 124900,
  "head_dim": 128,
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 7680,
  "layer_types": [
    "conv",
    "conv",
    "full_attention",
    "conv",
    "sliding_attention",
    "conv",
    "full_attention",
    "conv",
    "sliding_attention",
    "conv",
    "full_attention",
    "conv",
    "sliding_attention",
    "conv",
    "full_attention",
    "conv",
    "sliding_attention",
    "conv",
    "full_attention",
    "conv",
    "sliding_attention",
    "conv",
    "full_attention",
    "conv"
  ],
  "max_position_embeddings": 262144,
  "model_type": "lfm3",
  "norm_eps": 1e-05,
  "num_attention_heads": 16,
  "num_hidden_layers": 24,
  "num_key_value_heads": 4,
  "pad_token_id": 124893,
  "rope_parameters": {
    "full_attention": {
      "partial_rotary_factor": 0.5,
      "rope_theta": 1000000.0,
      "rope_type": "default"
    },
    "sliding_attention": {
      "rope_theta": 10000,
      "rope_type": "default"
    }
  },
  "sliding_window": 256,
  "tie_word_embeddings": true,
  "transformers_version": "5.20.2",
  "use_cache": true,
  "use_sandwich_norm": true,
  "vocab_size": 128000
}