Instructions to use FrontiersMind/Lumma-0.6B-Extract with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FrontiersMind/Lumma-0.6B-Extract with Transformers:

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

pipe = pipeline("text-generation", model="FrontiersMind/Lumma-0.6B-Extract", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("FrontiersMind/Lumma-0.6B-Extract", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use FrontiersMind/Lumma-0.6B-Extract with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "FrontiersMind/Lumma-0.6B-Extract"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "FrontiersMind/Lumma-0.6B-Extract",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/FrontiersMind/Lumma-0.6B-Extract

SGLang

How to use FrontiersMind/Lumma-0.6B-Extract 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 "FrontiersMind/Lumma-0.6B-Extract" \
    --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": "FrontiersMind/Lumma-0.6B-Extract",
		"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 "FrontiersMind/Lumma-0.6B-Extract" \
        --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": "FrontiersMind/Lumma-0.6B-Extract",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use FrontiersMind/Lumma-0.6B-Extract with Docker Model Runner:
```
docker model run hf.co/FrontiersMind/Lumma-0.6B-Extract
```

Lumma-0.6B-Extract / config.json

vishesh-t27

Upload folder using huggingface_hub

f901822 verified 3 days ago

Raw

History Blame Contribute Delete

1.11 kB

	{
	"architectures": [
	"NandiForCausalLM"
	],
	"attention_bias": false,
	"attention_dropout": 0.0,
	"auto_map": {
	"AutoConfig": "configuration_nandi.NandiConfig",
	"AutoModel": "modeling_nandi.NandiModel",
	"AutoModelForCausalLM": "modeling_nandi.NandiForCausalLM"
	},
	"bos_token_id": 1,
	"dtype": "bfloat16",
	"embedding_rank": 512,
	"eos_token_id": 0,
	"factorized_embedding": true,
	"head_dim": 90,
	"hidden_act": "silu",
	"hidden_size": 1440,
	"initializer_range": 0.02,
	"intermediate_size": 3280,
	"kv_cache_mode": "shared",
	"layer_sharing": false,
	"layer_sharing_repeats": 1,
	"max_position_embeddings": 12288,
	"mlp_bias": false,
	"model_type": "nandi",
	"num_attention_heads": 16,
	"num_hidden_layers": 30,
	"num_key_value_heads": 8,
	"pad_token_id": 3,
	"pretraining_tp": 1,
	"q_norm": true,
	"qk_norm": false,
	"rms_norm_eps": 1e-06,
	"rope_parameters": {
	"rope_theta": 1000000.0,
	"rope_type": "default"
	},
	"shared_kv": true,
	"tie_word_embeddings": true,
	"transformers_version": "5.14.1",
	"use_cache": false,
	"vocab_size": 131072
	}