Instructions to use lballore/llimba-3b-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lballore/llimba-3b-instruct with Transformers:

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

pipe = pipeline("text-generation", model="lballore/llimba-3b-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lballore/llimba-3b-instruct")
model = AutoModelForCausalLM.from_pretrained("lballore/llimba-3b-instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use lballore/llimba-3b-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lballore/llimba-3b-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lballore/llimba-3b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lballore/llimba-3b-instruct

SGLang

How to use lballore/llimba-3b-instruct 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 "lballore/llimba-3b-instruct" \
    --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": "lballore/llimba-3b-instruct",
		"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 "lballore/llimba-3b-instruct" \
        --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": "lballore/llimba-3b-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lballore/llimba-3b-instruct with Docker Model Runner:
```
docker model run hf.co/lballore/llimba-3b-instruct
```

llimba-3b-instruct

File size: 641 Bytes

c86e92d

{
  "base_model": "/workspaces/LLiMba/models/cpt-pretrain-qwen2.5-3b",
  "dataset": "/workspaces/LLiMba/data/curated/sft/sft_dataset.jsonl",
  "mode": "lora",
  "rank": 256,
  "alpha": 256,
  "dropout": 0.05,
  "target_modules": [
    "q_proj",
    "k_proj",
    "v_proj",
    "o_proj",
    "gate_proj",
    "up_proj",
    "down_proj"
  ],
  "lr": 2e-05,
  "epochs": 2,
  "batch_size": 1,
  "grad_accum": 16,
  "effective_batch": 16,
  "max_length": 4096,
  "warmup_steps": 50,
  "attention": "flash_attention_2",
  "eval_split": 0.05,
  "train_loss": 0.867611675270807,
  "eval_loss": null,
  "train_samples": 13683,
  "eval_samples": 721
}