Instructions to use RichardErkhov/bigcode_-_starcoder2-3b-4bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RichardErkhov/bigcode_-_starcoder2-3b-4bits with Transformers:

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

pipe = pipeline("text-generation", model="RichardErkhov/bigcode_-_starcoder2-3b-4bits")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RichardErkhov/bigcode_-_starcoder2-3b-4bits")
model = AutoModelForCausalLM.from_pretrained("RichardErkhov/bigcode_-_starcoder2-3b-4bits")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use RichardErkhov/bigcode_-_starcoder2-3b-4bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RichardErkhov/bigcode_-_starcoder2-3b-4bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RichardErkhov/bigcode_-_starcoder2-3b-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/RichardErkhov/bigcode_-_starcoder2-3b-4bits

SGLang

How to use RichardErkhov/bigcode_-_starcoder2-3b-4bits 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 "RichardErkhov/bigcode_-_starcoder2-3b-4bits" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RichardErkhov/bigcode_-_starcoder2-3b-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "RichardErkhov/bigcode_-_starcoder2-3b-4bits" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RichardErkhov/bigcode_-_starcoder2-3b-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use RichardErkhov/bigcode_-_starcoder2-3b-4bits with Docker Model Runner:
```
docker model run hf.co/RichardErkhov/bigcode_-_starcoder2-3b-4bits
```

bigcode_-_starcoder2-3b-4bits

File size: 1,240 Bytes

f2b5b1d

{
  "_name_or_path": "starcoder2-3b",
  "architectures": [
    "Starcoder2ForCausalLM"
  ],
  "attention_dropout": 0.1,
  "bos_token_id": 0,
  "embedding_dropout": 0.1,
  "eos_token_id": 0,
  "hidden_act": "gelu_pytorch_tanh",
  "hidden_size": 3072,
  "initializer_range": 0.018042,
  "intermediate_size": 12288,
  "max_position_embeddings": 16384,
  "mlp_type": "default",
  "model_type": "starcoder2",
  "norm_epsilon": 1e-05,
  "norm_type": "layer_norm",
  "num_attention_heads": 24,
  "num_hidden_layers": 30,
  "num_key_value_heads": 2,
  "quantization_config": {
    "_load_in_4bit": true,
    "_load_in_8bit": false,
    "bnb_4bit_compute_dtype": "float32",
    "bnb_4bit_quant_storage": "uint8",
    "bnb_4bit_quant_type": "fp4",
    "bnb_4bit_use_double_quant": false,
    "llm_int8_enable_fp32_cpu_offload": false,
    "llm_int8_has_fp16_weight": false,
    "llm_int8_skip_modules": null,
    "llm_int8_threshold": 6.0,
    "load_in_4bit": true,
    "load_in_8bit": false,
    "quant_method": "bitsandbytes"
  },
  "residual_dropout": 0.1,
  "rope_theta": 999999.4420358813,
  "sliding_window": 4096,
  "torch_dtype": "float16",
  "transformers_version": "4.39.3",
  "use_bias": true,
  "use_cache": true,
  "vocab_size": 49152
}