Instructions to use RedHatAI/starcoder2-15b-quantized.w8a16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RedHatAI/starcoder2-15b-quantized.w8a16 with Transformers:

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

pipe = pipeline("text-generation", model="RedHatAI/starcoder2-15b-quantized.w8a16")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RedHatAI/starcoder2-15b-quantized.w8a16")
model = AutoModelForCausalLM.from_pretrained("RedHatAI/starcoder2-15b-quantized.w8a16")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use RedHatAI/starcoder2-15b-quantized.w8a16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RedHatAI/starcoder2-15b-quantized.w8a16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RedHatAI/starcoder2-15b-quantized.w8a16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/RedHatAI/starcoder2-15b-quantized.w8a16

SGLang

How to use RedHatAI/starcoder2-15b-quantized.w8a16 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 "RedHatAI/starcoder2-15b-quantized.w8a16" \
    --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": "RedHatAI/starcoder2-15b-quantized.w8a16",
		"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 "RedHatAI/starcoder2-15b-quantized.w8a16" \
        --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": "RedHatAI/starcoder2-15b-quantized.w8a16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use RedHatAI/starcoder2-15b-quantized.w8a16 with Docker Model Runner:
```
docker model run hf.co/RedHatAI/starcoder2-15b-quantized.w8a16
```

starcoder2-15b-quantized.w8a16 / config.json

mgoin

Updated compression_config to quantization_config

25ac565 verified over 1 year ago

raw

history blame contribute delete

1.62 kB

	{
	"_name_or_path": "/root/.cache/huggingface/hub/models--bigcode--starcoder2-15b/snapshots/46d44742909c03ac8cee08eb03fdebce02e193ec",
	"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": 6144,
	"initializer_range": 0.01275,
	"intermediate_size": 24576,
	"max_position_embeddings": 16384,
	"mlp_type": "default",
	"model_type": "starcoder2",
	"norm_epsilon": 1e-05,
	"norm_type": "layer_norm",
	"num_attention_heads": 48,
	"num_hidden_layers": 40,
	"num_key_value_heads": 4,
	"residual_dropout": 0.1,
	"rope_theta": 100000,
	"sliding_window": 4096,
	"tie_word_embeddings": false,
	"torch_dtype": "float32",
	"transformers_version": "4.43.3",
	"use_bias": true,
	"use_cache": true,
	"vocab_size": 49152,
	"quantization_config": {
	"config_groups": {
	"group_0": {
	"input_activations": null,
	"output_activations": null,
	"targets": [
	"Linear"
	],
	"weights": {
	"block_structure": null,
	"dynamic": false,
	"group_size": null,
	"num_bits": 8,
	"observer": "minmax",
	"observer_kwargs": {},
	"strategy": "channel",
	"symmetric": true,
	"type": "int"
	}
	}
	},
	"format": "pack-quantized",
	"global_compression_ratio": 1.80713464481227,
	"ignore": [
	"lm_head"
	],
	"kv_cache_scheme": null,
	"quant_method": "compressed-tensors",
	"quantization_status": "frozen"
	}
	}