Instructions to use ShiftAddLLM/opt66b-3bit-acc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ShiftAddLLM/opt66b-3bit-acc with Transformers:

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

pipe = pipeline("text-generation", model="ShiftAddLLM/opt66b-3bit-acc")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ShiftAddLLM/opt66b-3bit-acc")
model = AutoModelForCausalLM.from_pretrained("ShiftAddLLM/opt66b-3bit-acc")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ShiftAddLLM/opt66b-3bit-acc with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ShiftAddLLM/opt66b-3bit-acc"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ShiftAddLLM/opt66b-3bit-acc",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ShiftAddLLM/opt66b-3bit-acc

SGLang

How to use ShiftAddLLM/opt66b-3bit-acc 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 "ShiftAddLLM/opt66b-3bit-acc" \
    --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": "ShiftAddLLM/opt66b-3bit-acc",
		"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 "ShiftAddLLM/opt66b-3bit-acc" \
        --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": "ShiftAddLLM/opt66b-3bit-acc",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ShiftAddLLM/opt66b-3bit-acc with Docker Model Runner:
```
docker model run hf.co/ShiftAddLLM/opt66b-3bit-acc
```

opt66b-3bit-acc / config.json

ShiftAddLLM

Upload folder using huggingface_hub

a0101c0 verified almost 2 years ago

raw

history blame contribute delete

749 Bytes

	{
	"_name_or_path": "facebook/opt-66b",
	"_remove_final_layer_norm": false,
	"activation_dropout": 0.0,
	"activation_function": "relu",
	"architectures": [
	"OPTForCausalLM"
	],
	"attention_dropout": 0.0,
	"bos_token_id": 2,
	"do_layer_norm_before": true,
	"dropout": 0.1,
	"enable_bias": true,
	"eos_token_id": 2,
	"ffn_dim": 36864,
	"hidden_size": 9216,
	"init_std": 0.02,
	"layer_norm_elementwise_affine": true,
	"layerdrop": 0.0,
	"max_position_embeddings": 2048,
	"model_type": "opt",
	"num_attention_heads": 72,
	"num_hidden_layers": 64,
	"pad_token_id": 1,
	"prefix": "</s>",
	"torch_dtype": "float16",
	"transformers_version": "4.38.1",
	"use_cache": true,
	"vocab_size": 50272,
	"word_embed_proj_dim": 9216
	}