Instructions to use derek-thomas/jais-13b-chat-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use derek-thomas/jais-13b-chat-hf with Transformers:

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

pipe = pipeline("text-generation", model="derek-thomas/jais-13b-chat-hf", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("derek-thomas/jais-13b-chat-hf", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use derek-thomas/jais-13b-chat-hf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "derek-thomas/jais-13b-chat-hf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "derek-thomas/jais-13b-chat-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/derek-thomas/jais-13b-chat-hf

SGLang

How to use derek-thomas/jais-13b-chat-hf 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 "derek-thomas/jais-13b-chat-hf" \
    --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": "derek-thomas/jais-13b-chat-hf",
		"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 "derek-thomas/jais-13b-chat-hf" \
        --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": "derek-thomas/jais-13b-chat-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use derek-thomas/jais-13b-chat-hf with Docker Model Runner:
```
docker model run hf.co/derek-thomas/jais-13b-chat-hf
```

derek-thomas commited on Nov 10, 2023

Commit

e888c38

1 Parent(s): d29dd9f

Fixing local mistake

Browse files

Files changed (1) hide show

handler.py +5 -7

handler.py CHANGED Viewed

@@ -27,13 +27,11 @@ class EndpointHandler:
         self.device = "cuda" if torch.cuda.is_available() else "cpu"
-        # self.tokenizer = AutoTokenizer.from_pretrained(path)
-        # self.model = AutoModelForCausalLM.from_pretrained(path, device_map="auto",
-        #                                                   offload_folder='offload',
-        #                                                   trust_remote_code=True,
-        #                                                   load_in_8bit=True)
-        self.tokenizer = tokenizer
-        self.model = model
     def __call__(self, data: Dict[str, Any]) -> Dict[str, Any]:

         self.device = "cuda" if torch.cuda.is_available() else "cpu"
+        self.tokenizer = AutoTokenizer.from_pretrained(path)
+        self.model = AutoModelForCausalLM.from_pretrained(path, device_map="auto",
+                                                          offload_folder='offload',
+                                                          trust_remote_code=True,
+                                                          load_in_8bit=True)
     def __call__(self, data: Dict[str, Any]) -> Dict[str, Any]: