Instructions to use lmdeploy/internlm-chat-7b-w4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lmdeploy/internlm-chat-7b-w4 with Transformers:

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

pipe = pipeline("text-generation", model="lmdeploy/internlm-chat-7b-w4", trust_remote_code=True)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("lmdeploy/internlm-chat-7b-w4", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lmdeploy/internlm-chat-7b-w4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lmdeploy/internlm-chat-7b-w4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lmdeploy/internlm-chat-7b-w4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lmdeploy/internlm-chat-7b-w4

SGLang

How to use lmdeploy/internlm-chat-7b-w4 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 "lmdeploy/internlm-chat-7b-w4" \
    --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": "lmdeploy/internlm-chat-7b-w4",
		"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 "lmdeploy/internlm-chat-7b-w4" \
        --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": "lmdeploy/internlm-chat-7b-w4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use lmdeploy/internlm-chat-7b-w4 with Docker Model Runner:
```
docker model run hf.co/lmdeploy/internlm-chat-7b-w4
```

support transformers 4.34.0

by unsubscribe - opened Oct 9, 2023

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

-7

Files changed (1) hide show

tokenization_internlm.py +7 -7

tokenization_internlm.py CHANGED Viewed

@@ -65,6 +65,13 @@ class InternLMTokenizer(PreTrainedTokenizer):
         **kwargs,
     ):
         self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
         super().__init__(
             bos_token=bos_token,
             eos_token=eos_token,
@@ -73,13 +80,6 @@ class InternLMTokenizer(PreTrainedTokenizer):
             clean_up_tokenization_spaces=clean_up_tokenization_spaces,
             **kwargs,
         )
-        self.vocab_file = vocab_file
-        self.add_bos_token = add_bos_token
-        self.add_eos_token = add_eos_token
-        self.decode_with_prefix_space = decode_with_prefix_space
-        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
-        self.sp_model.Load(vocab_file)
-        self._no_prefix_space_tokens = None
         """ Initialisation"""

         **kwargs,
     ):
         self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.decode_with_prefix_space = decode_with_prefix_space
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        self._no_prefix_space_tokens = None
         super().__init__(
             bos_token=bos_token,
             eos_token=eos_token,
             clean_up_tokenization_spaces=clean_up_tokenization_spaces,
             **kwargs,
         )
         """ Initialisation"""