Instructions to use internlm/internlm-chat-20b-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use internlm/internlm-chat-20b-4bit with Transformers:

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

pipe = pipeline("text-generation", model="internlm/internlm-chat-20b-4bit", trust_remote_code=True)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("internlm/internlm-chat-20b-4bit", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use internlm/internlm-chat-20b-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "internlm/internlm-chat-20b-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "internlm/internlm-chat-20b-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/internlm/internlm-chat-20b-4bit

SGLang

How to use internlm/internlm-chat-20b-4bit 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 "internlm/internlm-chat-20b-4bit" \
    --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": "internlm/internlm-chat-20b-4bit",
		"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 "internlm/internlm-chat-20b-4bit" \
        --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": "internlm/internlm-chat-20b-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use internlm/internlm-chat-20b-4bit with Docker Model Runner:
```
docker model run hf.co/internlm/internlm-chat-20b-4bit
```

although tokenizer.model exists , deploy is complaining that it does not

by TeaCult - opened Sep 29, 2023

Discussion

TeaCult

Sep 29, 2023

python3 -m lmdeploy.serve.turbomind.deploy
--model-name internlm-chat-20b
--model-path ./internlm-chat-20b
--model-format awq
--group-size 128
create workspace in directory ./workspace
copy triton model templates from "/home/gediz/.pyenv/versions/lmdeploy/lib/python3.10/site-packages/lmdeploy/serve/turbomind/triton_models" to "./workspace/triton_models" successfully
tokenizer model ./internlm-chat-20b/tokenizer.model does not exist

ls -lah
drwxr-xr-x 3 gediz gediz 4.0K Sep 29 19:41 .
drwx------ 24 gediz gediz 4.0K Sep 29 19:57 ..
-rw-r--r-- 1 gediz gediz 905 Sep 29 19:35 config.json
-rw-r--r-- 1 gediz gediz 5.1K Sep 29 19:35 configuration_internlm.py
-rw-r--r-- 1 gediz gediz 111 Sep 29 19:35 generation_config.json
drwxr-xr-x 9 gediz gediz 4.0K Sep 29 19:41 .git
-rw-r--r-- 1 gediz gediz 1.5K Sep 29 19:35 .gitattributes
-rw-r--r-- 1 gediz gediz 27M Sep 29 19:35 inputs_stats.pth
-rw-r--r-- 1 gediz gediz 1.9M Sep 29 19:35 key_stats.pth
-rw-r--r-- 1 gediz gediz 43K Sep 29 19:35 modeling_internlm.py
-rw-r--r-- 1 gediz gediz 38M Sep 29 19:35 outputs_stats.pth
-rw-r--r-- 1 gediz gediz 9.4G Sep 29 19:41 pytorch_model-00001-of-00002.bin
-rw-r--r-- 1 gediz gediz 1.9G Sep 29 19:36 pytorch_model-00002-of-00002.bin
-rw-r--r-- 1 gediz gediz 111K Sep 29 19:35 pytorch_model.bin.index.json
-rw-r--r-- 1 gediz gediz 4.0K Sep 29 19:35 README.md
-rw-r--r-- 1 gediz gediz 95 Sep 29 19:35 special_tokens_map.json
-rw-r--r-- 1 gediz gediz 8.8K Sep 29 19:35 tokenization_internlm.py
-rw-r--r-- 1 gediz gediz 343 Sep 29 19:35 tokenizer_config.json
-rw-r--r-- 1 gediz gediz 1.6M Sep 29 19:35 tokenizer.model
-rw-r--r-- 1 gediz gediz 1.9M Sep 29 19:35 value_stats.pth

unsubscribe

Intern Large Models org Oct 9, 2023

May pass absolute path for --model-path

allenzhong1214

Sep 9, 2024

You may want to check the official website for the .bin file

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment