Instructions to use NewBreaker/chatglm-6b-int4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NewBreaker/chatglm-6b-int4 with Transformers:

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

pipe = pipeline("text-generation", model="NewBreaker/chatglm-6b-int4", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("NewBreaker/chatglm-6b-int4", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NewBreaker/chatglm-6b-int4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NewBreaker/chatglm-6b-int4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NewBreaker/chatglm-6b-int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NewBreaker/chatglm-6b-int4

SGLang

How to use NewBreaker/chatglm-6b-int4 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 "NewBreaker/chatglm-6b-int4" \
    --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": "NewBreaker/chatglm-6b-int4",
		"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 "NewBreaker/chatglm-6b-int4" \
        --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": "NewBreaker/chatglm-6b-int4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use NewBreaker/chatglm-6b-int4 with Docker Model Runner:
```
docker model run hf.co/NewBreaker/chatglm-6b-int4
```

NewBreaker commited on May 1, 2023

Commit

9768cd8

1 Parent(s): 3f712ba

auto git

Browse files

Files changed (3) hide show

config.json +0 -1
demo_pipeline.py +1 -1
load_model.py +8 -7

config.json CHANGED Viewed

@@ -8,7 +8,6 @@
     "AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration",
     "AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration"
   },
-  "device": "cuda",
   "bos_token_id": 130004,
   "eos_token_id": 130005,
   "mask_token_id": 130000,

     "AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration",
     "AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration"
   },
   "bos_token_id": 130004,
   "eos_token_id": 130005,
   "mask_token_id": 130000,

demo_pipeline.py CHANGED Viewed

@@ -1,4 +1,4 @@
 from transformers import pipeline
-nlp = pipeline('text2text-generation',model ='.\\',trust_remote_code=True)
 # response = nlp

 from transformers import pipeline
+nlp = pipeline('text2text-generation',model ='THUDM/chatglm-6b',trust_remote_code=True)
 # response = nlp

load_model.py CHANGED Viewed

@@ -1,9 +1,10 @@
-from transformers import AutoTokenizer, AutoModel
 #
-tokenizer = AutoTokenizer.from_pretrained(".\\", trust_remote_code=True)
-# model = AutoModel.from_pretrained(".\\", trust_remote_code=True).float()
-model = AutoModel.from_pretrained(".\\", trust_remote_code=True)
-model = model.eval()
-response, history = model.chat(tokenizer, "你好", history=[])
-print("response:", response)

+from transformers import AutoTokenizer, AutoModel,pipeline
 #
+# tokenizer = AutoTokenizer.from_pretrained(".\\", trust_remote_code=True)
+# # model = AutoModel.from_pretrained(".\\", trust_remote_code=True).float()
+# model = AutoModel.from_pretrained(".\\", trust_remote_code=True)
+# model = model.eval()
+# response, history = model.chat(tokenizer, "你好", history=[])
+# print("response:", response)
+npl = pipeline('')