Instructions to use csdc-atl/buffer-baichuan2-13B-rag-4bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use csdc-atl/buffer-baichuan2-13B-rag-4bits with Transformers:

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

pipe = pipeline("text-generation", model="csdc-atl/buffer-baichuan2-13B-rag-4bits", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("csdc-atl/buffer-baichuan2-13B-rag-4bits", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use csdc-atl/buffer-baichuan2-13B-rag-4bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "csdc-atl/buffer-baichuan2-13B-rag-4bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "csdc-atl/buffer-baichuan2-13B-rag-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/csdc-atl/buffer-baichuan2-13B-rag-4bits

SGLang

How to use csdc-atl/buffer-baichuan2-13B-rag-4bits 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 "csdc-atl/buffer-baichuan2-13B-rag-4bits" \
    --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": "csdc-atl/buffer-baichuan2-13B-rag-4bits",
		"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 "csdc-atl/buffer-baichuan2-13B-rag-4bits" \
        --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": "csdc-atl/buffer-baichuan2-13B-rag-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use csdc-atl/buffer-baichuan2-13B-rag-4bits with Docker Model Runner:
```
docker model run hf.co/csdc-atl/buffer-baichuan2-13B-rag-4bits
```

Xiaoting He commited on Dec 5, 2023

Commit

3c8bbd7

1 Parent(s): 0b8561b

4bits

Browse files

Files changed (3) hide show

config.json +1 -1
quantize_config.json +1 -1
tokenization_baichuan.py +5 -6

config.json CHANGED Viewed

@@ -35,6 +35,6 @@
       "sym": true,
       "true_sequential": true,
       "model_name_or_path": null,
-      "model_file_base_name": null
   }
 }

       "sym": true,
       "true_sequential": true,
       "model_name_or_path": null,
+      "model_file_base_name": "model"
   }
 }

quantize_config.json CHANGED Viewed

@@ -7,5 +7,5 @@
   "sym": true,
   "true_sequential": true,
   "model_name_or_path": null,
-  "model_file_base_name": null
 }

   "sym": true,
   "true_sequential": true,
   "model_name_or_path": null,
+  "model_file_base_name": "model"
 }

tokenization_baichuan.py CHANGED Viewed

@@ -68,6 +68,11 @@ class BaichuanTokenizer(PreTrainedTokenizer):
             if isinstance(pad_token, str)
             else pad_token
         )
         super().__init__(
             bos_token=bos_token,
             eos_token=eos_token,
@@ -79,12 +84,6 @@ class BaichuanTokenizer(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.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
-        self.sp_model.Load(vocab_file)
     def __getstate__(self):
         state = self.__dict__.copy()
         state["sp_model"] = None

             if isinstance(pad_token, str)
             else pad_token
         )
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
         super().__init__(
             bos_token=bos_token,
             eos_token=eos_token,
             clean_up_tokenization_spaces=clean_up_tokenization_spaces,
             **kwargs,
         )
     def __getstate__(self):
         state = self.__dict__.copy()
         state["sp_model"] = None