Instructions to use verseAI/databricks-dolly-v2-3b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use verseAI/databricks-dolly-v2-3b with Transformers:

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

pipe = pipeline("text-generation", model="verseAI/databricks-dolly-v2-3b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("verseAI/databricks-dolly-v2-3b")
model = AutoModelForCausalLM.from_pretrained("verseAI/databricks-dolly-v2-3b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use verseAI/databricks-dolly-v2-3b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "verseAI/databricks-dolly-v2-3b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "verseAI/databricks-dolly-v2-3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/verseAI/databricks-dolly-v2-3b

SGLang

How to use verseAI/databricks-dolly-v2-3b 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 "verseAI/databricks-dolly-v2-3b" \
    --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": "verseAI/databricks-dolly-v2-3b",
		"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 "verseAI/databricks-dolly-v2-3b" \
        --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": "verseAI/databricks-dolly-v2-3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use verseAI/databricks-dolly-v2-3b with Docker Model Runner:
```
docker model run hf.co/verseAI/databricks-dolly-v2-3b
```

m-vai commited on May 3, 2023

Commit

68cef6f

1 Parent(s): 3984963

adjust print statements

Browse files

Files changed (1) hide show

tester.py +9 -9

tester.py CHANGED Viewed

@@ -26,7 +26,7 @@ Following Sugarman’s death, the Embassy of the United States issued a statemen
     return s
 def mainSimple():
-    print("In main simple...")
     # print(pathlib.Path().resolve())
     # generate_text_pipline = pipeline(model="databricks/dolly-v2-3b", torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto")
     # generate_text_pipline = pipeline(model="verseAI/databricks-dolly-v2-3b", torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto")
@@ -36,10 +36,10 @@ def mainSimple():
     resp = generate_text_pipline(inputText)
     respStr = resp[0]["generated_text"]
-    print(f'Input: {inputText}')
-    print(f'Response: {respStr}')
-    print("All Done!")
 def getInstrAndContext():
     context = '''George Washington (February 22, 1732[b] - December 14, 1799) was an American military officer, statesman,
@@ -50,7 +50,7 @@ and Founding Father who served as the first president of the United States from
     return instr, context
 def mainContext():
-    print("In main context...")
     workingDir = pathlib.Path().resolve()
     generate_text_pipline = pipeline(model=workingDir, torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto", return_full_text=True)
@@ -77,11 +77,11 @@ def mainContext():
     else:
         resp = llm_chain.predict(instruction=instr).lstrip()
-    print(f'Input-Instr: {instr}')
-    print(f'Input-Context: {context}')
-    print(f'Response: {resp}')
-    print("All Done!")
 if __name__ == "__main__":
     mainContext()

     return s
 def mainSimple():
+    print("\nIn main simple...")
     # print(pathlib.Path().resolve())
     # generate_text_pipline = pipeline(model="databricks/dolly-v2-3b", torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto")
     # generate_text_pipline = pipeline(model="verseAI/databricks-dolly-v2-3b", torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto")
     resp = generate_text_pipline(inputText)
     respStr = resp[0]["generated_text"]
+    print(f'\nInput: {inputText}')
+    print(f'\nResponse: {respStr}')
+    print("\nAll Done!\n")
 def getInstrAndContext():
     context = '''George Washington (February 22, 1732[b] - December 14, 1799) was an American military officer, statesman,
     return instr, context
 def mainContext():
+    print("\nIn main context...")
     workingDir = pathlib.Path().resolve()
     generate_text_pipline = pipeline(model=workingDir, torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto", return_full_text=True)
     else:
         resp = llm_chain.predict(instruction=instr).lstrip()
+    print(f'\nInput-Context: {context}')
+    print(f'\nInput-Instr: {instr}')
+    print(f'\nResponse: {resp}')
+    print("\nAll Done!\n")
 if __name__ == "__main__":
     mainContext()