Instructions to use llmware/slim-summary-phi-3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use llmware/slim-summary-phi-3 with Transformers:

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

pipe = pipeline("text-generation", model="llmware/slim-summary-phi-3", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("llmware/slim-summary-phi-3", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("llmware/slim-summary-phi-3", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use llmware/slim-summary-phi-3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "llmware/slim-summary-phi-3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "llmware/slim-summary-phi-3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/llmware/slim-summary-phi-3

SGLang

How to use llmware/slim-summary-phi-3 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 "llmware/slim-summary-phi-3" \
    --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": "llmware/slim-summary-phi-3",
		"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 "llmware/slim-summary-phi-3" \
        --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": "llmware/slim-summary-phi-3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use llmware/slim-summary-phi-3 with Docker Model Runner:
```
docker model run hf.co/llmware/slim-summary-phi-3
```

doberst commited on Oct 9, 2024

Commit

5ec08a9

verified ·

1 Parent(s): d71088d

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -10

README.md CHANGED Viewed

@@ -1,27 +1,23 @@
 ---
 license: apache-2.0
 inference: false
-tags: [green, p1, llmware-fx, ov, emerald]
 ---
-# slim-summary-tiny-ov
-**slim-summary-tiny-ov** is a specialized function calling model that summarizes a given text and generates as output a Python list of summary points.
-This is an OpenVino int4 quantized version of slim-summary-tiny, providing a very fast, very small inference implementation, optimized for AI PCs using Intel GPU, CPU and NPU.
 ### Model Description
 - **Developed by:** llmware
-- **Model type:** tinyllama
-- **Parameters:** 1.1 billion
-- **Model Parent:** llmware/slim-summary-tiny
 - **Language(s) (NLP):** English
 - **License:** Apache 2.0
 - **Uses:** Summary bulletpoints extracted from complex business documents
-- **RAG Benchmark Accuracy Score:** NA
-- **Quantization:** int4
 ## Model Card Contact

 ---
 license: apache-2.0
 inference: false
 ---
+# slim-summary-phi-3
+**slim-summary-phi-3** is a specialized function calling model that summarizes a given text and generates as output a Python list of summary points.
+This is the base Pytorch version of the model, useful for further fine-tuning.  For faster inference, we would recommend using either the GGUF or OpenVino version of the model.
 ### Model Description
 - **Developed by:** llmware
+- **Model type:** phi-3
+- **Parameters:** 3.8 billion
+- **Model Parent:** llmware/slim-summary-phi-3
 - **Language(s) (NLP):** English
 - **License:** Apache 2.0
 - **Uses:** Summary bulletpoints extracted from complex business documents
 ## Model Card Contact