Instructions to use microsoft/Phi-3-small-128k-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/Phi-3-small-128k-instruct with Transformers:

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

pipe = pipeline("text-generation", model="microsoft/Phi-3-small-128k-instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-small-128k-instruct", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use microsoft/Phi-3-small-128k-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "microsoft/Phi-3-small-128k-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/Phi-3-small-128k-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/microsoft/Phi-3-small-128k-instruct

SGLang

How to use microsoft/Phi-3-small-128k-instruct 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 "microsoft/Phi-3-small-128k-instruct" \
    --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": "microsoft/Phi-3-small-128k-instruct",
		"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 "microsoft/Phi-3-small-128k-instruct" \
        --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": "microsoft/Phi-3-small-128k-instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use microsoft/Phi-3-small-128k-instruct with Docker Model Runner:
```
docker model run hf.co/microsoft/Phi-3-small-128k-instruct
```

Move flash_attn assert from init into calling func

#32

by rogerxfeng8 - opened Sep 4, 2024

base: refs/heads/main

←

from: refs/pr/32

Discussion Files changed

-1

Files changed (1) hide show

modeling_phi3_small.py +2 -1

modeling_phi3_small.py CHANGED Viewed

@@ -215,7 +215,6 @@ class Phi3SmallSelfAttention(nn.Module):
                 f"Layer {layer_idx + 1} is using dense attention since it is divisible by "
                 f"{self.config.dense_attention_every_n_layers}"
             )
-            assert is_flash_attention_available, "Flash Attention is not available, but is needed for dense attention"
         else:
             # BlockSparse related Parameters
             self.blocksparse_params = BlockSparseParams.from_config(config)
@@ -419,6 +418,8 @@ class Phi3SmallSelfAttention(nn.Module):
             avoid doing that.
         """
         attention_dropout_prob = self.attention_dropout_rate if self.training else 0.0
         # Get into the correct shape for the Flash Attention API
         # shape: (bs, seq_len, nqp, hn)

                 f"Layer {layer_idx + 1} is using dense attention since it is divisible by "
                 f"{self.config.dense_attention_every_n_layers}"
             )
         else:
             # BlockSparse related Parameters
             self.blocksparse_params = BlockSparseParams.from_config(config)
             avoid doing that.
         """
+        assert is_flash_attention_available, "Flash Attention is not available, but is needed for dense attention"
         attention_dropout_prob = self.attention_dropout_rate if self.training else 0.0
         # Get into the correct shape for the Flash Attention API
         # shape: (bs, seq_len, nqp, hn)

Move flash_attn assert from __init__ into calling func

Move flash_attn assert from init into calling func