Instructions to use BAAI/AquilaMoE-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BAAI/AquilaMoE-SFT with Transformers:

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

pipe = pipeline("text-generation", model="BAAI/AquilaMoE-SFT", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("BAAI/AquilaMoE-SFT", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use BAAI/AquilaMoE-SFT with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/BAAI/AquilaMoE-SFT

SGLang

How to use BAAI/AquilaMoE-SFT 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 "BAAI/AquilaMoE-SFT" \
    --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": "BAAI/AquilaMoE-SFT",
		"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 "BAAI/AquilaMoE-SFT" \
        --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": "BAAI/AquilaMoE-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use BAAI/AquilaMoE-SFT with Docker Model Runner:
```
docker model run hf.co/BAAI/AquilaMoE-SFT
```

ldwang commited on Jun 15, 2024

Commit

03ccd11

verified ·

1 Parent(s): 0918388

Update README.md

Browse files

Files changed (1) hide show

README.md +12 -8

README.md CHANGED Viewed

@@ -1,11 +1,11 @@
----
-license: apache-2.0
-language:
-- en
-- zh
-tags:
-- moe
----
 # AquilaMoE: Efficient Training for MoE Models with Scale-Up and Scale-Out Strategies
 <p align="center">
     <br>
@@ -142,6 +142,10 @@ The performance of the AquilaMoE model series improves significantly across mult
 | mmlu-ppl         | 59.93         |
 | winograd-ppl     | 57.5          |
 *Table: Performance of AquilaMoE-SFT (16\*8B) on various benchmarks.*
 ## License Agreement

+---
+license: apache-2.0
+language:
+- en
+- zh
+tags:
+- moe
+---
 # AquilaMoE: Efficient Training for MoE Models with Scale-Up and Scale-Out Strategies
 <p align="center">
     <br>
 | mmlu-ppl         | 59.93         |
 | winograd-ppl     | 57.5          |
+| Model            | GPT 3.5 Turbo (11/06) | GPT 3.5 Turbo (03/01) | AquilaMoE-SFT |
+|------------------|-----------------------|-----------------------|---------------|
+| AlpacaEval 2.0   | 19.3                  | 18.1                  | 21.1          |
 *Table: Performance of AquilaMoE-SFT (16\*8B) on various benchmarks.*
 ## License Agreement