Instructions to use YeungNLP/firefly-mixtral-8x7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use YeungNLP/firefly-mixtral-8x7b with Transformers:

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

pipe = pipeline("text-generation", model="YeungNLP/firefly-mixtral-8x7b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("YeungNLP/firefly-mixtral-8x7b")
model = AutoModelForCausalLM.from_pretrained("YeungNLP/firefly-mixtral-8x7b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use YeungNLP/firefly-mixtral-8x7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "YeungNLP/firefly-mixtral-8x7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "YeungNLP/firefly-mixtral-8x7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/YeungNLP/firefly-mixtral-8x7b

SGLang

How to use YeungNLP/firefly-mixtral-8x7b 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 "YeungNLP/firefly-mixtral-8x7b" \
    --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": "YeungNLP/firefly-mixtral-8x7b",
		"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 "YeungNLP/firefly-mixtral-8x7b" \
        --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": "YeungNLP/firefly-mixtral-8x7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use YeungNLP/firefly-mixtral-8x7b with Docker Model Runner:
```
docker model run hf.co/YeungNLP/firefly-mixtral-8x7b
```

YeungNLP commited on Dec 24, 2023

Commit

cccf0fe

1 Parent(s): 30abf8d

Update README.md

Browse files

Files changed (1) hide show

README.md +16 -1

README.md CHANGED Viewed

@@ -4,7 +4,22 @@ language:
 - en
 ---
-This model is finetuend on "mistralai/Mixtral-8x7B-v0.1" with Firefly
 ## Run the model

 - en
 ---
+This model is finetuend based on "mistralai/Mixtral-8x7B-v0.1" with [Firefly](https://github.com/yangjianxin1/Firefly) and 48k data from ultrachat.
+## Evaluation
+Though we finetune with only 48k data, our model can also achieve excellent performance.
+| Model                                                                                        | Open LLM Leaderboard         |
+|------------------------------------------------------------------------------------------------|---------------------------------------------|
+| Qwen-72B  | 73.6  |
+| Mixtral-8x7B-Instruct-v0.1 |        72.62                                   |
+|**Firefly-Mixtral-8x7B**|**70.34**|
+|Yi-34B|69.42|
+|Mixtral-8x7B-v0.1|68.42|
+|Llama2-65B-Chat|67.87|
+|Qwen-14B|65.86|
+|Vicuna-33B-v1.3 |58.54|
 ## Run the model