Instructions to use MugoSquero/LMCocktail-phi-2-v1.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MugoSquero/LMCocktail-phi-2-v1.1 with Transformers:

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

pipe = pipeline("text-generation", model="MugoSquero/LMCocktail-phi-2-v1.1", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("MugoSquero/LMCocktail-phi-2-v1.1", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MugoSquero/LMCocktail-phi-2-v1.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MugoSquero/LMCocktail-phi-2-v1.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MugoSquero/LMCocktail-phi-2-v1.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MugoSquero/LMCocktail-phi-2-v1.1

SGLang

How to use MugoSquero/LMCocktail-phi-2-v1.1 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 "MugoSquero/LMCocktail-phi-2-v1.1" \
    --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": "MugoSquero/LMCocktail-phi-2-v1.1",
		"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 "MugoSquero/LMCocktail-phi-2-v1.1" \
        --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": "MugoSquero/LMCocktail-phi-2-v1.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use MugoSquero/LMCocktail-phi-2-v1.1 with Docker Model Runner:
```
docker model run hf.co/MugoSquero/LMCocktail-phi-2-v1.1
```

LM-Cocktail phi-2 v1.1

This is a 0.5-0.5 merge of two models based on phi-2. Here are the models used to create this merge:

I named this model "LMCocktail phi-2 v1.1" because I see it as a continuation of the v1.

I used Yhyu13/phi-2-sft-dpo-gpt4_en-ep1 and it "outputs significantly longer result" than the one used in v1 by Yhyu13.

I also used venkycs/phi-2-instruct "a fine-tuned version of microsoft/phi-2 on the filtered ultrachat200k dataset using the SFT technique".

The main reason I created this model was to merge it with cognitivecomputations/dolphin-2_6-phi-2, and I will create a repo for it when I do it.

Code

The LM-cocktail is novel technique for merging multiple models: https://arxiv.org/abs/2311.13534

Code is backed up by this repo: https://github.com/FlagOpen/FlagEmbedding.git

Merging script is available under the ./scripts folder.

Downloads last month: 3

Safetensors

Model size

3B params

Tensor type

F32

Paper for MugoSquero/LMCocktail-phi-2-v1.1

LM-Cocktail: Resilient Tuning of Language Models via Model Merging

Paper • 2311.13534 • Published Nov 22, 2023 • 3