Instructions to use lgaalves/llama-2-13b-chat-platypus with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lgaalves/llama-2-13b-chat-platypus with Transformers:

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

pipe = pipeline("text-generation", model="lgaalves/llama-2-13b-chat-platypus")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lgaalves/llama-2-13b-chat-platypus")
model = AutoModelForCausalLM.from_pretrained("lgaalves/llama-2-13b-chat-platypus", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lgaalves/llama-2-13b-chat-platypus with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lgaalves/llama-2-13b-chat-platypus"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lgaalves/llama-2-13b-chat-platypus",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lgaalves/llama-2-13b-chat-platypus

SGLang

How to use lgaalves/llama-2-13b-chat-platypus 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 "lgaalves/llama-2-13b-chat-platypus" \
    --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": "lgaalves/llama-2-13b-chat-platypus",
		"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 "lgaalves/llama-2-13b-chat-platypus" \
        --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": "lgaalves/llama-2-13b-chat-platypus",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use lgaalves/llama-2-13b-chat-platypus with Docker Model Runner:
```
docker model run hf.co/lgaalves/llama-2-13b-chat-platypus
```

lgaalves commited on Sep 12, 2023

Commit

6170e7f

1 Parent(s): 828aa10

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -6

README.md CHANGED Viewed

@@ -19,13 +19,13 @@ language:
 | Metric                | llama-2-13b-chat-platypus | garage-bAInd/Platypus2-13B| llama-2-13b-chat-hf  (base) |
 |-----------------------|-------|-------|-------|
-| Avg.                  | -|61.35| 59.93 |
-| ARC (25-shot)         | -|61.26| 59.04 |
-| HellaSwag (10-shot)   | -|82.56| 81.94 |
-| MMLU (5-shot)         | -|56.7| 54.64 |
-| TruthfulQA (0-shot)   | -|44.86| 44.12 |
 We use state-of-the-art [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) to run the benchmark tests above, using the same version as the HuggingFace LLM Leaderboard. Please see below for detailed instructions on reproducing benchmark results.
 ### Model Details

 | Metric                | llama-2-13b-chat-platypus | garage-bAInd/Platypus2-13B| llama-2-13b-chat-hf  (base) |
 |-----------------------|-------|-------|-------|
+| Avg.                  | 58.8 |**61.35**| 59.93 |
+| ARC (25-shot)         | 53.84|**61.26**| 59.04 |
+| HellaSwag (10-shot)   | 80.67|**82.56**| 81.94 |
+| MMLU (5-shot)         | 54.44|**56.7**| 54.64 |
+| TruthfulQA (0-shot)   | **46.23**|44.86| 44.12 |
 We use state-of-the-art [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) to run the benchmark tests above, using the same version as the HuggingFace LLM Leaderboard. Please see below for detailed instructions on reproducing benchmark results.
 ### Model Details