garage-bAInd/Open-Platypus
Viewer โข Updated โข 24.9k โข 13.4k โข 432
How to use kyujinpy/SOLAR-Platypus-10.7B-v2 with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="kyujinpy/SOLAR-Platypus-10.7B-v2") # pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("kyujinpy/SOLAR-Platypus-10.7B-v2")
model = AutoModelForCausalLM.from_pretrained("kyujinpy/SOLAR-Platypus-10.7B-v2", device_map="auto")How to use kyujinpy/SOLAR-Platypus-10.7B-v2 with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kyujinpy/SOLAR-Platypus-10.7B-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "kyujinpy/SOLAR-Platypus-10.7B-v2",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/kyujinpy/SOLAR-Platypus-10.7B-v2
How to use kyujinpy/SOLAR-Platypus-10.7B-v2 with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "kyujinpy/SOLAR-Platypus-10.7B-v2" \
--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": "kyujinpy/SOLAR-Platypus-10.7B-v2",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'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 "kyujinpy/SOLAR-Platypus-10.7B-v2" \
--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": "kyujinpy/SOLAR-Platypus-10.7B-v2",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use kyujinpy/SOLAR-Platypus-10.7B-v2 with Docker Model Runner:
docker model run hf.co/kyujinpy/SOLAR-Platypus-10.7B-v2
Model Developers Kyujin Han (kyujinpy)
Input Models input text only.
Output Models generate text only.
Model Architecture
SOLAR-Platypus-10.7B-v2 is an auto-regressive language model based on the Llama2 architecture.
Base Model
upstage/SOLAR-10.7B-v1.0
Training Dataset
garage-bAInd/Open-Platypus.
While training, I used Q-LoRA.
The lora_r values is 64.
## Human:
## Assistant:
| Model | Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K |
|---|---|---|---|---|---|---|---|
| SOLAR-Platypus-10.7B-v1 | 58.62 | 61.69 | 84.23 | 60.37 | 51.58 | 82.79 | 11.07 |
| SOLAR-Platypus-10.7B-v2 | 55.25 | 59.39 | 83.57 | 59.93 | 43.15 | 81.45 | 4.02 |
| upstage/SOLAR-10.7B-v1.0 | 66.04 | 61.95 | 84.60 | 65.48 | 45.04 | 83.66 | 55.50 |
### KO-Platypus
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
repo = "kyujinpy/SOLAR-Platypus-10.7B-v2"
OpenOrca = AutoModelForCausalLM.from_pretrained(
repo,
return_dict=True,
torch_dtype=torch.float16,
device_map='auto'
)
OpenOrca_tokenizer = AutoTokenizer.from_pretrained(repo)