Instructions to use TheBigBlender/Orca2myth7.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBigBlender/Orca2myth7.2 with Transformers:

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

pipe = pipeline("text-generation", model="TheBigBlender/Orca2myth7.2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TheBigBlender/Orca2myth7.2")
model = AutoModelForCausalLM.from_pretrained("TheBigBlender/Orca2myth7.2")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use TheBigBlender/Orca2myth7.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TheBigBlender/Orca2myth7.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBigBlender/Orca2myth7.2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TheBigBlender/Orca2myth7.2

SGLang

How to use TheBigBlender/Orca2myth7.2 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 "TheBigBlender/Orca2myth7.2" \
    --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": "TheBigBlender/Orca2myth7.2",
		"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 "TheBigBlender/Orca2myth7.2" \
        --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": "TheBigBlender/Orca2myth7.2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TheBigBlender/Orca2myth7.2 with Docker Model Runner:
```
docker model run hf.co/TheBigBlender/Orca2myth7.2
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

A product of an amateur merger. I like it due to the fact that it combines both Orca2 understanding and Pyg's dialogue style (using mythalion for consistency). - shotmisser64

This model was made by ShotMisser64 using the following mergekit yaml:

slices:
  - sources:
    - model: output/Orca2flat
      layer_range: [0, 13]
  - sources:
    - model: PygmalionAI/mythalion-13b
      layer_range: [3, 22]
  - sources:
    - model: output/Orca2flat
      layer_range: [14, 27]
  - sources:
    - model: PygmalionAI/mythalion-13b
      layer_range: [23, 40]
merge_method: passthrough
dtype: float16

The Orca2flat model uses the following mergekit yaml:

merge_method: task_arithmetic
base_model: TheBloke/Llama-2-13B-fp16
models:
  - model: TheBloke/Llama-2-13B-fp16
  - model: microsoft/Orca-2-13b
    parameters:
      weight: 1.0
dtype: float16

Found something interesting or would you like your own custom merge? Visit our community at https://koboldai.org/discord

Please respect the license of the origin models.

Downloads last month: 5

Safetensors

Model size

20B params

Tensor type

F16

Model tree for TheBigBlender/Orca2myth7.2

Quantizations

3 models