Instructions to use eren23/ogno-monarch-jaskier-merge-7b-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use eren23/ogno-monarch-jaskier-merge-7b-v2 with Transformers:

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

pipe = pipeline("text-generation", model="eren23/ogno-monarch-jaskier-merge-7b-v2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("eren23/ogno-monarch-jaskier-merge-7b-v2")
model = AutoModelForCausalLM.from_pretrained("eren23/ogno-monarch-jaskier-merge-7b-v2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use eren23/ogno-monarch-jaskier-merge-7b-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "eren23/ogno-monarch-jaskier-merge-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": "eren23/ogno-monarch-jaskier-merge-7b-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/eren23/ogno-monarch-jaskier-merge-7b-v2

SGLang

How to use eren23/ogno-monarch-jaskier-merge-7b-v2 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 "eren23/ogno-monarch-jaskier-merge-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": "eren23/ogno-monarch-jaskier-merge-7b-v2",
		"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 "eren23/ogno-monarch-jaskier-merge-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": "eren23/ogno-monarch-jaskier-merge-7b-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use eren23/ogno-monarch-jaskier-merge-7b-v2 with Docker Model Runner:
```
docker model run hf.co/eren23/ogno-monarch-jaskier-merge-7b-v2
```

Good job

by supercharge19 - opened Mar 7, 2024

Discussion

supercharge19

Mar 7, 2024

Nice work, however, I have seen that mistral instruct v0.2 is better than simple mistral instruct, and mistral instruct (simple) is based on mistral 7b (pretrained), so models based on v0.2 are generally better. Do you think your merges have (or merge tree anywhere) has v0.2 in it or they all are simple instruct model?

eren23

Owner Mar 8, 2024

hey, I think the owner of the original models I used had the merge tree somewhere but I couldn't find it, you might want to check that out.

supercharge19

Mar 8, 2024

I tried looking for the tree but I could not find the complete tree, on twitter only small portion of tree was there (before latest merges) but that is ok, no need to look further, if this works for me I will report back happily otherwise it is too much to dig in it.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment