Instructions to use ohyeah1/Violet-Lyra-Gutenberg-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ohyeah1/Violet-Lyra-Gutenberg-v2 with Transformers:

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

pipe = pipeline("text-generation", model="ohyeah1/Violet-Lyra-Gutenberg-v2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ohyeah1/Violet-Lyra-Gutenberg-v2")
model = AutoModelForCausalLM.from_pretrained("ohyeah1/Violet-Lyra-Gutenberg-v2")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ohyeah1/Violet-Lyra-Gutenberg-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ohyeah1/Violet-Lyra-Gutenberg-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ohyeah1/Violet-Lyra-Gutenberg-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ohyeah1/Violet-Lyra-Gutenberg-v2

SGLang

How to use ohyeah1/Violet-Lyra-Gutenberg-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 "ohyeah1/Violet-Lyra-Gutenberg-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": "ohyeah1/Violet-Lyra-Gutenberg-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 "ohyeah1/Violet-Lyra-Gutenberg-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": "ohyeah1/Violet-Lyra-Gutenberg-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ohyeah1/Violet-Lyra-Gutenberg-v2 with Docker Model Runner:
```
docker model run hf.co/ohyeah1/Violet-Lyra-Gutenberg-v2
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Violet-Lyra-Gutenberg-v2-12b

Update: Actually I do not really like its prose. It is more stable, though. Interestingly it scored amazing on UGI leaderboard. I do think this model is really smart though.

Prompt format:

ChatML

models:
  - model: redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS
    parameters:
      weight: 0.3
  - model: Nitral-AI/Captain_Eris_Noctis-12B-v0.420
    parameters:
      weight: 0.3
  - model: ohyeah1/Violet-Lyra-Gutenberg
    parameters:
      weight: 0.6
base_model: mistralai/Mistral-Nemo-Base-2407
parameters:
  density: 0.5
  epsilon: 0.1
  lambda: 1.0
  normalize: false
  rescale: true
merge_method: della_linear
tokenizer:
  source: union
dtype: bfloat16

Downloads last month: 15

Safetensors

Model size

12B params

Tensor type

BF16

Model tree for ohyeah1/Violet-Lyra-Gutenberg-v2

Merges

13 models

Quantizations

3 models