Instructions to use athirdpath/Hestia-20b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use athirdpath/Hestia-20b with Transformers:

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

pipe = pipeline("text-generation", model="athirdpath/Hestia-20b")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("athirdpath/Hestia-20b")
model = AutoModelForCausalLM.from_pretrained("athirdpath/Hestia-20b", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use athirdpath/Hestia-20b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "athirdpath/Hestia-20b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "athirdpath/Hestia-20b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/athirdpath/Hestia-20b

SGLang

How to use athirdpath/Hestia-20b 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 "athirdpath/Hestia-20b" \
    --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": "athirdpath/Hestia-20b",
		"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 "athirdpath/Hestia-20b" \
        --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": "athirdpath/Hestia-20b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use athirdpath/Hestia-20b with Docker Model Runner:
```
docker model run hf.co/athirdpath/Hestia-20b
```

This is a task_arithmetic merge of Harmonia (my 20b faux base model) with Noromaid and my LORA-glued Nethena. Solidly outperforms Harmonia.

merge_method: task_arithmetic

base_model: athirdpath/Harmonia-20b

models:

model: athirdpath/Harmonia-20b
model: NeverSleep/Noromaid-20b-v0.1.1
- parameters: weight: 0.25
model: athirdpath/Nethena-20b-Glued
- parameters: weight: 0.2

dtype: float16

Thanks to Undi95 for pioneering the 20B recipe, and for most of the models involved.

Downloads last month: 65

Safetensors

Model size

20B params

Tensor type

F16

Collection including athirdpath/Hestia-20b

Roleplay Deployment Models

Collection

Models I think are cool and worth using. A lot of what I make is intended only as a part in a further merge or as a test, these are the others. • 13 items • Updated Sep 3, 2024 • 3