Instructions to use finis-est/L3.3-Faust-70B-exp.001 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use finis-est/L3.3-Faust-70B-exp.001 with Transformers:

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

pipe = pipeline("text-generation", model="finis-est/L3.3-Faust-70B-exp.001")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("finis-est/L3.3-Faust-70B-exp.001")
model = AutoModelForCausalLM.from_pretrained("finis-est/L3.3-Faust-70B-exp.001", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use finis-est/L3.3-Faust-70B-exp.001 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "finis-est/L3.3-Faust-70B-exp.001"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "finis-est/L3.3-Faust-70B-exp.001",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/finis-est/L3.3-Faust-70B-exp.001

SGLang

How to use finis-est/L3.3-Faust-70B-exp.001 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 "finis-est/L3.3-Faust-70B-exp.001" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "finis-est/L3.3-Faust-70B-exp.001",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "finis-est/L3.3-Faust-70B-exp.001" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "finis-est/L3.3-Faust-70B-exp.001",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use finis-est/L3.3-Faust-70B-exp.001 with Docker Model Runner:
```
docker model run hf.co/finis-est/L3.3-Faust-70B-exp.001
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

L3.3-Faust-70B-exp.001

This is a merge of pre-trained language models created using mergekit.

In case it wasn't obvious, I'm still experimenting. Hopefully the next few ones in this Faust series have actual reasoning in this section for how and why I did things, but for now it's just a silly merge by a noob.

I don't know what I'm doing yet - still reading and picking up on a lot of things, so I thought I'd give runpod and mergekit a quick shot.

Found some good and bad during testing. Will need to work on that first before continuing to merge. It was fun to learn how to merge and host this, however.

Merge Details

Merge Method

This model was merged using the Model Stock merge method using unsloth/Llama-3.3-70B-Instruct as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

base_model: unsloth/Llama-3.3-70B-Instruct
merge_method: model_stock
dtype: bfloat16
models:
  - model: EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
  - model: Sao10K/70B-L3.3-Cirrus-x1
  - model: TheDrummer/Anubis-70B-v1