Instructions to use Fizzarolli/L3.1-70b-glitz-v0.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Fizzarolli/L3.1-70b-glitz-v0.2 with Transformers:

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

pipe = pipeline("text-generation", model="Fizzarolli/L3.1-70b-glitz-v0.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Fizzarolli/L3.1-70b-glitz-v0.2")
model = AutoModelForCausalLM.from_pretrained("Fizzarolli/L3.1-70b-glitz-v0.2", 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 Fizzarolli/L3.1-70b-glitz-v0.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Fizzarolli/L3.1-70b-glitz-v0.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Fizzarolli/L3.1-70b-glitz-v0.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Fizzarolli/L3.1-70b-glitz-v0.2

SGLang

How to use Fizzarolli/L3.1-70b-glitz-v0.2 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 "Fizzarolli/L3.1-70b-glitz-v0.2" \
    --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": "Fizzarolli/L3.1-70b-glitz-v0.2",
		"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 "Fizzarolli/L3.1-70b-glitz-v0.2" \
        --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": "Fizzarolli/L3.1-70b-glitz-v0.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Fizzarolli/L3.1-70b-glitz-v0.2 with Docker Model Runner:
```
docker model run hf.co/Fizzarolli/L3.1-70b-glitz-v0.2
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Glitz v0.2 ✨💅

art by spindlehorse toons. i do not own this image. all credit goes to them!

compute sponsored by ShuttleAI

details

this is an experimental l3.1 70b finetuning run... that crashed midway through. however, the results are still interesting, so i wanted to publish them :3

prompting

use l3 instruct. not writing an example, almost everything supports it by now and i don't feel like remembering the stupid tags

datasets

a mix of publicly available claude synth data in various domains, mostly. also systemchat but i refuse to credit cognitive computations for anything decent so forget i said it

Downloads last month: 17

Safetensors

Model size

71B params

Tensor type

BF16

Model tree for Fizzarolli/L3.1-70b-glitz-v0.2

Merges

5 models

Quantizations

2 models