Instructions to use AxiomicLabs/GPT-X3-150M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AxiomicLabs/GPT-X3-150M with Transformers:

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

pipe = pipeline("text-generation", model="AxiomicLabs/GPT-X3-150M", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("AxiomicLabs/GPT-X3-150M", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AxiomicLabs/GPT-X3-150M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AxiomicLabs/GPT-X3-150M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AxiomicLabs/GPT-X3-150M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/AxiomicLabs/GPT-X3-150M

SGLang

How to use AxiomicLabs/GPT-X3-150M 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 "AxiomicLabs/GPT-X3-150M" \
    --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": "AxiomicLabs/GPT-X3-150M",
		"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 "AxiomicLabs/GPT-X3-150M" \
        --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": "AxiomicLabs/GPT-X3-150M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use AxiomicLabs/GPT-X3-150M with Docker Model Runner:
```
docker model run hf.co/AxiomicLabs/GPT-X3-150M
```

Question about the tokenizer

by FlameF0X - opened 10 days ago

Discussion

FlameF0X

10 days ago

Why is the BOS token removed?

Datdanboi25

Axiomic Labs org 9 days ago

Good question,
I did a bit of research into the impacts of a BOS, while useful as an attention sink, without it models learn to simply use the 1st token as said sink without any real loss penalty, thus it was decided to remove it

Datdanboi25 changed discussion status to closed 9 days ago

FlameF0X

9 days ago

Aah, yeah, makes sense

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