Instructions to use k050506koch/GPT3-dev-350m-2805 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use k050506koch/GPT3-dev-350m-2805 with Transformers:

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

pipe = pipeline("text-generation", model="k050506koch/GPT3-dev-350m-2805", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("k050506koch/GPT3-dev-350m-2805", trust_remote_code=True, dtype="auto")

llama-cpp-python

How to use k050506koch/GPT3-dev-350m-2805 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="k050506koch/GPT3-dev-350m-2805",
	filename="model_medium2805-Q2_K.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use k050506koch/GPT3-dev-350m-2805 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf k050506koch/GPT3-dev-350m-2805:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf k050506koch/GPT3-dev-350m-2805:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf k050506koch/GPT3-dev-350m-2805:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf k050506koch/GPT3-dev-350m-2805:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf k050506koch/GPT3-dev-350m-2805:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf k050506koch/GPT3-dev-350m-2805:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf k050506koch/GPT3-dev-350m-2805:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf k050506koch/GPT3-dev-350m-2805:Q4_K_M

Use Docker

docker model run hf.co/k050506koch/GPT3-dev-350m-2805:Q4_K_M

LM Studio
Jan

vLLM

How to use k050506koch/GPT3-dev-350m-2805 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "k050506koch/GPT3-dev-350m-2805"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "k050506koch/GPT3-dev-350m-2805",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/k050506koch/GPT3-dev-350m-2805:Q4_K_M

SGLang

How to use k050506koch/GPT3-dev-350m-2805 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 "k050506koch/GPT3-dev-350m-2805" \
    --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": "k050506koch/GPT3-dev-350m-2805",
		"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 "k050506koch/GPT3-dev-350m-2805" \
        --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": "k050506koch/GPT3-dev-350m-2805",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use k050506koch/GPT3-dev-350m-2805 with Ollama:
```
ollama run hf.co/k050506koch/GPT3-dev-350m-2805:Q4_K_M
```

Unsloth Studio

How to use k050506koch/GPT3-dev-350m-2805 with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for k050506koch/GPT3-dev-350m-2805 to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for k050506koch/GPT3-dev-350m-2805 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for k050506koch/GPT3-dev-350m-2805 to start chatting

Docker Model Runner
How to use k050506koch/GPT3-dev-350m-2805 with Docker Model Runner:
```
docker model run hf.co/k050506koch/GPT3-dev-350m-2805:Q4_K_M
```

Lemonade

How to use k050506koch/GPT3-dev-350m-2805 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull k050506koch/GPT3-dev-350m-2805:Q4_K_M

Run and chat with the model

lemonade run user.GPT3-dev-350m-2805-Q4_K_M

List all available models

lemonade list

GPT3

Welcome to the GPT3 repository! This project is an attempt to recreate the architecture and approach from the original OpenAI GPT-3 paper. The repository includes scripts for training, fine-tuning, and inference of a GPT-3-like model using PyTorch and the Hugging Face Transformers library. Here are located weights of dev checkpoints of my models. You can always download a folder, paste it's path inside inference.py and chat with them.

You can find all code on GitHub

This is a model with 350 million parameters (attempt to replicate GPT-3 Medium). (it's very undertrained.)

Note 2: This is a model checkpoint released on 28th of May 2025 (192 batch size, 512 tokens and 10000 steps under Lion optimizer). It scores 28.55% on MMLU which is slightly higher than 25% (random guess)

Note 3: This model already demonstrates basic abilities in generating text. It's not perfect and I will continue working on it. Expect Instruct models soon.

inference:

from transformers import AutoTokenizer, AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained('k050506koch/GPT3-dev-350m-2805', trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained('k050506koch/GPT3-dev-350m-2805')
tokenizer.pad_token_id = tokenizer.eos_token_id
print("\n", tokenizer.decode(model.generate(tokenizer.encode("He is a doctor. His main goal is", return_tensors='pt'),
    max_length=128, temperature=0.7, top_p=0.9, repetition_penalty=1.2, no_repeat_ngram_size=3,
    num_return_sequences=1, do_sample=True)[0], skip_special_tokens=True))

Contributing

Contributions are welcome! I'm just a student who is interested in AI so my code may be incorrect or have logical issues. Please open an issue or submit a pull request for any improvements or bug fixes, I will be happy.

License

This project is licensed under the MIT License. See the LICENSE file for details. Everyone can use and modify this code at their discretion.

Acknowledgements

Thanks OpenAI, HuggingFace and Pytorch for making this project possible!

Downloads last month: 543

Safetensors

Model size

0.4B params

Tensor type

BF16

Dataset used to train k050506koch/GPT3-dev-350m-2805

Spaces using k050506koch/GPT3-dev-350m-2805 2

Paper for k050506koch/GPT3-dev-350m-2805

Language Models are Few-Shot Learners

Paper • 2005.14165 • Published May 28, 2020 • 20