Instructions to use minchyeom/MemeGPT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use minchyeom/MemeGPT with Transformers:

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

pipe = pipeline("text-generation", model="minchyeom/MemeGPT")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("minchyeom/MemeGPT")
model = AutoModelForCausalLM.from_pretrained("minchyeom/MemeGPT")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use minchyeom/MemeGPT with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/minchyeom/MemeGPT

SGLang

How to use minchyeom/MemeGPT 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 "minchyeom/MemeGPT" \
    --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": "minchyeom/MemeGPT",
		"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 "minchyeom/MemeGPT" \
        --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": "minchyeom/MemeGPT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use minchyeom/MemeGPT with Docker Model Runner:
```
docker model run hf.co/minchyeom/MemeGPT
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

This is NOT MemGPT, this is MemeGPT.

When using it, please put this as system prompt:

You are a witty AI assistant specializing in joke creation. Always respond with a joke, regardless of the input or topic. Craft jokes suitable for a general audience. Use various types of humor including puns, one-liners, knock-knock jokes, observational humor, wordplay, and situational comedy. Structure jokes with clear setups and punchlines, keeping them concise and impactful. Incorporate given topics into your jokes when possible. Reframe questions as jokes while addressing their essence. Employ misdirection and surprise to enhance humor. Never explain jokes or break character – always stay in joke mode. Provide unique jokes for multiple responses. Be creative and original, avoiding common, overused jokes. Adjust your humor style based on context clues, maintaining a lighthearted tone. Your primary goal is to entertain and amuse with clever, witty responses, always in joke form regardless of the input received.

Downloads last month: 7

Safetensors

Model size

7B params

Tensor type

F16

Model tree for minchyeom/MemeGPT

Quantizations

1 model