Instructions to use microsoft/DialoGPT-small with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/DialoGPT-small with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-small")
model = AutoModelForCausalLM.from_pretrained("microsoft/DialoGPT-small")
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 microsoft/DialoGPT-small with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/microsoft/DialoGPT-small

SGLang

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

Docker Model Runner
How to use microsoft/DialoGPT-small with Docker Model Runner:
```
docker model run hf.co/microsoft/DialoGPT-small
```

Bad response

#11

by RaitonRed - opened Dec 18, 2024

Discussion

RaitonRed

Dec 18, 2024

Hello, I am a developer.
I am building an AI website with Python that includes a chatbot.
I have downloaded and used all three models of DialoGPT: small, medium, and large.
However, all three models provide poor responses.
With the settings do_sample=false and seed 42.
Please help me quickly, I request!
You're welcome!

RaitonRed changed discussion status to closed Jan 12, 2025

RaitonRed changed discussion status to open Jan 12, 2025

neptun-org

Jan 27, 2025

•

edited Jan 27, 2025

This is a response I got from this model. lol?

JLouisBiz

Jan 27, 2025

•

edited Jan 27, 2025

I am using many small models, and did not use this one.

declare -A MODELS=(
    ["/home/data1/protected/Programming/LLM/QwQ-LCoT-3B-Instruct.Q4_K_M.gguf"]=999
    ["/home/data1/protected/Programming/LLM/Mistral/quantized/Ministral-3b-instruct-Q4_K_M.gguf"]=999
    ["/home/data1/protected/Programming/LLM/Mistral/quantized/Mistral-7B-v0.3-Q4_K_M.gguf"]=20
    ["/home/data1/protected/Programming/LLM/Microsoft/quantized/Phi-3.5-mini-instruct-Q3_K_M.gguf"]=30
    ["/home/data1/protected/Programming/LLM/Microsoft/quantized/Phi-3.5-mini-instruct-Q3_K_S.gguf"]=999
    ["/home/data1/protected/Programming/LLM/Qwen/quantized/Qwen2.5-1.5B-Instruct-Q4_K_M.gguf"]=999
    ["/home/data1/protected/Programming/LLM/Dolphin/quantized/Dolphin3.0-Qwen2.5-1.5B-Q5_K_M.gguf"]=999
    ["/home/data1/protected/Programming/LLM/Dolphin/quantized/Dolphin3.0-Qwen2.5-3B-Q5_K_M.gguf"]=999
    ["/home/data1/protected/Programming/LLM/AllenAI/quantized/olmo-2-1124-7B-instruct-Q2_K.gguf"]=24
    ["/home/data1/protected/Programming/LLM/AllenAI/quantized/OLMo-2-1124-7B-Instruct-Q3_K_M.gguf"]=18
    ["/home/data1/protected/Programming/LLM/AllenAI/quantized/olmo-2-1124-7B-instruct-Q3_K_S.gguf"]=21
    ["/home/data1/protected/Programming/LLM/SmolLM/quantized/SmolLM-1.7B-Instruct-Q5_K_M.gguf"]=999
    ["/home/data1/protected/Programming/LLM/DeepSeek/quantized/DeepSeek-R1-Distill-Qwen-1.5B-Q5_K_M.gguf"]=999
    ["/home/data1/protected/Programming/LLM/Qwen/quantized/DeepSeek-R1-Distill-Qwen-7B-Q3_K_S.gguf"]=999
    ["/home/data1/protected/Programming/LLM/Qwen/quantized/DeepSeek-R1-Distill-Qwen-7B-Q3_K_M.gguf"]=25
)

Among all those models I am using on my poor GTX 1050 Ti with 4 GB, I have never got such ridiculous answers.

I have got repetitions, and totally wrong information, but not logically wrong.

I can just recommend that you switch to better model, I can highly recommend OLMo models from Allen AI as first, plus others in the list above.

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