Instructions to use mazesmazes/tiny-turn with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mazesmazes/tiny-turn with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mazesmazes/tiny-turn")
model = AutoModelForCausalLM.from_pretrained("mazesmazes/tiny-turn")
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 Settings

vLLM

How to use mazesmazes/tiny-turn with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/mazesmazes/tiny-turn

SGLang

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

Docker Model Runner
How to use mazesmazes/tiny-turn with Docker Model Runner:
```
docker model run hf.co/mazesmazes/tiny-turn
```

tiny-turn / README.md

mazesmazes

Model save

1dc885b verified about 1 month ago

preview code

raw

history blame contribute delete

2.37 kB

	---
	library_name: transformers
	license: apache-2.0
	base_model: HuggingFaceTB/SmolLM2-135M-Instruct
	tags:
	- generated_from_trainer
	model-index:
	- name: tiny-turn
	results: []
	---

	<!-- This model card has been generated automatically according to the information the Trainer had access to. You
	should probably proofread and complete it, then remove this comment. -->

	# tiny-turn

	This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct) on the None dataset.
	It achieves the following results on the evaluation set:
	- Loss: 1.5122

	## Model description

	More information needed

	## Intended uses & limitations

	More information needed

	## Training and evaluation data

	More information needed

	## Training procedure

	### Training hyperparameters

	The following hyperparameters were used during training:
	- learning_rate: 5e-05
	- train_batch_size: 64
	- eval_batch_size: 64
	- seed: 42
	- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
	- lr_scheduler_type: cosine
	- lr_scheduler_warmup_steps: 500
	- num_epochs: 10

	### Training results

	\| Training Loss \| Epoch \| Step \| Validation Loss \|
	\|:-------------:\|:------:\|:-----:\|:---------------:\|
	\| 1.5988 \| 0.4819 \| 1000 \| 1.6036 \|
	\| 1.5668 \| 0.9639 \| 2000 \| 1.5508 \|
	\| 1.5337 \| 1.4458 \| 3000 \| 1.5326 \|
	\| 1.5210 \| 1.9277 \| 4000 \| 1.5227 \|
	\| 1.5078 \| 2.4096 \| 5000 \| 1.5176 \|
	\| 1.5247 \| 2.8916 \| 6000 \| 1.5157 \|
	\| 1.5034 \| 3.3735 \| 7000 \| 1.5140 \|
	\| 1.5273 \| 3.8554 \| 8000 \| 1.5137 \|
	\| 1.5209 \| 4.3373 \| 9000 \| 1.5127 \|
	\| 1.5108 \| 4.8193 \| 10000 \| 1.5123 \|
	\| 1.5134 \| 5.3012 \| 11000 \| 1.5123 \|
	\| 1.5131 \| 5.7831 \| 12000 \| 1.5124 \|
	\| 1.5197 \| 6.2651 \| 13000 \| 1.5122 \|
	\| 1.5254 \| 6.7470 \| 14000 \| 1.5122 \|
	\| 1.5139 \| 7.2289 \| 15000 \| 1.5120 \|
	\| 1.5068 \| 7.7108 \| 16000 \| 1.5122 \|
	\| 1.5128 \| 8.1928 \| 17000 \| 1.5121 \|
	\| 1.5168 \| 8.6747 \| 18000 \| 1.5122 \|


	### Framework versions

	- Transformers 5.5.4
	- Pytorch 2.8.0+cu128
	- Datasets 3.6.0
	- Tokenizers 0.22.2