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

Libraries

How to use minpeter/tiny-ko-sft with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("minpeter/tiny-ko-sft")
model = AutoModelForCausalLM.from_pretrained("minpeter/tiny-ko-sft")
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 minpeter/tiny-ko-sft with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/minpeter/tiny-ko-sft

SGLang

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

Docker Model Runner
How to use minpeter/tiny-ko-sft with Docker Model Runner:
```
docker model run hf.co/minpeter/tiny-ko-sft
```

minpeter commited on Jun 3, 2025

Commit

2eb873e

verified ·

1 Parent(s): 6abd0b9

End of training

Browse files

Files changed (1) hide show

README.md +16 -9

README.md CHANGED Viewed

@@ -6,6 +6,7 @@ tags:
 - generated_from_trainer
 datasets:
 - lemon-mint/Korean-FineTome-100k
 model-index:
 - name: ko-tiny-exp
   results: []
@@ -30,6 +31,13 @@ datasets:
     message_property_mappings:
       role: role
       content: content
 dataset_prepared_path: last_run_prepared
 val_set_size: 0.05
@@ -45,7 +53,6 @@ pad_to_sequence_len: true
 gradient_accumulation_steps: 4
 micro_batch_size: 16
-num_epochs: 2
 optimizer: paged_adamw_8bit
 lr_scheduler: cosine
 learning_rate: 2e-5
@@ -61,6 +68,7 @@ logging_steps: 1
 flash_attention: true
 warmup_steps: 100
 evals_per_epoch: 2
 saves_per_epoch: 1
 weight_decay: 0.0
@@ -71,9 +79,9 @@ weight_decay: 0.0
 # ko-tiny-exp
-This model is a fine-tuned version of [minpeter/pretrained-tiny-ko](https://huggingface.co/minpeter/pretrained-tiny-ko) on the lemon-mint/Korean-FineTome-100k dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.9471
 ## Model description
@@ -101,17 +109,16 @@ The following hyperparameters were used during training:
 - optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT 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: 100
-- training_steps: 176
 ### Training results
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 3.0538        | 0.0113 | 1    | 3.0136          |
-| 2.9454        | 0.4972 | 44   | 3.0099          |
-| 2.9657        | 0.9944 | 88   | 2.9874          |
-| 3.0426        | 1.4859 | 132  | 2.9506          |
-| 2.9403        | 1.9831 | 176  | 2.9471          |
 ### Framework versions

 - generated_from_trainer
 datasets:
 - lemon-mint/Korean-FineTome-100k
+- lemon-mint/smol-koreantalk
 model-index:
 - name: ko-tiny-exp
   results: []
     message_property_mappings:
       role: role
       content: content
+  - path: lemon-mint/smol-koreantalk
+    type: chat_template
+    split: train[:20%]
+    field_messages: messages
+    message_property_mappings:
+      role: role
+      content: content
 dataset_prepared_path: last_run_prepared
 val_set_size: 0.05
 gradient_accumulation_steps: 4
 micro_batch_size: 16
 optimizer: paged_adamw_8bit
 lr_scheduler: cosine
 learning_rate: 2e-5
 flash_attention: true
 warmup_steps: 100
+num_epochs: 2
 evals_per_epoch: 2
 saves_per_epoch: 1
 weight_decay: 0.0
 # ko-tiny-exp
+This model is a fine-tuned version of [minpeter/pretrained-tiny-ko](https://huggingface.co/minpeter/pretrained-tiny-ko) on the lemon-mint/Korean-FineTome-100k and the lemon-mint/smol-koreantalk datasets.
 It achieves the following results on the evaluation set:
+- Loss: 2.8226
 ## Model description
 - optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT 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: 100
+- training_steps: 1498
 ### Training results
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
+| 3.0001        | 0.0013 | 1    | 2.9904          |
+| 2.8288        | 0.5002 | 375  | 2.8669          |
+| 2.8188        | 1.0    | 750  | 2.8255          |
+| 2.8012        | 1.5002 | 1125 | 2.8226          |
 ### Framework versions