Instructions to use jeehwon/Qwen3-0.6B-Distill with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jeehwon/Qwen3-0.6B-Distill with Transformers:

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

pipe = pipeline("text-generation", model="jeehwon/Qwen3-0.6B-Distill")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jeehwon/Qwen3-0.6B-Distill")
model = AutoModelForCausalLM.from_pretrained("jeehwon/Qwen3-0.6B-Distill")
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 jeehwon/Qwen3-0.6B-Distill with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jeehwon/Qwen3-0.6B-Distill"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jeehwon/Qwen3-0.6B-Distill",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jeehwon/Qwen3-0.6B-Distill

SGLang

How to use jeehwon/Qwen3-0.6B-Distill 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 "jeehwon/Qwen3-0.6B-Distill" \
    --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": "jeehwon/Qwen3-0.6B-Distill",
		"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 "jeehwon/Qwen3-0.6B-Distill" \
        --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": "jeehwon/Qwen3-0.6B-Distill",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jeehwon/Qwen3-0.6B-Distill with Docker Model Runner:
```
docker model run hf.co/jeehwon/Qwen3-0.6B-Distill
```

jeehwon commited on May 12, 2025

Commit

725d71f

verified ·

1 Parent(s): 0423eb5

Model save

Browse files

Files changed (6) hide show

README.md +2 -4
all_results.json +4 -4
config.json +1 -1
train_results.json +4 -4
trainer_state.json +4 -5
training_args.bin +1 -1

README.md CHANGED Viewed

@@ -1,11 +1,9 @@
 ---
 base_model: Qwen/Qwen3-0.6B
-datasets: dnotitia/Magpie_Claude35_no_think_1k_samples
 library_name: transformers
 model_name: Qwen3-0.6B-Distill
 tags:
 - generated_from_trainer
-- open-r1
 - trl
 - sft
 licence: license
@@ -13,7 +11,7 @@ licence: license
 # Model Card for Qwen3-0.6B-Distill
-This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B) on the [dnotitia/Magpie_Claude35_no_think_1k_samples](https://huggingface.co/datasets/dnotitia/Magpie_Claude35_no_think_1k_samples) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Quick start
@@ -29,7 +27,7 @@ print(output["generated_text"])
 ## Training procedure
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/jeehwon/huggingface/runs/7aj0t15a)
 This model was trained with SFT.

 ---
 base_model: Qwen/Qwen3-0.6B
 library_name: transformers
 model_name: Qwen3-0.6B-Distill
 tags:
 - generated_from_trainer
 - trl
 - sft
 licence: license
 # Model Card for Qwen3-0.6B-Distill
+This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Quick start
 ## Training procedure
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/jeehwon/huggingface/runs/wmyciqe4)
 This model was trained with SFT.

all_results.json CHANGED Viewed

@@ -1,8 +1,8 @@
 {
     "total_flos": 241025679360.0,
-    "train_loss": 1.798710823059082,
-    "train_runtime": 28.1411,
     "train_samples": 1000,
-    "train_samples_per_second": 35.535,
-    "train_steps_per_second": 0.284
 }

 {
     "total_flos": 241025679360.0,
+    "train_loss": 0.0,
+    "train_runtime": 1.8142,
     "train_samples": 1000,
+    "train_samples_per_second": 551.204,
+    "train_steps_per_second": 4.41
 }

config.json CHANGED Viewed

@@ -24,7 +24,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.52.0.dev0",
-  "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 151936
 }

   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.52.0.dev0",
+  "use_cache": false,
   "use_sliding_window": false,
   "vocab_size": 151936
 }

train_results.json CHANGED Viewed

@@ -1,8 +1,8 @@
 {
     "total_flos": 241025679360.0,
-    "train_loss": 1.798710823059082,
-    "train_runtime": 28.1411,
     "train_samples": 1000,
-    "train_samples_per_second": 35.535,
-    "train_steps_per_second": 0.284
 }

 {
     "total_flos": 241025679360.0,
+    "train_loss": 0.0,
+    "train_runtime": 1.8142,
     "train_samples": 1000,
+    "train_samples_per_second": 551.204,
+    "train_steps_per_second": 4.41
 }

trainer_state.json CHANGED Viewed

@@ -19,13 +19,12 @@
     },
     {
       "epoch": 1.0,
-      "num_tokens": 341202.0,
       "step": 8,
       "total_flos": 241025679360.0,
-      "train_loss": 1.798710823059082,
-      "train_runtime": 28.1411,
-      "train_samples_per_second": 35.535,
-      "train_steps_per_second": 0.284
     }
   ],
   "logging_steps": 5,

     },
     {
       "epoch": 1.0,
       "step": 8,
       "total_flos": 241025679360.0,
+      "train_loss": 0.0,
+      "train_runtime": 1.8142,
+      "train_samples_per_second": 551.204,
+      "train_steps_per_second": 4.41
     }
   ],
   "logging_steps": 5,

training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ef742d59d27360db3fe16667eb3c58136e19765240fb3cc3b01780091d55532
 size 7288

 version https://git-lfs.github.com/spec/v1
+oid sha256:bcc81f5e1ddde42d1d153d91507165b3de763cc592147a0d5a963c5e5a807b85
 size 7288