Instructions to use vsan/tiny-pickle-v3-coder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use vsan/tiny-pickle-v3-coder with Transformers:

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

pipe = pipeline("text-generation", model="vsan/tiny-pickle-v3-coder")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("vsan/tiny-pickle-v3-coder")
model = AutoModelForCausalLM.from_pretrained("vsan/tiny-pickle-v3-coder", device_map="auto")
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 vsan/tiny-pickle-v3-coder with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "vsan/tiny-pickle-v3-coder"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "vsan/tiny-pickle-v3-coder",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/vsan/tiny-pickle-v3-coder

SGLang

How to use vsan/tiny-pickle-v3-coder 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 "vsan/tiny-pickle-v3-coder" \
    --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": "vsan/tiny-pickle-v3-coder",
		"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 "vsan/tiny-pickle-v3-coder" \
        --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": "vsan/tiny-pickle-v3-coder",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use vsan/tiny-pickle-v3-coder with Docker Model Runner:
```
docker model run hf.co/vsan/tiny-pickle-v3-coder
```

tiny-pickle-v3-coder / training_metrics.json

vsan

Publish merged Tiny Pickle v3 Coder Safetensors

5db6877 verified about 22 hours ago

Raw

History Blame Contribute Delete

751 Bytes

	{
	"model_name": "Tiny Pickle v3 Coder",
	"base_model": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
	"dataset": "nvidia/OpenCodeInstruct",
	"filter": "average_test_score == 1.0 and every recorded test passed",
	"optimizer_steps": 9920,
	"approximate_packed_sequences_processed": 39680,
	"maximum_sequence_length": 4096,
	"lora_rank": 64,
	"lora_alpha": 128,
	"learning_rate": 2e-05,
	"training_started_utc": "2026-08-05T14:21:34.316358+00:00",
	"training_finished_utc": "2026-08-06T00:21:43.664269+00:00",
	"train_metrics": {
	"train_runtime": 36009.1351,
	"train_samples_per_second": 111.083,
	"train_steps_per_second": 27.771,
	"total_flos": 2.9005106965157708e+19,
	"train_loss": 0.21270046533956644,
	"epoch": 0.00992
	}
	}