Instructions to use finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-Coder-30B-A3B-Instruct")
model = PeftModel.from_pretrained(base_model, "finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct")

Transformers

How to use finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct

SGLang

How to use finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct 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 "finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct" \
    --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": "finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct",
		"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 "finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct" \
        --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": "finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct with Docker Model Runner:
```
docker model run hf.co/finalform/decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct
```

decompose_subtasks_Qwen3-Coder-30B-A3B-Instruct

File size: 1,830 Bytes

34152df

{
  "best_global_step": 10,
  "best_metric": 0.011018680408596992,
  "best_model_checkpoint": "decompose_subtasks_results/checkpoint-10",
  "epoch": 2.0,
  "eval_steps": 10,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4,
      "grad_norm": 0.6372967790496565,
      "learning_rate": 0.0004,
      "loss": 0.1392,
      "step": 2
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.13711821605679056,
      "learning_rate": 0.00036923076923076927,
      "loss": 0.0764,
      "step": 4
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.059600875108830706,
      "learning_rate": 0.0003076923076923077,
      "loss": 0.0168,
      "step": 6
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.08242293936785632,
      "learning_rate": 0.0002461538461538462,
      "loss": 0.0125,
      "step": 8
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.07358354465430926,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.0151,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9957547539335891,
      "eval_loss": 0.011018680408596992,
      "eval_runtime": 68.1767,
      "eval_samples_per_second": 0.484,
      "eval_steps_per_second": 0.073,
      "step": 10
    }
  ],
  "logging_steps": 2,
  "max_steps": 15,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1696630308864.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}