Instructions to use BirdToast/qwen3.5-9b-marvin with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries
PEFT
How to use BirdToast/qwen3.5-9b-marvin with PEFT:
```
Task type is invalid.
```

How to use BirdToast/qwen3.5-9b-marvin with Transformers:

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

pipe = pipeline("text-generation", model="BirdToast/qwen3.5-9b-marvin")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("BirdToast/qwen3.5-9b-marvin")
model = AutoModelForImageTextToText.from_pretrained("BirdToast/qwen3.5-9b-marvin")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use BirdToast/qwen3.5-9b-marvin with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BirdToast/qwen3.5-9b-marvin"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BirdToast/qwen3.5-9b-marvin",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/BirdToast/qwen3.5-9b-marvin

SGLang

How to use BirdToast/qwen3.5-9b-marvin 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 "BirdToast/qwen3.5-9b-marvin" \
    --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": "BirdToast/qwen3.5-9b-marvin",
		"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 "BirdToast/qwen3.5-9b-marvin" \
        --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": "BirdToast/qwen3.5-9b-marvin",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use BirdToast/qwen3.5-9b-marvin with Docker Model Runner:
```
docker model run hf.co/BirdToast/qwen3.5-9b-marvin
```

output-marvin

This model is a fine-tuned version of Qwen/Qwen3.5-9B.

W&B run: https://wandb.ai/cooawoo-personal/Qwen9B/runs/agujtrjb

Training procedure

Hyperparameters

Parameter	Value
Learning rate	`0.0001`
LR scheduler	SchedulerType.COSINE
Per-device batch size	1
Gradient accumulation	4
Effective batch size	4
Epochs	1
Max sequence length	16384
Optimizer	OptimizerNames.PAGED_ADAMW_8BIT
Weight decay	0.01
Warmup ratio	0.05
Max gradient norm	1.0
Precision	bf16
Loss type	nll
Chunked cross-entropy	yes

LoRA configuration

Parameter	Value
Rank (r)	32
Alpha	64
Target modules	attn.proj, down_proj, gate_proj, in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, linear_fc1, linear_fc2, o_proj, out_proj, q_proj, qkv, up_proj, v_proj
Quantization	4-bit (nf4)

Dataset statistics

Dataset	Samples	Total tokens	Trainable tokens
rpDungeon/marvin/marvin_no_anthologies.json	1,480	22,939,354	22,939,354

Training config

model_name_or_path: Qwen/Qwen3.5-9B
bf16: true
gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
use_liger: true
use_cce: true
neftune_noise_alpha: 5
dataloader_num_workers: 4
dataloader_pin_memory: true
max_length: 16384
learning_rate: 0.0001
warmup_ratio: 0.05
weight_decay: 0.01
lr_scheduler_type: cosine
per_device_train_batch_size: 1
gradient_accumulation_steps: 4
optim: paged_adamw_8bit
max_grad_norm: 1.0
use_peft: true
load_in_4bit: true
bnb_4bit_quant_type: nf4
lora_r: 32
lora_alpha: 64
lora_dropout: 0
logging_steps: 1
disable_tqdm: false
save_strategy: steps
save_steps: 500
save_total_limit: 3
report_to: wandb
output_dir: output-marvin
data_config: data.yaml
prepared_dataset: prepared
attn_implementation: flash_attention_2
num_train_epochs: 1
saves_per_epoch: 3
run_name: qwen35-9b-qlora-marvin

Data config

datasets:
- path: rpDungeon/marvin
  data_files: marvin_no_anthologies.json
  type: text
  truncation_strategy: split
shuffle_datasets: true
shuffle_combined: true
shuffle_seed: 42
eval_split: 0.0
split_seed: 42
assistant_only_loss: false

Framework versions

PEFT 0.18.1
Loft: 0.1.0
Transformers: 5.2.0
Pytorch: 2.6.0+cu124
Datasets: 4.6.1
Tokenizers: 0.22.2

Downloads last month: -

Safetensors

Model size

10B params

Tensor type

BF16

F32

Model tree for BirdToast/qwen3.5-9b-marvin

Base model

Qwen/Qwen3.5-9B-Base

Finetuned

Qwen/Qwen3.5-9B

Adapter

(211)

this model