Instructions to use zhongweixie/claw-rl-v13-gdpo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zhongweixie/claw-rl-v13-gdpo with Transformers:

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

pipe = pipeline("image-text-to-text", model="zhongweixie/claw-rl-v13-gdpo")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("zhongweixie/claw-rl-v13-gdpo")
model = AutoModelForMultimodalLM.from_pretrained("zhongweixie/claw-rl-v13-gdpo", device_map="auto")
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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zhongweixie/claw-rl-v13-gdpo with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zhongweixie/claw-rl-v13-gdpo"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zhongweixie/claw-rl-v13-gdpo",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/zhongweixie/claw-rl-v13-gdpo

SGLang

How to use zhongweixie/claw-rl-v13-gdpo 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 "zhongweixie/claw-rl-v13-gdpo" \
    --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": "zhongweixie/claw-rl-v13-gdpo",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "zhongweixie/claw-rl-v13-gdpo" \
        --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": "zhongweixie/claw-rl-v13-gdpo",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use zhongweixie/claw-rl-v13-gdpo with Docker Model Runner:
```
docker model run hf.co/zhongweixie/claw-rl-v13-gdpo
```

qwen3vl-8b-claw-rl-gdpo-v13-iter560

在内部 agent 任务集上做 RL 微调得到的 Qwen3-VL-8B 模型。

这不是 Qwen3-VL-8B-Instruct 本身，上游那份模型卡里的任何跑分都不适用于本模型。本仓此前误挂了上游 Qwen 的模型卡，现已更正。

训练配置

项	值
实验名	`exp_rl_v5_v13_gdpo`
算法	GDPO
变体	冷启动基线。无 rubric overlay，无 dim scoring。
起点	原始 Qwen3-VL-8B-Instruct(冷启动，非 SFT checkpoint)
最终 checkpoint	`iter_0000560`
训练量	561 个 rollout step，约 3 个 epoch，187 条 prompt
samples / prompt	8
rollout batch	1
global batch	8
硬件	8xH800，TP=8，colocate
奖励 judge	`qwen/qwen-2.5-72b-instruct` (OpenRouter)

学习率、KL 系数等由 slime 的 formal profile 提供，未在启动脚本中显式设置，故此处不列出，以免给出未经核实的数值。

与同期 5 个 *_stage3 模型的区别：那些以 stage3 SFT checkpoint (exp_stage3_v2_merged) 为起点，本模型是从原始 instruct 冷启动的。

评测状态

目前没有可信的 held-out 评测数字。 不要把本模型当作已验证优于其起点。

原因：

cold-start 基线作业(373736)因 sglang JIT 缺 ninja 未能启动，缺少可比基线。
held-out 驱动脚本会主动拒绝一部分任务(沙箱快照类、多模态输入类)，覆盖率不足 30 题全集，其总分不能当作全集分数引用。

用法

from transformers import Qwen3VLForConditionalGeneration, AutoProcessor

model = Qwen3VLForConditionalGeneration.from_pretrained(
    "zhongweixie/qwen3vl-8b-claw-rl-gdpo-v13-iter560", dtype="auto", device_map="auto"
)
processor = AutoProcessor.from_pretrained("zhongweixie/qwen3vl-8b-claw-rl-gdpo-v13-iter560")

来源

由 slime/tools/convert_torch_dist_to_hf.py 从 Megatron 分布式 checkpoint /aifs4su/hansirui_3rd/rl-checkpoints/exp_rl_v5_v13_gdpo/iter_0000560 转换而来，缺失权重与 tokenizer/config 取自原始 Qwen3-VL-8B-Instruct。

Downloads last month: 7

Safetensors

Model size

9B params

Tensor type

BF16

Model tree for zhongweixie/claw-rl-v13-gdpo

Base model

Qwen/Qwen3-VL-8B-Instruct

Finetuned

(688)

this model