Instructions to use end000/Qwen3.5-0.8B-WSPForge-Math-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use end000/Qwen3.5-0.8B-WSPForge-Math-v1 with Transformers:

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

pipe = pipeline("text-generation", model="end000/Qwen3.5-0.8B-WSPForge-Math-v1")
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("end000/Qwen3.5-0.8B-WSPForge-Math-v1")
model = AutoModelForMultimodalLM.from_pretrained("end000/Qwen3.5-0.8B-WSPForge-Math-v1", 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 end000/Qwen3.5-0.8B-WSPForge-Math-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "end000/Qwen3.5-0.8B-WSPForge-Math-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "end000/Qwen3.5-0.8B-WSPForge-Math-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/end000/Qwen3.5-0.8B-WSPForge-Math-v1

SGLang

How to use end000/Qwen3.5-0.8B-WSPForge-Math-v1 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 "end000/Qwen3.5-0.8B-WSPForge-Math-v1" \
    --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": "end000/Qwen3.5-0.8B-WSPForge-Math-v1",
		"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 "end000/Qwen3.5-0.8B-WSPForge-Math-v1" \
        --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": "end000/Qwen3.5-0.8B-WSPForge-Math-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use end000/Qwen3.5-0.8B-WSPForge-Math-v1 with Docker Model Runner:
```
docker model run hf.co/end000/Qwen3.5-0.8B-WSPForge-Math-v1
```

Qwen3.5-0.8B-WSPForge-Math-v1 / wspforge /research_code /scripts /validate_coding_tasks.py

end000

Release Sol014 mathematical specialization with paired evaluations, regressions and exact replay artifacts

fc32096 verified about 22 hours ago

Raw History Blame Contribute Delete

1.91 kB

import hashlib
import json
from common import ROOT, rows, save_json
from code_sandbox import check_code

directory = ROOT/'data/coding'
records = rows(directory/'all_tasks.jsonl')
accepted, outcomes = [], []
for i,row in enumerate(records):
    result = check_code(row['reference_code'],row['entry_point'],row['test_code'])
    outcomes.append({'id':row['id'],**result})
    if result['passed']:
        accepted.append(row)
    if i%50==0:
        print('Canonical task checks',i+1,'accepted',len(accepted),flush=True)
save_json(directory/'canonical_validation.json',outcomes)
groups = {'discovery':[],'validation':[],'confirmation':[]}
for source in ['humaneval','mbpp']:
    pool = sorted([r for r in accepted if r['source']==source],key=lambda r:hashlib.sha256(('wspforge-code-v1:'+r['id']).encode()).hexdigest())
    for split,start,count in [('discovery',0,12),('validation',12,12),('confirmation',24,24)]:
        if len(pool)<start+count:
            raise ValueError('Not enough eligible canonical tasks')
        groups[split]+=pool[start:start+count]
for name,items in groups.items():
    (directory/f'{name}.jsonl').write_text(''.join(json.dumps(r,ensure_ascii=False)+'\n' for r in items))
manifest = {name:{'count':len(items),'ids':[r['id'] for r in items],
                  'sha256':hashlib.sha256((directory/f'{name}.jsonl').read_bytes()).hexdigest()} for name,items in groups.items()}
manifest['_eligibility'] = {'original_tasks':len(records),'canonical_passed_sandbox':len(accepted),
                            'criterion':'canonical code passes fixed sandbox and standard-library guard before any model answer is measured',
                            'note':'Compact custom subsets, not official full-benchmark leaderboard scores; pretraining exposure is unknown.'}
save_json(directory/'splits_manifest.json',manifest)
print(json.dumps({k:len(v) for k,v in groups.items()}),flush=True)

	import hashlib
	import json
	from common import ROOT, rows, save_json
	from code_sandbox import check_code

	directory = ROOT/'data/coding'
	records = rows(directory/'all_tasks.jsonl')
	accepted, outcomes = [], []
	for i,row in enumerate(records):
	result = check_code(row['reference_code'],row['entry_point'],row['test_code'])
	outcomes.append({'id':row['id'],**result})
	if result['passed']:
	accepted.append(row)
	if i%50==0:
	print('Canonical task checks',i+1,'accepted',len(accepted),flush=True)
	save_json(directory/'canonical_validation.json',outcomes)
	groups = {'discovery':[],'validation':[],'confirmation':[]}
	for source in ['humaneval','mbpp']:
	pool = sorted([r for r in accepted if r['source']==source],key=lambda r:hashlib.sha256(('wspforge-code-v1:'+r['id']).encode()).hexdigest())
	for split,start,count in [('discovery',0,12),('validation',12,12),('confirmation',24,24)]:
	if len(pool)<start+count:
	raise ValueError('Not enough eligible canonical tasks')
	groups[split]+=pool[start:start+count]
	for name,items in groups.items():
	(directory/f'{name}.jsonl').write_text(''.join(json.dumps(r,ensure_ascii=False)+'\n' for r in items))
	manifest = {name:{'count':len(items),'ids':[r['id'] for r in items],
	'sha256':hashlib.sha256((directory/f'{name}.jsonl').read_bytes()).hexdigest()} for name,items in groups.items()}
	manifest['_eligibility'] = {'original_tasks':len(records),'canonical_passed_sandbox':len(accepted),
	'criterion':'canonical code passes fixed sandbox and standard-library guard before any model answer is measured',
	'note':'Compact custom subsets, not official full-benchmark leaderboard scores; pretraining exposure is unknown.'}
	save_json(directory/'splits_manifest.json',manifest)
	print(json.dumps({k:len(v) for k,v in groups.items()}),flush=True)