Instructions to use VeigaPunk/qanta-tossup-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use VeigaPunk/qanta-tossup-v1 with Transformers:

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

pipe = pipeline("text-generation", model="VeigaPunk/qanta-tossup-v1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("VeigaPunk/qanta-tossup-v1")
model = AutoModelForCausalLM.from_pretrained("VeigaPunk/qanta-tossup-v1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use VeigaPunk/qanta-tossup-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "VeigaPunk/qanta-tossup-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VeigaPunk/qanta-tossup-v1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/VeigaPunk/qanta-tossup-v1

SGLang

How to use VeigaPunk/qanta-tossup-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 "VeigaPunk/qanta-tossup-v1" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VeigaPunk/qanta-tossup-v1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "VeigaPunk/qanta-tossup-v1" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VeigaPunk/qanta-tossup-v1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use VeigaPunk/qanta-tossup-v1 with Docker Model Runner:
```
docker model run hf.co/VeigaPunk/qanta-tossup-v1
```

VeigaPunk QANTA 2026 Agent (`qanta-tossup-v1`)

Public submission artifact for QANTA 2026 when the official submission Space is down (RUNTIME_ERROR).

Load

from transformers import pipeline
tossup = pipeline("quizbowl-tossup", model="VeigaPunk/qanta-tossup-v1", trust_remote_code=True)
print(tossup({"question_text": "This French capital is known as the City of Light."}))
bonus = pipeline("quizbowl-bonus", model="VeigaPunk/qanta-tossup-v1", trust_remote_code=True)
print(bonus({"leadin": "French geography.", "part": "Name the capital city."}))

Organizer eval request

HF discussions: spaces/qanta-challenge/quizbowl-submission#1
Queue PRs: datasets/qanta-challenge/advcal-jobs discussions #3-#6
Contact: qanta@googlegroups.com

Base weights resolved from Qwen/Qwen2.5-1.5B-Instruct via config _name_or_path.

Downloads last month: 41

Model tree for VeigaPunk/qanta-tossup-v1

Base model

Qwen/Qwen2.5-1.5B

Finetuned

Qwen/Qwen2.5-1.5B-Instruct

Finetuned

(1704)

this model

VeigaPunk QANTA 2026 Agent (qanta-tossup-v1)

Load

Organizer eval request

Model tree for VeigaPunk/qanta-tossup-v1

VeigaPunk QANTA 2026 Agent (`qanta-tossup-v1`)