Instructions to use Ba2han/math-test-maxx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Ba2han/math-test-maxx with Transformers:

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

pipe = pipeline("text-generation", model="Ba2han/math-test-maxx")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Ba2han/math-test-maxx")
model = AutoModelForCausalLM.from_pretrained("Ba2han/math-test-maxx", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Ba2han/math-test-maxx with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Ba2han/math-test-maxx"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ba2han/math-test-maxx",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Ba2han/math-test-maxx

SGLang

How to use Ba2han/math-test-maxx 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 "Ba2han/math-test-maxx" \
    --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": "Ba2han/math-test-maxx",
		"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 "Ba2han/math-test-maxx" \
        --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": "Ba2han/math-test-maxx",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Ba2han/math-test-maxx with Docker Model Runner:
```
docker model run hf.co/Ba2han/math-test-maxx
```

math-test-maxx

Merged 16-bit weights: unsloth/Qwen2.5-3B-Instruct + ReasonMaxxer offline-search LoRA (v2 recipe).

Source: ansz42/ReasonMaxxer (offline_search/, pack configs/test_pack_qwen25_3b.yaml).

Same-protocol greedy 0-shot

vLLM, boxed chat prompt, MathVerifier. Not the official 8-shot Qwen numbers.

Model	GSM8K (n=1319)	MATH-500 (n=500)
Base `unsloth/Qwen2.5-3B-Instruct`	84.6%	61.6%
This merge	85.0%	62.8%
vs base	+0.4 pp	+1.2 pp

An earlier train recipe (lr 2e-4, batch 1×4, clip 1.0) regressed to 82.9% / 53.8%. This upload is the milder retry.

Train recipe (preferred)

Knob	Value
LoRA	r=16, α=32, QKVO
learning_rate	`2e-5`
batch × grad accum	2 × 4 (effective 8)
max_grad_norm	0.1
data	300 MATH-500 items (seed 42), 12 offline search samples, entropy-weighted signed loss
steps	774 micro-steps / 194 Adam updates (one pass over informative rows)

In-loop 300-item eval (temp 0.6, n=4): pass@1 61.3%, pass@4 74.0%.

Downloads last month: -

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for Ba2han/math-test-maxx

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-3B-Instruct

Finetuned

unsloth/Qwen2.5-3B-Instruct

Adapter

(477)

this model