Instructions to use LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025 with Transformers:

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

pipe = pipeline("text-generation", model="LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025")
model = AutoModelForCausalLM.from_pretrained("LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025")
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

vLLM

How to use LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025

SGLang

How to use LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025 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 "LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025" \
    --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": "LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025",
		"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 "LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025" \
        --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": "LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025 with Docker Model Runner:
```
docker model run hf.co/LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025
```

QThink-Qwen3-1.7B-AIME2025

QThink: Parallel Latent Reasoning via Per-Step Distillation of Multiple Rollouts

Trained on DAPO-Math-17K (14.1K problems) for competition-level math (AIME 2025).

Caveat: This is a preliminary model trained with short-context data (768 token budget). AIME problems require long reasoning chains (16K+ tokens). A full-length version (R18) is in progress. Results below reflect the short-context constraint, not the method's ceiling.

Results (AIME 2025, 30 problems, 10 runs, temperature=0.6)

Model	Mean ± SE
CODI (paper method)	8.7% ± 1.3%
QThink (ours)	8.3% ± 1.1%
SFT	7.3% ± 1.0%
Base Qwen3-1.7B	6.3% ± 1.3%

All methods trained with matched 768-token budget for fair comparison. Differences are within standard error on this small test set.

For reference, the official Qwen3-1.7B in thinking mode achieves 65.6% on AIME 2025 using 32K+ generation tokens (Qwen3 Technical Report, arXiv 2505.09388).

Cross-Benchmark Results (QThink best)

Benchmark	QThink	SFT	Base	CODI
GSM8k	83.2%	80.7%	77.3%	78.2%
MATH-500	43.6%	38.2%	33.6%	31.2%
Tooluse	48.5%	45.6%	47.1%	42.6%
AIME 2025	8.3%	7.3%	6.3%	8.7%

Training Config

Base model: Qwen/Qwen3-1.7B with LoRA (rank=32, alpha=16)
Mode: uniform multi-rollout per-step distillation (gamma=2.0, K=6 latent steps)
Training data: DAPO-Math-17K English (14.1K problems, 16 rollouts each)
Student budget: max_prompt=512, max_answer=256 (768 total)
Teacher states: precomputed from full rollouts (up to 4096 tokens)

Downloads last month: 1

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for LakshyAAAgrawal/QThink-Qwen3-1.7B-AIME2025

Base model

Qwen/Qwen3-1.7B-Base

Finetuned

Qwen/Qwen3-1.7B

Finetuned

(788)

this model