Achieving Superior Performance over Qwen3-32B and QwQ-32B Using Only 800 Strategically Curated Samples
Model description
NTele-R1-32B-V1 is the continuation of NTele-R1-32B-Preview, you can visit for more information. We have made great improvements on the base by using less corpus in mathematics and code (only 800 items, including 400 mathematics and 400 codes), and surpassed the industry's advanced models Qwen3-32B and QwQ-32B.
| Model | Release Date | AIME2024 | AIME2025 | MATH500 | GPQA-Diamond | LCB（24.08-25.02） |
|---|---|---|---|---|---|---|
| DeepSeek-R1-Distill-Qwen-32B | 25.1.20 | 64.17 | 55.21 | 89.8 | 62.1 | 50.26 |
| QwQ-32B | 25.3.6 | 76.25 | 67.30 | 94.6 | 63.6 | 60.94 |
| Qwen3-32B(think) | 25.4.29 | 78.75 | 73.33 | 95 | 69.7 | 53.24 |
| NTele-R1-32B-V1(ours) | 25.5.10 | 82.5 | 74.49 | 95.2 | 67.17 | 63.69 |
Data
You can access our dataset to get 800 training data and visit the NTele-R1-32B-Preview to learn about the data synthesis and screening process.
Evaluation
We evaluate models with SkyThought.
Training Details
NTele-R1-32B-V1 was trained from DeepSeek-32B-Distill on 8xH800.
Training hyperparameter
- learning_rate: 1e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 6
- total_train_batch_size: 48
- total_eval_batch_size: 48
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10.0