wandermay commited on
Commit
63450fa
·
verified ·
1 Parent(s): 86e0dea

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -3,6 +3,10 @@ license: mit
3
  ---
4
  ## Achieving Superior Performance over QwQ-32B Using Only 965 Strategically Curated Samples
5
 
 
 
 
 
6
  ### Model description
7
  Most existing mthods focused on distilling DeepSeek-R1 to improve reasoning ability. However, as far as we know, there is no distilled model could surpass DeepSeek-R1 or QwQ-32B. We introduce NTele-R1-32B-DS , a state-of-the-art mathematical reasoning model that outperforms QwQ-32B across common reasoning benchmarks, including AIME2024/2025, MATH500 and GPQA-Diamond.
8
  Notely, NTele-R1-32B-DS is the first that achieves **more than 80/70 in challenging AIME2024/2025**.
 
3
  ---
4
  ## Achieving Superior Performance over QwQ-32B Using Only 965 Strategically Curated Samples
5
 
6
+ ### NTele-R1-32B-V1
7
+ [NTele-R1-32B-V1](https://huggingface.co/ZTE-AIM/NTele-R1-32B-V1) is the continuation of NTele-R1-32B-Preview, and its capabilities can be accessed [here](https://huggingface.co/ZTE-AIM/NTele-R1-32B-V1).
8
+
9
+
10
  ### Model description
11
  Most existing mthods focused on distilling DeepSeek-R1 to improve reasoning ability. However, as far as we know, there is no distilled model could surpass DeepSeek-R1 or QwQ-32B. We introduce NTele-R1-32B-DS , a state-of-the-art mathematical reasoning model that outperforms QwQ-32B across common reasoning benchmarks, including AIME2024/2025, MATH500 and GPQA-Diamond.
12
  Notely, NTele-R1-32B-DS is the first that achieves **more than 80/70 in challenging AIME2024/2025**.