LiangJiang commited on
Commit
1a27ba5
·
verified ·
1 Parent(s): c5c9fea

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -52,7 +52,7 @@ To compare the performance of Ring-lite-2507 and Ring-lite, we evaluate the two
52
 
53
  ### Math
54
 
55
- | **Benchmark** | **Ring-lite-2507** | **Ring-lite-2506** | **Qwen3-8B-Thinking**
56
  | :-------------: | :---------------: | :-----------: | :-------------------: |
57
  | MATH-500 (Pass@1) | 97.60 | 97.95 | 97.30 |
58
  | CNMO 2024 (Pass@1) | 76.91 | 77.78 | 75.09 |
@@ -64,14 +64,14 @@ To compare the performance of Ring-lite-2507 and Ring-lite, we evaluate the two
64
 
65
  ### Coding
66
 
67
- | **Benchmark** | **Ring-lite-2507** | **Ring-lite-2506** | **Qwen3-8B-Thinking**
68
  | :-------------: | :---------------: | :-----------: | :-------------------: |
69
  | LiveCodeBench(2408-2505) (Pass@1) |62.56 | 63.27 | 56.94 |
70
  | Codeforces | 84.80 | 89.09 | 73.31 |
71
 
72
  ### Reasoning \& Agentic
73
 
74
- | **Benchmark** | **Ring-lite-2507** | **Ring-lite-2506** | **Qwen3-8B-Thinking**
75
  | :-------------: | :---------------: | :-----------: | :-------------------: |
76
  | DROP (zero-shot F1) | 88.55 | 89.27 | 87.13 |
77
  | BBH (EM) | 87.59 | 88.65 | 87.30 |
@@ -81,7 +81,7 @@ To compare the performance of Ring-lite-2507 and Ring-lite, we evaluate the two
81
 
82
  ### Alignment
83
 
84
- | **Benchmark** | **Ring-lite-2507** | **Ring-lite-2506** | **Qwen3-8B-Thinking**
85
  | :-------------: | :---------------: | :-----------: | :-------------------: |
86
  | IFEval (Prompt Strict) | 78.93 | 82.99 | 85.0 |
87
  | AlignBench v1.1(gpt-4.1) | 80.69 | 80.90 | 74.70 |
 
52
 
53
  ### Math
54
 
55
+ | **Benchmark** | **Ring-mini-2.0** | **Ring-lite-2507** | **Qwen3-8B-Thinking**
56
  | :-------------: | :---------------: | :-----------: | :-------------------: |
57
  | MATH-500 (Pass@1) | 97.60 | 97.95 | 97.30 |
58
  | CNMO 2024 (Pass@1) | 76.91 | 77.78 | 75.09 |
 
64
 
65
  ### Coding
66
 
67
+ | **Benchmark** | **Ring-mini-2.0** | **Ring-lite-2507** | **Qwen3-8B-Thinking**
68
  | :-------------: | :---------------: | :-----------: | :-------------------: |
69
  | LiveCodeBench(2408-2505) (Pass@1) |62.56 | 63.27 | 56.94 |
70
  | Codeforces | 84.80 | 89.09 | 73.31 |
71
 
72
  ### Reasoning \& Agentic
73
 
74
+ | **Benchmark** | **Ring-mini-2.0** | **Ring-lite-2507** | **Qwen3-8B-Thinking**
75
  | :-------------: | :---------------: | :-----------: | :-------------------: |
76
  | DROP (zero-shot F1) | 88.55 | 89.27 | 87.13 |
77
  | BBH (EM) | 87.59 | 88.65 | 87.30 |
 
81
 
82
  ### Alignment
83
 
84
+ | **Benchmark** | **Ring-mini-2.0** | **Ring-lite-2507** | **Qwen3-8B-Thinking**
85
  | :-------------: | :---------------: | :-----------: | :-------------------: |
86
  | IFEval (Prompt Strict) | 78.93 | 82.99 | 85.0 |
87
  | AlignBench v1.1(gpt-4.1) | 80.69 | 80.90 | 74.70 |