| 注册会员 | 1331 |
| 主题 | 846 |
| 模型 | 3026 |
| 技能包 | 13874 |
| 数据集 | 1047 |
| 论文 | 545 |
| 开源项目 | 649 |
| 旗舰模型 | 厂商 | 智能评分 |
|---|---|---|
| Claude Opus 5 (Adaptive Reasoning, Max Effort) |
63.05
|
|
| GPT-5.6 Sol (max) |
60.93
|
|
| Grok 4.6 (high) |
60.92
|
|
| Kimi K3 (max) |
59.70
|
|
| GLM-5.3 (max) |
59.51
|
|
| Qwen3.8 Max |
58.08
|
|
| Muse Spark 1.2 (xhigh) |
56.76
|
|
| Gemini 3.7 Flash (high) |
56.03
|
|
| DeepSeek V4 Pro 0813 (Reasoning, Max Effort) |
53.20
|
|
| Motif 3 |
47.36
|
|
| MiniMax-M3 |
45.40
|
|
| MiMo-V2.5-Pro |
42.88
|
|
| Inkling (xhigh) |
42.29
|
|
| Hy3 |
42.21
|
|
| Nex-N2-Pro |
41.74
|
|
| Solar Pro 4 |
41.64
|
|
| JT-4.1 Flash 236B A21B |
39.85
|
|
| Agnes 2.5 Pro Alpha |
39.71
|
|
| Nemotron 3 Ultra 550B A55B (Reasoning) |
38.32
|
|
| Ling 3.0 Flash |
37.82
|
|
| A.X-K2 |
35.01
|
|
| G9v3-39A5B |
34.02
|
|
| LongCat 2.0 |
33.97
|
|
| KAT Coder Pro V2 |
33.73
|
|
| K-EXAONE 2.0 0803 |
30.98
|
|
| Step 3.7 Flash |
30.90
|
|
| Mistral Medium 3.5 |
30.39
|
|
| Doubao Seed Code |
26.49
|
|
| Command A+ |
22.77
|
|
| ERNIE 5.0 Thinking Preview |
22.28
|
|
| Nova 2.0 Pro Preview (medium) |
22.12
|
|
| Mercury 2 |
21.90
|
|
| Apriel-v1.6-15B-Thinker |
20.85
|
|
| Trinity Large Thinking |
18.66
|
|
| HyperNova 60B 2605 |
18.32
|
|
| K2 Think V2 |
17.43
|
|
| HyperCLOVA X SEED Think (32B) | Naver |
17.19
|
| Mi:dm K 2.5 Pro |
16.56
|
|
| INTELLECT-3 |
15.72
|
|
| Celeris-1 |
12.35
|
|
| Tri-21B-Think |
12.34
|
|
| MiniCPM5-1B (Reasoning) |
11.92
|
|
| Sarvam 105B (high) |
11.90
|
|
| Nanbeige4.1-3B |
11.02
|
|
| LFM2.5-2.6B |
10.99
|
|
| Hermes 4 - Llama-3.1 70B (Reasoning) |
9.85
|
|
| Falcon-H1R-7B |
9.66
|
|
| Granite 4.1 30B |
8.70
|
|
| Olmo 3.1 32B Think |
7.90
|
|
| R1 1776 |
6.05
|
|
| Phi-4 Mini Instruct |
5.74
|
|
| Jamba 1.7 Large |
4.99
|
|
| Reka Flash 3 |
3.71
|
|
| Apertus 70B Instruct |
1.98
|