社区运行状况
| 注册会员 | 1100 |
| 主题 | 846 |
| 模型 | 3026 |
| 技能包 | 13874 |
| 数据集 | 1047 |
| 论文 | 331 |
| 开源项目 | 532 |
| 模型名称 | 厂商 | 智能评分 | 编程能力 | 智能体能力 | 速度评分 |
|---|---|---|---|---|---|
| GPT-5.5 (xhigh) |  OpenAI |
60.24
|
59.12
|
74.12
|
63.29
|
| GPT-5.5 (high) | OpenAI |
58.87
|
58.53
|
71.95
|
62.09
|
| Claude Opus 4.7 (Adaptive Reasoning, Max Effort) |  Anthropic |
57.28
|
52.51
|
71.29
|
43.50
|
| Gemini 3.1 Pro Preview |  Google |
57.18
|
55.50
|
59.09
|
119.49
|
| GPT-5.5 (medium) | OpenAI |
56.71
|
56.21
|
69.39
|
57.08
|
| Kimi K2.6 |  Kimi |
53.90
|
47.12
|
65.97
|
30.06
|
| MiMo-V2.5-Pro |  Xiaomi |
53.83
|
45.53
|
67.44
|
56.19
|
| GPT-5.3 Codex (xhigh) | OpenAI |
53.56
|
53.10
|
60.54
|
72.85
|
| Grok 4.3 |  xAI |
53.20
|
41.03
|
65.89
|
78.35
|
| Muse Spark |  Meta |
52.15
|
47.47
|
61.99
|
—
|
| Claude Opus 4.7 (Non-reasoning, High Effort) | Anthropic |
51.82
|
53.07
|
64.64
|
38.57
|
| Qwen3.6 Max Preview |  Alibaba |
51.81
|
44.92
|
64.83
|
36.37
|
| Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) | Anthropic |
51.72
|
50.94
|
63.00
|
51.38
|
| DeepSeek V4 Pro (Reasoning, Max Effort) |  DeepSeek |
51.51
|
47.47
|
67.19
|
33.86
|
| GLM-5.1 (Reasoning) |  Z AI |
51.41
|
43.37
|
67.05
|
51.94
|
| GPT-5.5 (low) | OpenAI |
50.78
|
52.06
|
59.69
|
56.85
|
| Qwen3.6 Plus | Alibaba |
49.98
|
42.87
|
61.67
|
52.75
|
| DeepSeek V4 Pro (Reasoning, High Effort) | DeepSeek |
49.79
|
43.25
|
66.65
|
33.24
|
| GLM-5 (Reasoning) | Z AI |
49.77
|
44.18
|
63.14
|
63.52
|
| MiniMax-M2.7 |  MiniMax |
49.62
|
41.93
|
61.49
|
47.12
|
| MiMo-V2.5 | Xiaomi |
49.03
|
42.13
|
65.53
|
92.55
|
| GPT-5.4 mini (xhigh) | OpenAI |
48.90
|
51.48
|
58.88
|
155.83
|
| GPT-5.4 (low) | OpenAI |
47.94
|
45.57
|
58.22
|
67.61
|
| GLM-5-Turbo | Z AI |
46.76
|
36.77
|
66.13
|
—
|
| DeepSeek V4 Flash (Reasoning, Max Effort) | DeepSeek |
46.52
|
38.71
|
61.28
|
72.49
|
| Gemini 3 Flash Preview (Reasoning) | Google |
46.43
|
42.62
|
49.66
|
160.81
|
| Qwen3.6 27B (Reasoning) | Alibaba |
45.82
|
36.50
|
62.85
|
61.72
|
| Qwen3.5 397B A17B (Reasoning) | Alibaba |
45.05
|
41.28
|
55.83
|
52.84
|
| MiMo-V2-Omni-0327 | Xiaomi |
44.93
|
36.89
|
58.63
|
116.21
|
| DeepSeek V4 Flash (Reasoning, High Effort) | DeepSeek |
44.87
|
39.76
|
57.82
|
—
|
| Claude Sonnet 4.6 (Non-reasoning, High Effort) | Anthropic |
44.38
|
46.43
|
61.62
|
42.37
|
| GPT-5.4 nano (xhigh) | OpenAI |
43.98
|
43.91
|
47.60
|
144.86
|
| GLM-5.1 (Non-reasoning) | Z AI |
43.82
|
35.77
|
66.04
|
42.37
|
| Qwen3.6 35B A3B (Reasoning) | Alibaba |
43.49
|
35.15
|
58.34
|
193.59
|
| MiMo-V2-Omni | Xiaomi |
43.40
|
35.46
|
58.56
|
118.97
|
| Kimi K2.6 (Non-reasoning) | Kimi |
42.95
|
38.41
|
58.73
|
29.23
|
| GLM 5V Turbo (Reasoning) | Z AI |
42.85
|
36.22
|
61.07
|
—
|
| Claude Sonnet 4.6 (Non-reasoning, Low Effort) | Anthropic |
42.60
|
42.98
|
57.45
|
44.18
|
| Hy3-preview (Reasoning) |  Tencent |
41.85
|
36.46
|
55.67
|
80.66
|
| Qwen3.5 122B A10B (Reasoning) | Alibaba |
41.60
|
34.71
|
53.00
|
145.09
|
| MiMo-V2-Flash (Feb 2026) | Xiaomi |
41.46
|
33.48
|
48.76
|
133.76
|
| Gemini 3 Pro Preview (low) | Google |
41.30
|
39.36
|
45.05
|
—
|
| GPT-5.5 (Non-reasoning) | OpenAI |
40.94
|
48.61
|
50.24
|
58.51
|
| GLM-5 (Non-reasoning) | Z AI |
40.57
|
39.03
|
60.27
|
56.82
|
| Qwen3.5 397B A17B (Non-reasoning) | Alibaba |
40.10
|
37.43
|
53.32
|
53.56
|
| DeepSeek V4 Pro (Non-reasoning) | DeepSeek |
39.27
|
38.36
|
63.27
|
33.09
|
| Mistral Medium 3.5 |  Mistral |
39.23
|
35.42
|
53.16
|
143.22
|
| Gemma 4 31B (Reasoning) | Google |
39.18
|
38.71
|
40.94
|
35.42
|
| Qwen3.5 Omni Plus | Alibaba |
38.63
|
27.64
|
52.83
|
53.33
|
| Grok 4.1 Fast (Reasoning) | xAI |
38.61
|
30.90
|
49.31
|
77.52
|
| Step 3.5 Flash 2603 |  StepFun |
38.47
|
34.56
|
48.23
|
180.77
|
| o3 | OpenAI |
38.37
|
38.40
|
36.09
|
71.84
|
| GPT-5.4 nano (medium) | OpenAI |
38.11
|
35.03
|
41.64
|
143.78
|
| GPT-5.4 mini (medium) | OpenAI |
37.73
|
37.46
|
40.27
|
158.68
|
| Kimi K2.5 (Non-reasoning) | Kimi |
37.27
|
25.82
|
52.84
|
43.92
|
| Qwen3.6 27B (Non-reasoning) | Alibaba |
37.14
|
26.56
|
60.86
|
61.66
|
| Claude 4.5 Haiku (Reasoning) | Anthropic |
37.09
|
32.61
|
40.16
|
91.26
|
| DeepSeek V4 Flash (Non-reasoning) | DeepSeek |
36.46
|
35.15
|
61.33
|
72.61
|
| NVIDIA Nemotron 3 Super 120B A12B (Reasoning) |  NVIDIA |
35.97
|
31.19
|
40.18
|
178.03
|
| Qwen3.5 122B A10B (Non-reasoning) | Alibaba |
35.87
|
31.58
|
49.51
|
142.67
|
| Nova 2.0 Pro Preview (medium) |  Amazon |
35.71
|
30.40
|
46.95
|
143.32
|
| MiMo-V2.5-Pro (Non-reasoning) | Xiaomi |
35.59
|
36.78
|
50.75
|
60.53
|
| GPT-5.4 (Non-reasoning) | OpenAI |
35.39
|
40.95
|
39.14
|
59.29
|
| Gemini 3 Flash Preview (Non-reasoning) | Google |
35.05
|
37.84
|
35.01
|
168.74
|
| Gemini 2.5 Pro | Google |
34.63
|
31.95
|
32.68
|
125.20
|
| Nova 2.0 Lite (high) | Amazon |
34.54
|
23.42
|
37.34
|
153.06
|
| Hy3-preview (Non-reasoning) | Tencent |
33.66
|
34.33
|
46.70
|
81.47
|
| Ling-2.6-1T |  InclusionAI |
33.61
|
33.05
|
48.21
|
—
|
| Doubao Seed Code |  ByteDance Seed |
33.52
|
31.26
|
36.45
|
—
|
| Gemini 3.1 Flash-Lite Preview | Google |
33.52
|
30.13
|
25.67
|
325.76
|
| gpt-oss-120B (high) | OpenAI |
33.27
|
28.62
|
37.87
|
230.89
|
| Mercury 2 |  Inception |
32.82
|
30.56
|
39.74
|
687.39
|
| Qwen3.5 9B (Reasoning) | Alibaba |
32.43
|
25.34
|
37.42
|
50.20
|
| Gemma 4 31B (Non-reasoning) | Google |
32.29
|
33.90
|
39.39
|
—
|
| K-EXAONE (Reasoning) |  LG AI Research |
32.12
|
27.03
|
38.14
|
—
|
| Grok 3 mini Reasoning (high) | xAI |
32.08
|
25.16
|
31.15
|
152.37
|
| Nova 2.0 Pro Preview (low) | Amazon |
31.90
|
24.50
|
37.71
|
139.49
|
| Trinity Large Thinking |  Arcee AI |
31.87
|
27.19
|
42.61
|
117.20
|
| Qwen3.6 35B A3B (Non-reasoning) | Alibaba |
31.53
|
17.60
|
52.53
|
186.88
|
| Gemma 4 26B A4B (Reasoning) | Google |
31.21
|
22.44
|
32.15
|
—
|
| Claude 4.5 Haiku (Non-reasoning) | Anthropic |
31.05
|
29.64
|
32.59
|
86.26
|
| Qwen3.5 35B A3B (Non-reasoning) | Alibaba |
30.69
|
16.83
|
48.04
|
153.02
|
| MiMo-V2-Flash (Non-reasoning) | Xiaomi |
30.35
|
25.81
|
47.34
|
128.40
|
| EXAONE 4.5 33B | LG AI Research |
30.23
|
22.97
|
36.50
|
—
|
| Nova 2.0 Lite (medium) | Amazon |
29.73
|
23.88
|
32.85
|
173.44
|
| ERNIE 5.0 Thinking Preview |  Baidu |
29.09
|
29.17
|
39.74
|
—
|
| Grok 4.20 0309 v2 (Non-reasoning) | xAI |
28.99
|
22.03
|
38.31
|
91.07
|
| Grok Code Fast 1 | xAI |
28.74
|
23.69
|
35.63
|
74.49
|
| Nemotron Cascade 2 30B A3B | NVIDIA |
28.35
|
25.75
|
26.15
|
—
|
| Qwen3 Coder Next | Alibaba |
28.28
|
22.89
|
42.10
|
151.23
|
| Nova 2.0 Omni (medium) | Amazon |
28.02
|
15.11
|
38.20
|
—
|
| Mistral Small 4 (Reasoning) | Mistral |
27.80
|
24.27
|
25.87
|
150.24
|
| Qwen3.5 9B (Non-reasoning) | Alibaba |
27.33
|
21.35
|
41.10
|
—
|
| Magistral Medium 1.2 | Mistral |
27.10
|
21.66
|
24.45
|
40.12
|
| Gemma 4 26B A4B (Non-reasoning) | Google |
27.09
|
29.09
|
28.93
|
—
|
| Qwen3.5 4B (Reasoning) | Alibaba |
27.08
|
17.49
|
32.46
|
205.10
|
| DeepSeek R1 0528 (May '25) | DeepSeek |
27.07
|
24.03
|
20.84
|
—
|
| Qwen3 Next 80B A3B (Reasoning) | Alibaba |
26.72
|
19.49
|
23.56
|
157.55
|
| Ling 2.6 Flash | InclusionAI |
26.16
|
23.17
|
38.06
|
208.75
|
| Qwen3.5 Omni Flash | Alibaba |
25.87
|
14.04
|
41.63
|
240.15
|
| Solar Pro 3 |  Upstage |
25.87
|
13.27
|
34.92
|
—
|
| JT-MINI |  China Mobile |
25.37
|
21.19
|
42.36
|
—
|
| Nova 2.0 Lite (low) | Amazon |
24.59
|
13.64
|
27.77
|
167.01
|
| gpt-oss-120B (low) | OpenAI |
24.47
|
15.53
|
28.04
|
241.61
|
| gpt-oss-20B (high) | OpenAI |
24.47
|
18.53
|
27.60
|
249.55
|
| GPT-5.4 nano (Non-Reasoning) | OpenAI |
24.36
|
27.89
|
25.92
|
146.34
|
| NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) | NVIDIA |
24.27
|
18.97
|
19.14
|
187.69
|
| LongCat Flash Lite |  LongCat |
23.93
|
16.52
|
38.76
|
120.64
|
| Grok 4.1 Fast (Non-reasoning) | xAI |
23.56
|
19.47
|
32.95
|
71.70
|
| K-EXAONE (Non-reasoning) | LG AI Research |
23.41
|
13.53
|
31.22
|
—
|
| GPT-5.4 mini (Non-Reasoning) | OpenAI |
23.28
|
25.32
|
25.01
|
144.30
|
| Nova 2.0 Omni (low) | Amazon |
23.22
|
13.95
|
22.61
|
—
|
| Mi:dm K 2.5 Pro |  Korea Telecom |
23.06
|
12.57
|
36.76
|
—
|
| Nova 2.0 Pro Preview (Non-reasoning) | Amazon |
23.06
|
20.49
|
23.88
|
113.00
|
| Mistral Large 3 | Mistral |
22.80
|
22.68
|
21.70
|
48.24
|
| Ring-1T | InclusionAI |
22.78
|
16.78
|
18.31
|
—
|
| Qwen3.5 4B (Non-reasoning) | Alibaba |
22.60
|
13.67
|
36.26
|
206.42
|
| INTELLECT-3 |  Prime Intellect |
22.17
|
19.10
|
19.81
|
—
|
| Devstral 2 | Mistral |
22.04
|
23.66
|
21.86
|
56.68
|
| Solar Open 100B (Reasoning) | Upstage |
21.67
|
10.47
|
24.29
|
—
|
| Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) | Google |
21.65
|
18.15
|
11.67
|
—
|
| Nemotron 3 Nano Omni 30B A3B Reasoning | NVIDIA |
21.43
|
14.81
|
23.87
|
311.10
|
| gpt-oss-20B (low) | OpenAI |
20.79
|
14.37
|
21.86
|
247.30
|
| Qwen3 Next 80B A3B Instruct | Alibaba |
20.11
|
15.27
|
14.19
|
148.66
|
| Devstral Small 2 | Mistral |
19.47
|
20.72
|
20.79
|
56.74
|
| Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) | Google |
19.42
|
14.54
|
10.14
|
—
|
| Motif-2-12.7B-Reasoning |  Motif Technologies |
19.08
|
11.94
|
19.17
|
—
|
| Ling-1T | InclusionAI |
19.04
|
18.80
|
11.84
|
—
|
| Nova Premier | Amazon |
19.01
|
13.84
|
16.43
|
27.70
|
| Gemma 4 E4B (Reasoning) | Google |
18.76
|
13.70
|
6.92
|
56.22
|
| Llama Nemotron Super 49B v1.5 (Reasoning) | NVIDIA |
18.68
|
15.15
|
9.36
|
47.19
|
| Mistral Small 4 (Non-reasoning) | Mistral |
18.62
|
16.45
|
18.55
|
143.36
|
| Llama 3.3 Nemotron Super 49B v1 (Reasoning) | NVIDIA |
18.49
|
9.41
|
—
|
—
|
| Llama 4 Maverick | Meta |
18.36
|
15.58
|
7.22
|
107.36
|
| Magistral Small 1.2 | Mistral |
18.16
|
14.76
|
17.33
|
106.10
|
| Sarvam 105B (high) |  Sarvam |
18.16
|
9.81
|
24.69
|
100.02
|
| Nova 2.0 Lite (Non-reasoning) | Amazon |
18.03
|
12.53
|
21.06
|
149.10
|
| Llama 3.1 Instruct 405B | Meta |
17.38
|
14.50
|
6.34
|
29.27
|
| EXAONE 4.0 32B (Reasoning) | LG AI Research |
16.68
|
13.98
|
9.53
|
—
|
| Nova 2.0 Omni (Non-reasoning) | Amazon |
16.61
|
13.84
|
14.91
|
191.72
|
| Qwen3.5 2B (Reasoning) | Alibaba |
16.29
|
3.45
|
23.00
|
—
|
| Nanbeige4.1-3B |  Nanbeige |
16.08
|
8.87
|
7.21
|
—
|
| Ministral 3 14B | Mistral |
15.98
|
10.90
|
17.39
|
137.45
|
| DeepSeek R1 Distill Llama 70B | DeepSeek |
15.95
|
11.43
|
—
|
43.52
|
| Falcon-H1R-7B |  TII UAE |
15.80
|
9.81
|
9.26
|
—
|
| Ling-flash-2.0 | InclusionAI |
15.74
|
16.72
|
8.35
|
83.49
|
| Qwen3 Omni 30B A3B (Reasoning) | Alibaba |
15.62
|
12.71
|
10.61
|
79.71
|
| Step3 VL 10B | StepFun |
15.45
|
13.91
|
5.36
|
—
|
| Gemma 4 E2B (Reasoning) | Google |
15.21
|
9.00
|
6.92
|
—
|
| Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) | NVIDIA |
15.02
|
13.09
|
3.80
|
40.57
|
| ERNIE 4.5 300B A47B | Baidu |
14.96
|
14.53
|
0.00
|
21.79
|
| Solar Pro 2 (Reasoning) | Upstage |
14.92
|
12.09
|
11.43
|
—
|
| NVIDIA Nemotron Nano 12B v2 VL (Reasoning) | NVIDIA |
14.89
|
11.75
|
7.12
|
131.71
|
| Ministral 3 8B | Mistral |
14.84
|
9.97
|
16.66
|
156.66
|
| Gemma 4 E4B (Non-reasoning) | Google |
14.83
|
6.36
|
8.67
|
53.59
|
| NVIDIA Nemotron Nano 9B V2 (Reasoning) | NVIDIA |
14.76
|
8.34
|
9.43
|
121.38
|
| Granite 4.1 30B |  IBM |
14.69
|
10.12
|
14.04
|
—
|
| NVIDIA Nemotron 3 Nano 4B | NVIDIA |
14.68
|
10.02
|
9.75
|
—
|
| Qwen3.5 2B (Non-reasoning) | Alibaba |
14.67
|
4.92
|
27.19
|
357.48
|
| Llama Nemotron Super 49B v1.5 (Non-reasoning) | NVIDIA |
14.59
|
10.47
|
8.38
|
45.43
|
| Llama 3.3 Instruct 70B | Meta |
14.49
|
10.70
|
9.09
|
82.99
|
| Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning) | NVIDIA |
14.43
|
—
|
—
|
—
|
| Kimi Linear 48B A3B Instruct | Kimi |
14.41
|
14.21
|
—
|
—
|
| Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) | NVIDIA |
14.35
|
7.64
|
—
|
—
|
| Ring-flash-2.0 | InclusionAI |
14.02
|
10.64
|
0.00
|
89.35
|
| Solar Pro 2 (Non-reasoning) | Upstage |
13.59
|
11.29
|
12.71
|
—
|
| Llama 4 Scout | Meta |
13.52
|
6.68
|
5.17
|
140.34
|
| Command A |  Cohere |
13.48
|
9.88
|
5.07
|
35.96
|
| Llama 3.1 Nemotron Instruct 70B | NVIDIA |
13.44
|
10.78
|
7.70
|
37.09
|
| NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) | NVIDIA |
13.17
|
15.76
|
8.48
|
87.32
|
| NVIDIA Nemotron Nano 9B V2 (Non-reasoning) | NVIDIA |
13.16
|
7.49
|
7.80
|
125.60
|
| Granite 4.1 8B | IBM |
12.38
|
7.25
|
10.67
|
89.50
|
| Sarvam 30B (high) | Sarvam |
12.34
|
7.92
|
11.50
|
255.39
|
| Gemma 4 E2B (Non-reasoning) | Google |
12.10
|
8.31
|
7.41
|
—
|
| R1 1776 |  Perplexity |
11.99
|
—
|
—
|
—
|
| Llama 3.2 Instruct 90B (Vision) | Meta |
11.90
|
—
|
—
|
57.76
|
| EXAONE 4.0 32B (Non-reasoning) | LG AI Research |
11.66
|
9.42
|
1.36
|
—
|
| Ministral 3 3B | Mistral |
11.24
|
4.78
|
11.44
|
270.97
|
| Jamba 1.7 Large |  AI21 Labs |
10.88
|
7.77
|
4.48
|
61.27
|
| Granite 4.0 H Small | IBM |
10.81
|
8.50
|
5.75
|
248.06
|
| Qwen3 Omni 30B A3B Instruct | Alibaba |
10.68
|
7.22
|
5.46
|
95.12
|
| Qwen3.5 0.8B (Reasoning) | Alibaba |
10.52
|
0.00
|
15.89
|
—
|
| LFM2 24B A2B |  Liquid AI |
10.49
|
3.63
|
3.70
|
86.52
|
| Phi-4 |  Microsoft |
10.41
|
11.21
|
0.00
|
35.86
|
| Nova Micro | Amazon |
10.27
|
4.14
|
4.68
|
282.15
|
| NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) | NVIDIA |
10.09
|
5.86
|
6.43
|
168.19
|
| Phi-4 Multimodal Instruct | Microsoft |
10.04
|
—
|
—
|
16.54
|
| Qwen3.5 0.8B (Non-reasoning) | Alibaba |
9.91
|
0.96
|
21.73
|
358.99
|
| Jamba Reasoning 3B | AI21 Labs |
9.60
|
2.47
|
5.26
|
—
|
| Reka Flash 3 |  Reka AI |
9.52
|
8.91
|
0.00
|
—
|
| Ling-mini-2.0 | InclusionAI |
9.19
|
5.02
|
4.39
|
—
|
| Llama 3.2 Instruct 11B (Vision) | Meta |
8.73
|
4.25
|
4.87
|
48.90
|
| Granite 4.1 3B | IBM |
8.54
|
5.49
|
6.53
|
—
|
| Phi-4 Mini Instruct | Microsoft |
8.39
|
3.59
|
2.73
|
43.59
|
| Exaone 4.0 1.2B (Reasoning) | LG AI Research |
8.26
|
3.09
|
5.46
|
—
|
| Exaone 4.0 1.2B (Non-reasoning) | LG AI Research |
8.11
|
2.47
|
6.82
|
—
|
| LFM2.5-1.2B-Thinking | Liquid AI |
8.08
|
1.39
|
6.53
|
—
|
| Jamba 1.7 Mini | AI21 Labs |
8.07
|
3.09
|
4.19
|
—
|
| LFM2 2.6B | Liquid AI |
8.04
|
1.35
|
4.48
|
—
|
| LFM2.5-1.2B-Instruct | Liquid AI |
8.04
|
0.77
|
3.61
|
—
|
| Granite 4.0 H 1B | IBM |
7.99
|
2.74
|
6.53
|
—
|
| Gemma 3 270M | Google |
7.71
|
0.00
|
3.02
|
—
|
| Apertus 70B Instruct |  Swiss AI Initiative |
7.70
|
1.89
|
4.29
|
—
|
| Granite 4.0 Micro | IBM |
7.67
|
4.98
|
4.19
|
—
|
| Granite 4.0 1B | IBM |
7.34
|
2.89
|
7.60
|
—
|
| LFM2 8B A1B | Liquid AI |
7.03
|
2.28
|
3.51
|
—
|
| LFM2.5-VL-1.6B | Liquid AI |
6.18
|
1.00
|
2.83
|
—
|
| Granite 4.0 350M | IBM |
6.10
|
0.31
|
4.39
|
—
|
| Apertus 8B Instruct | Swiss AI Initiative |
5.88
|
1.35
|
3.80
|
—
|
| Granite 4.0 H 350M | IBM |
5.44
|
0.58
|
4.87
|
—
|
| Tiny Aya Global | Cohere |
4.74
|
1.20
|
0.00
|
—
|
| EXAONE 4.5 33B (Non-reasoning) | LG AI Research |
—
|
—
|
—
|
—
|
| GPT-5.5 Pro (xhigh) | OpenAI |
—
|
—
|
—
|
—
|
| Gemini 3 Deep Think | Google |
—
|
—
|
—
|
—
|
| Mi:dm K 2.5 Pro Preview | Korea Telecom |
—
|
11.94
|
—
|
—
|