社区运行状况
| 注册会员 | 1285 |
| 主题 | 846 |
| 模型 | 3026 |
| 技能包 | 13874 |
| 数据集 | 1047 |
| 论文 | 443 |
| 开源项目 | 633 |
| 模型名称 | 厂商 | 智能评分 | 编程能力 | 智能体能力 | 速度评分 |
|---|---|---|---|---|---|
| Claude Opus 5 (Adaptive Reasoning, Max Effort) |  Anthropic |
60.69
|
77.98
|
55.26
|
55.13
|
| Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) | Anthropic |
60.07
|
77.00
|
54.53
|
55.93
|
| Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) | Anthropic |
59.86
|
76.49
|
52.81
|
73.84
|
| GPT-5.6 Sol (max) |  OpenAI |
58.89
|
77.39
|
54.00
|
70.60
|
| Claude Opus 5 (Adaptive Reasoning, High Effort) | Anthropic |
58.86
|
76.52
|
52.07
|
54.25
|
| GPT-5.6 Sol (xhigh) | OpenAI |
57.65
|
78.35
|
51.80
|
72.41
|
| Kimi K3 (max) |  Kimi |
57.11
|
76.24
|
50.07
|
35.93
|
| Claude Opus 5 (Adaptive Reasoning, Medium Effort) | Anthropic |
56.28
|
74.33
|
47.08
|
53.12
|
| GPT-5.6 Sol (high) | OpenAI |
55.87
|
77.16
|
48.53
|
63.90
|
| GPT-5.6 Terra (max) | OpenAI |
54.95
|
76.66
|
47.38
|
146.31
|
| Grok 4.5 (high) |  SpaceXAI |
53.83
|
72.45
|
45.69
|
59.73
|
| GPT-5.6 Sol (medium) | OpenAI |
53.59
|
76.26
|
44.55
|
63.56
|
| Claude Sonnet 5 (Adaptive Reasoning, Max Effort) | Anthropic |
53.35
|
71.55
|
46.69
|
80.22
|
| GPT-5.6 Terra (xhigh) | OpenAI |
51.60
|
70.64
|
44.70
|
123.56
|
| GPT-5.6 Luna (max) | OpenAI |
51.24
|
71.45
|
45.60
|
180.51
|
| GLM-5.2 (max) |  Z AI |
51.09
|
68.76
|
43.06
|
196.27
|
| Muse Spark 1.1 (xhigh) |  Meta |
50.62
|
71.34
|
37.54
|
207.48
|
| Claude Opus 5 (Adaptive Reasoning, Low Effort) | Anthropic |
50.61
|
66.95
|
39.79
|
53.78
|
| Gemini 3.5 Flash (high) |  Google |
50.20
|
70.14
|
37.45
|
184.19
|
| Gemini 3.6 Flash (high) | Google |
50.07
|
69.24
|
38.72
|
213.50
|
| DeepSeek V4 Flash 0731 (Reasoning, Max Effort) |  DeepSeek |
49.93
|
69.06
|
45.67
|
113.48
|
| GPT-5.6 Sol (low) | OpenAI |
49.44
|
69.67
|
40.00
|
65.31
|
| GPT-5.6 Luna (xhigh) | OpenAI |
49.07
|
68.60
|
42.92
|
161.85
|
| GPT-5.6 Terra (high) | OpenAI |
48.95
|
67.14
|
41.33
|
117.51
|
| Kimi K3 (low) | Kimi |
46.60
|
71.98
|
37.04
|
32.90
|
| Gemini 3.1 Pro Preview | Google |
46.46
|
68.83
|
21.40
|
131.79
|
| GPT-5.6 Luna (high) | OpenAI |
46.06
|
63.34
|
40.07
|
163.44
|
| Qwen3.7 Max |  Alibaba |
45.99
|
65.97
|
30.59
|
207.23
|
| GPT-5.6 Terra (medium) | OpenAI |
45.57
|
64.74
|
37.04
|
119.52
|
| Gemini 3.5 Flash (medium) | Google |
45.38
|
—
|
—
|
180.31
|
| MiniMax-M3 |  MiniMax |
44.44
|
58.57
|
35.36
|
84.31
|
| DeepSeek V4 Pro (Reasoning, Max Effort) | DeepSeek |
44.27
|
59.36
|
36.36
|
64.63
|
| GPT-5.3 Codex (xhigh) | OpenAI |
44.27
|
—
|
—
|
136.05
|
| Motif 3 (Beta) |  Motif Technologies |
44.08
|
61.97
|
34.93
|
—
|
| DeepSeek V4 Pro (Reasoning, High Effort) | DeepSeek |
43.11
|
58.67
|
34.43
|
65.35
|
| Muse Spark | Meta |
43.06
|
58.62
|
28.69
|
—
|
| MiMo-V2.5-Pro |  Xiaomi |
42.24
|
60.19
|
29.11
|
46.60
|
| Kimi K2.7 Code | Kimi |
41.95
|
60.76
|
29.59
|
40.21
|
| Claude Sonnet 5 (Non-reasoning, High Effort) | Anthropic |
41.73
|
66.39
|
33.70
|
63.20
|
| Hy3 |  Tencent |
41.23
|
58.80
|
30.73
|
70.19
|
| GPT-5.6 Sol (Non-reasoning) | OpenAI |
41.20
|
65.14
|
34.85
|
63.77
|
| Nex-N2-Pro |  Nex AGI |
40.96
|
59.12
|
31.03
|
129.09
|
| Inkling (xhigh) |  Thinking Machines |
40.74
|
52.06
|
32.34
|
84.79
|
| GPT-5.6 Terra (low) | OpenAI |
40.47
|
58.09
|
30.60
|
116.43
|
| Inkling Small | Thinking Machines |
40.17
|
52.95
|
30.78
|
131.09
|
| Qwen3.6 Plus | Alibaba |
39.56
|
54.53
|
27.55
|
55.67
|
| Qwen3.7 Plus | Alibaba |
38.98
|
55.86
|
20.81
|
53.00
|
| JT-4.1 Flash 236B A21B |  China Mobile |
38.83
|
52.43
|
34.91
|
—
|
| Agnes 2.5 Pro Alpha |  Sapiens AI |
38.79
|
58.78
|
25.17
|
—
|
| GPT-5.6 Luna (medium) | OpenAI |
38.05
|
50.73
|
30.99
|
152.58
|
| Nemotron 3 Ultra 550B A55B (Reasoning) |  NVIDIA |
37.76
|
49.27
|
27.36
|
131.44
|
| MiMo-V2.5 | Xiaomi |
37.20
|
56.80
|
23.71
|
79.76
|
| Qwen3.6 27B (Reasoning) | Alibaba |
37.05
|
53.72
|
27.03
|
58.65
|
| Gemini 3.5 Flash-Lite | Google |
36.48
|
49.32
|
26.82
|
365.57
|
| MiMo-V2-Omni-0327 | Xiaomi |
36.39
|
—
|
—
|
—
|
| Grok 4.3 (medium) | SpaceXAI |
36.00
|
—
|
—
|
137.57
|
| Grok 4.3 (low) | SpaceXAI |
35.44
|
—
|
—
|
129.66
|
| MiMo-V2-Omni | Xiaomi |
34.99
|
—
|
—
|
—
|
| Gemini 3.5 Flash (minimal) | Google |
34.91
|
—
|
—
|
172.28
|
| Kimi K2.6 (Non-reasoning) | Kimi |
34.58
|
—
|
—
|
31.01
|
| Claude Sonnet 4.6 (Non-reasoning, Low Effort) | Anthropic |
34.26
|
—
|
—
|
43.37
|
| GLM-5.2 (Non-reasoning) | Z AI |
34.05
|
46.49
|
34.78
|
108.19
|
| GPT-5.6 Terra (Non-reasoning) | OpenAI |
33.97
|
52.31
|
29.33
|
112.09
|
| KAT Coder Pro V2 |  KwaiKAT |
33.69
|
59.46
|
15.53
|
105.06
|
| Qwen3.5 397B A17B (Reasoning) | Alibaba |
33.68
|
48.21
|
19.85
|
66.51
|
| Hy3-preview (Reasoning) | Tencent |
33.58
|
—
|
—
|
—
|
| LongCat 2.0 |  LongCat |
33.48
|
45.26
|
21.82
|
42.76
|
| GPT-5.6 Luna (low) | OpenAI |
33.26
|
44.16
|
25.40
|
152.78
|
| MiMo-V2-Flash (Feb 2026) | Xiaomi |
33.22
|
—
|
—
|
—
|
| Qwen3.5 122B A10B (Reasoning) | Alibaba |
32.28
|
45.71
|
20.72
|
135.34
|
| Qwen3.5 397B A17B (Non-reasoning) | Alibaba |
31.98
|
—
|
—
|
64.04
|
| Qwen3.6 35B A3B (Reasoning) | Alibaba |
31.65
|
41.88
|
21.41
|
154.76
|
| DeepSeek V4 Pro (Non-reasoning) | DeepSeek |
31.22
|
—
|
—
|
63.24
|
| G9v3-39A5B |  AI9Stars |
30.94
|
31.71
|
28.72
|
—
|
| Qwen3.5 Omni Plus | Alibaba |
30.64
|
—
|
—
|
53.54
|
| Ring-2.6-1T |  InclusionAI |
30.57
|
42.83
|
18.86
|
—
|
| Qwen3.6 27B (Non-reasoning) | Alibaba |
30.46
|
46.63
|
23.29
|
59.49
|
| o3 | OpenAI |
30.40
|
—
|
—
|
124.67
|
| Step 3.7 Flash |  StepFun |
30.27
|
39.57
|
21.53
|
—
|
| Mistral Medium 3.5 |  Mistral |
29.95
|
46.90
|
19.00
|
143.06
|
| Claude 4.5 Haiku (Reasoning) | Anthropic |
29.58
|
43.89
|
16.40
|
107.79
|
| Gemma 4 31B (Reasoning) | Google |
29.35
|
43.43
|
14.45
|
35.03
|
| GPT-5.5 Instant (June 2026) | OpenAI |
28.86
|
39.42
|
11.02
|
193.90
|
| DeepSeek V4 Flash (Non-reasoning) | DeepSeek |
28.65
|
—
|
—
|
119.34
|
| JT-35B-Flash | China Mobile |
28.36
|
—
|
—
|
—
|
| KAT-Coder-Pro V1 | KwaiKAT |
28.26
|
—
|
—
|
—
|
| MiMo-V2.5-Pro (Non-reasoning) | Xiaomi |
27.86
|
—
|
—
|
40.90
|
| Qwen3.5 122B A10B (Non-reasoning) | Alibaba |
27.60
|
43.34
|
15.83
|
147.16
|
| GPT-5.6 Luna (Non-reasoning) | OpenAI |
26.56
|
39.28
|
22.01
|
163.57
|
| Hy3-preview (Non-reasoning) | Tencent |
26.10
|
—
|
—
|
—
|
| Ling-2.6-1T | InclusionAI |
26.05
|
—
|
—
|
—
|
| Doubao Seed Code |  ByteDance Seed |
25.98
|
—
|
—
|
—
|
| Gemini 2.5 Pro | Google |
25.82
|
33.25
|
7.10
|
143.74
|
| Gemma 4 26B A4B (Reasoning) | Google |
25.69
|
39.32
|
10.97
|
—
|
| NVIDIA Nemotron 3 Super 120B A12B (Reasoning) | NVIDIA |
25.41
|
37.72
|
8.72
|
142.38
|
| Gemini 3.1 Flash-Lite | Google |
25.04
|
34.69
|
6.17
|
311.54
|
| Grok 4.3 (Non-reasoning) | SpaceXAI |
24.75
|
35.18
|
22.81
|
129.92
|
| MiMo-V2-Flash (Non-reasoning) | Xiaomi |
24.67
|
49.84
|
11.99
|
—
|
| Qwen3.6 35B A3B (Non-reasoning) | Alibaba |
24.16
|
28.14
|
19.14
|
165.15
|
| Qwen3.5 35B A3B (Non-reasoning) | Alibaba |
23.99
|
36.98
|
11.82
|
174.53
|
| gpt-oss-120b (high) | OpenAI |
23.83
|
30.44
|
13.17
|
199.26
|
| Claude 4.5 Haiku (Non-reasoning) | Anthropic |
23.71
|
—
|
—
|
89.91
|
| Command A+ |  Cohere |
22.51
|
27.85
|
9.16
|
205.13
|
| K-EXAONE (Reasoning) |  LG AI Research |
22.10
|
32.11
|
8.00
|
—
|
| ERNIE 5.0 Thinking Preview |  Baidu |
21.92
|
—
|
—
|
—
|
| Gemma 4 12B (Reasoning) | Google |
21.81
|
30.96
|
7.93
|
111.37
|
| Gemma 4 31B (Non-reasoning) | Google |
21.77
|
33.17
|
10.96
|
69.19
|
| Nova 2.0 Pro Preview (medium) |  Amazon |
21.77
|
33.96
|
7.00
|
123.05
|
| Qwen3.5 9B (Reasoning) | Alibaba |
21.44
|
28.66
|
7.41
|
83.17
|
| Mercury 2 |  Inception |
21.43
|
31.11
|
9.60
|
769.75
|
| Qwen3 Coder Next | Alibaba |
21.11
|
36.23
|
8.79
|
135.54
|
| Nova 2.0 Omni (medium) | Amazon |
20.95
|
—
|
—
|
—
|
| Apriel-v1.6-15B-Thinker |  ServiceNow |
20.54
|
—
|
—
|
—
|
| Qwen3.5 9B (Non-reasoning) | Alibaba |
20.32
|
23.46
|
—
|
—
|
| EXAONE 4.5 33B | LG AI Research |
20.22
|
23.57
|
9.73
|
—
|
| Gemma 4 26B A4B (Non-reasoning) | Google |
20.10
|
—
|
—
|
65.45
|
| Qwen3.5 4B (Reasoning) | Alibaba |
20.09
|
22.60
|
—
|
45.87
|
| North Mini Code | Cohere |
19.80
|
36.45
|
3.12
|
23.86
|
| Nova 2.0 Pro Preview (low) | Amazon |
19.61
|
25.87
|
6.42
|
126.28
|
| Mistral Small 4 (Reasoning) | Mistral |
19.56
|
26.64
|
4.70
|
160.07
|
| Devstral 2 | Mistral |
19.16
|
31.26
|
10.57
|
43.06
|
| Nova 2.0 Lite (medium) | Amazon |
19.00
|
—
|
—
|
161.84
|
| Qwen3.5 Omni Flash | Alibaba |
18.99
|
—
|
—
|
242.99
|
| JT-MINI | China Mobile |
18.53
|
—
|
—
|
—
|
| Nova 2.0 Lite (high) | Amazon |
18.17
|
23.04
|
3.13
|
143.05
|
| Trinity Large Thinking |  Arcee AI |
18.16
|
25.77
|
3.65
|
197.00
|
| Magistral Medium 1.2 | Mistral |
17.87
|
21.32
|
1.86
|
114.43
|
| Nova 2.0 Lite (low) | Amazon |
17.81
|
—
|
—
|
159.53
|
| HyperNova 60B 2605 |  Multiverse Computing |
17.80
|
23.23
|
6.69
|
350.95
|
| Nemotron Cascade 2 30B A3B | NVIDIA |
17.58
|
25.35
|
3.67
|
—
|
| Devstral Small 2 | Mistral |
17.44
|
29.33
|
10.47
|
87.43
|
| K2 Think V2 |  MBZUAI Institute of Foundation Models |
17.26
|
20.98
|
1.79
|
—
|
| LongCat Flash Lite | LongCat |
17.22
|
—
|
—
|
—
|
| HyperCLOVA X SEED Think (32B) |  Naver |
17.02
|
—
|
—
|
—
|
| K-EXAONE (Non-reasoning) | LG AI Research |
16.74
|
—
|
—
|
—
|
| Qwen3 Next 80B A3B (Reasoning) | Alibaba |
16.72
|
17.42
|
2.06
|
207.08
|
| Nova 2.0 Omni (low) | Amazon |
16.56
|
—
|
—
|
—
|
| Mi:dm K 2.5 Pro |  Korea Telecom |
16.42
|
—
|
—
|
—
|
| G9v3-3B | AI9Stars |
16.06
|
9.90
|
14.05
|
—
|
| Qwen3.5 4B (Non-reasoning) | Alibaba |
16.00
|
20.33
|
—
|
31.86
|
| Mistral Large 3 | Mistral |
15.93
|
20.07
|
5.52
|
48.37
|
| INTELLECT-3 |  Prime Intellect |
15.61
|
—
|
—
|
—
|
| Solar Open 100B (Reasoning) |  Upstage |
15.15
|
—
|
—
|
—
|
| Nemotron 3 Nano Omni 30B A3B Reasoning | NVIDIA |
14.93
|
13.75
|
—
|
321.06
|
| gpt-oss-120b (low) | OpenAI |
14.91
|
21.24
|
0.96
|
266.53
|
| gpt-oss-20b (high) | OpenAI |
14.89
|
20.70
|
3.10
|
213.65
|
| Nova 2.0 Pro Preview (Non-reasoning) | Amazon |
14.36
|
20.86
|
2.85
|
107.31
|
| gpt-oss-20b (low) | OpenAI |
14.34
|
—
|
—
|
213.82
|
| Llama 4 Maverick | Meta |
14.27
|
16.28
|
1.31
|
110.84
|
| K2-V2 (high) | MBZUAI Institute of Foundation Models |
14.19
|
—
|
—
|
—
|
| NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) | NVIDIA |
14.18
|
14.37
|
1.99
|
269.80
|
| Solar Pro 3 | Upstage |
14.14
|
16.21
|
2.68
|
128.46
|
| Ling 2.6 Flash | InclusionAI |
14.05
|
25.26
|
2.25
|
91.16
|
| Qwen3 Next 80B A3B Instruct | Alibaba |
13.72
|
—
|
—
|
182.33
|
| DiffusionGemma 26B A4B | Google |
13.45
|
19.66
|
2.20
|
—
|
| Gemma 4 12B (Non-reasoning) | Google |
13.19
|
—
|
—
|
107.39
|
| Motif-2-12.7B-Reasoning | Motif Technologies |
12.79
|
—
|
—
|
—
|
| Nova Premier | Amazon |
12.73
|
—
|
—
|
32.42
|
| K2-V2 (medium) | MBZUAI Institute of Foundation Models |
12.43
|
—
|
—
|
—
|
| Llama Nemotron Super 49B v1.5 (Reasoning) | NVIDIA |
12.42
|
—
|
—
|
81.93
|
| Mistral Small 4 (Non-reasoning) | Mistral |
12.37
|
—
|
—
|
147.24
|
| Tri-21B-Think |  Trillion Labs |
12.36
|
—
|
—
|
—
|
| Gemma 4 E4B (Reasoning) | Google |
12.21
|
9.39
|
1.79
|
86.16
|
| MiniCPM5-1B (Reasoning) |  OpenBMB |
11.96
|
—
|
—
|
—
|
| Sarvam 105B (high) |  Sarvam |
11.94
|
—
|
—
|
—
|
| Nova 2.0 Lite (Non-reasoning) | Amazon |
11.83
|
—
|
—
|
147.09
|
| Celeris-1 |  Celeris |
11.81
|
14.40
|
2.77
|
2086.00
|
| MiniCPM5-1B (Non-reasoning) | OpenBMB |
11.74
|
—
|
—
|
—
|
| Magistral Small 1.2 | Mistral |
11.29
|
14.72
|
1.44
|
147.02
|
| Nanbeige4.1-3B |  Nanbeige |
11.09
|
9.62
|
0.00
|
—
|
| Ministral 3 14B | Mistral |
11.05
|
14.36
|
2.20
|
98.36
|
| EXAONE 4.0 32B (Reasoning) | LG AI Research |
10.59
|
—
|
—
|
—
|
| Nova 2.0 Omni (Non-reasoning) | Amazon |
10.53
|
—
|
—
|
—
|
| Llama 4 Scout | Meta |
10.04
|
8.17
|
1.10
|
133.60
|
| Hermes 4 - Llama-3.1 70B (Reasoning) |  Nous Research |
9.97
|
—
|
—
|
89.89
|
| Falcon-H1R-7B |  TII UAE |
9.79
|
—
|
—
|
—
|
| Gemma 4 E2B (Reasoning) | Google |
9.66
|
7.23
|
1.51
|
—
|
| Qwen3 Omni 30B A3B (Reasoning) | Alibaba |
9.63
|
—
|
—
|
100.44
|
| Step3 VL 10B | StepFun |
9.47
|
—
|
—
|
—
|
| Llama 3.3 Instruct 70B | Meta |
9.40
|
11.93
|
0.34
|
86.03
|
| Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) | NVIDIA |
9.08
|
—
|
—
|
52.36
|
| ERNIE 4.5 300B A47B | Baidu |
9.02
|
—
|
—
|
—
|
| Hermes 4 - Llama-3.1 405B (Reasoning) | Nous Research |
9.00
|
—
|
—
|
37.01
|
| NVIDIA Nemotron Nano 12B v2 VL (Reasoning) | NVIDIA |
8.96
|
—
|
—
|
89.71
|
| Ministral 3 8B | Mistral |
8.96
|
9.69
|
1.24
|
118.59
|
| Gemma 4 E4B (Non-reasoning) | Google |
8.91
|
—
|
—
|
88.01
|
| Granite 4.1 30B |  IBM |
8.86
|
10.35
|
1.31
|
—
|
| NVIDIA Nemotron Nano 9B V2 (Reasoning) | NVIDIA |
8.84
|
—
|
—
|
195.47
|
| Hermes 4 - Llama-3.1 405B (Non-reasoning) | Nous Research |
8.80
|
—
|
—
|
35.67
|
| NVIDIA Nemotron 3 Nano 4B | NVIDIA |
8.73
|
7.98
|
2.34
|
—
|
| Llama Nemotron Super 49B v1.5 (Non-reasoning) | NVIDIA |
8.68
|
—
|
—
|
90.73
|
| K2-V2 (low) | MBZUAI Institute of Foundation Models |
8.55
|
—
|
—
|
—
|
| Kimi Linear 48B A3B Instruct | Kimi |
8.53
|
—
|
—
|
—
|
| Llama 3.1 Instruct 405B | Meta |
8.50
|
—
|
—
|
—
|
| LFM2.5-8B-A1B |  Liquid AI |
8.32
|
—
|
—
|
336.80
|
| Ring-flash-2.0 | InclusionAI |
8.16
|
—
|
—
|
—
|
| Olmo 3.1 32B Think |  Allen Institute for AI |
8.09
|
—
|
—
|
—
|
| Command A | Cohere |
7.67
|
—
|
—
|
60.00
|
| Qwen3.5 2B (Reasoning) | Alibaba |
7.64
|
2.92
|
0.82
|
—
|
| Llama 3.1 Nemotron Instruct 70B | NVIDIA |
7.64
|
—
|
—
|
83.79
|
| NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) | NVIDIA |
7.39
|
—
|
—
|
229.21
|
| NVIDIA Nemotron Nano 9B V2 (Non-reasoning) | NVIDIA |
7.38
|
—
|
—
|
171.75
|
| Hermes 4 - Llama-3.1 70B (Non-reasoning) | Nous Research |
6.90
|
—
|
—
|
94.72
|
| Ministral 3 3B | Mistral |
6.80
|
4.78
|
1.58
|
205.71
|
| Granite 4.1 8B | IBM |
6.67
|
9.50
|
—
|
106.31
|
| Sarvam 30B (high) | Sarvam |
6.64
|
—
|
—
|
—
|
| Olmo 3.1 32B Instruct | Allen Institute for AI |
6.47
|
—
|
—
|
—
|
| Gemma 4 E2B (Non-reasoning) | Google |
6.41
|
—
|
—
|
—
|
| R1 1776 |  Perplexity |
6.31
|
—
|
—
|
—
|
| Llama 3.2 Instruct 90B (Vision) | Meta |
6.23
|
—
|
—
|
—
|
| Phi-4 Mini Instruct |  Microsoft |
6.01
|
3.84
|
0.34
|
43.58
|
| EXAONE 4.0 32B (Non-reasoning) | LG AI Research |
6.01
|
—
|
—
|
—
|
| Qwen3.5 2B (Non-reasoning) | Alibaba |
5.62
|
2.39
|
0.76
|
—
|
| Qwen3.5 0.8B (Reasoning) | Alibaba |
5.46
|
0.00
|
5.70
|
—
|
| DeepHermes 3 - Mistral 24B Preview (Non-reasoning) | Nous Research |
5.31
|
—
|
—
|
—
|
| Jamba 1.7 Large |  AI21 Labs |
5.30
|
—
|
—
|
56.91
|
| Granite 4.0 H Small | IBM |
5.24
|
—
|
—
|
48.05
|
| Qwen3 Omni 30B A3B Instruct | Alibaba |
5.11
|
—
|
—
|
95.48
|
| LFM2 24B A2B | Liquid AI |
4.95
|
—
|
—
|
—
|
| Phi-4 | Microsoft |
4.87
|
—
|
—
|
—
|
| Nova Micro | Amazon |
4.74
|
—
|
—
|
279.03
|
| Granite 4.1 3B | IBM |
4.70
|
4.72
|
0.41
|
—
|
| NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) | NVIDIA |
4.58
|
—
|
—
|
164.52
|
| Phi-4 Multimodal Instruct | Microsoft |
4.53
|
—
|
—
|
17.68
|
| MiniCPM-V 4.6 1.3B | OpenBMB |
4.19
|
0.69
|
1.44
|
—
|
| Jamba Reasoning 3B | AI21 Labs |
4.13
|
—
|
—
|
—
|
| Reka Flash 3 |  Reka AI |
4.06
|
—
|
—
|
—
|
| Olmo 3 7B Think | Allen Institute for AI |
3.97
|
—
|
—
|
—
|
| Molmo 7B-D | Allen Institute for AI |
3.81
|
—
|
—
|
—
|
| Ling-mini-2.0 | InclusionAI |
3.75
|
—
|
—
|
—
|
| Llama 3.2 Instruct 11B (Vision) | Meta |
3.33
|
—
|
—
|
49.55
|
| Qwen3.5 0.8B (Non-reasoning) | Alibaba |
3.31
|
1.21
|
0.41
|
—
|
| Exaone 4.0 1.2B (Reasoning) | LG AI Research |
2.90
|
—
|
—
|
—
|
| Olmo 3 7B Instruct | Allen Institute for AI |
2.81
|
—
|
—
|
—
|
| Exaone 4.0 1.2B (Non-reasoning) | LG AI Research |
2.77
|
—
|
—
|
—
|
| LFM2.5-1.2B-Thinking | Liquid AI |
2.75
|
—
|
—
|
—
|
| Jamba 1.7 Mini | AI21 Labs |
2.73
|
—
|
—
|
—
|
| LFM2 2.6B | Liquid AI |
2.71
|
—
|
—
|
—
|
| LFM2.5-1.2B-Instruct | Liquid AI |
2.71
|
—
|
—
|
—
|
| Granite 4.0 H 1B | IBM |
2.66
|
—
|
—
|
—
|
| Gemma 3 270M | Google |
2.41
|
—
|
—
|
—
|
| Apertus 70B Instruct |  Swiss AI Initiative |
2.40
|
—
|
—
|
—
|
| Granite 4.0 Micro | IBM |
2.37
|
—
|
—
|
—
|
| DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) | Nous Research |
2.29
|
—
|
—
|
—
|
| Granite 4.0 1B | IBM |
2.07
|
—
|
—
|
—
|
| Molmo2-8B | Allen Institute for AI |
2.03
|
—
|
—
|
—
|
| LFM2 8B A1B | Liquid AI |
1.78
|
—
|
—
|
—
|
| LFM2.5-VL-1.6B | Liquid AI |
1.01
|
—
|
—
|
395.21
|
| Apertus 8B Instruct | Swiss AI Initiative |
1.00
|
—
|
—
|
—
|
| Granite 4.0 350M | IBM |
1.00
|
—
|
—
|
—
|
| Granite 4.0 H 350M | IBM |
1.00
|
—
|
—
|
—
|
| Tiny Aya Global | Cohere |
1.00
|
—
|
—
|
—
|
| Claude Sonnet 5 (Adaptive Reasoning, High Effort) | Anthropic |
—
|
—
|
—
|
66.40
|
| Claude Sonnet 5 (Adaptive Reasoning, Low Effort) | Anthropic |
—
|
—
|
—
|
64.44
|
| Claude Sonnet 5 (Adaptive Reasoning, Medium Effort) | Anthropic |
—
|
—
|
—
|
64.76
|
| Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort) | Anthropic |
—
|
—
|
—
|
75.12
|
| Cogito v2.1 (Reasoning) |  Deep Cogito |
—
|
—
|
—
|
—
|
| 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 |
—
|
—
|
—
|
—
|