| 1 |
Qwen3.5-9B (no reasoning)Alibaba |
0.41
|
| 2 |
Command ACohere |
0.42
|
| 3 |
Nemotron 3 Nano 30B A3B (no reasoning)NVIDIA |
0.43
|
| 4 |
Mistral Small 4 (no reasoning)Mistral |
0.47
|
| 5 |
Ministral 3 8B 2512Mistral |
0.48
|
| 6 |
gemma-4-e4b-it (no reasoning)Google |
0.48
|
| 7 |
Ministral 3 14B 2512Mistral |
0.50
|
| 8 |
Claude Haiku 4.5 (no reasoning)Anthropic |
0.54
|
| 9 |
Nova Micro 1.0Amazon |
0.58
|
| 10 |
Ministral 3 3B 2512Mistral |
0.58
|
| 11 |
Llama 4 ScoutMeta |
0.63
|
| 12 |
Llama 3.3 70B InstructMeta |
0.64
|
| 12 |
Llama 4 MaverickMeta |
0.64
|
| 14 |
Gemma 4 26B A4B (off)Google |
0.64
|
| 15 |
Claude Sonnet 5.5 (medium)Anthropic |
0.67
|
| 16 |
qwen3.5-4b (no reasoning)Alibaba |
0.67
|
| 17 |
Claude Sonnet 5.5 (low reasoning)Anthropic |
0.68
|
| 18 |
Mistral Large 3 2512Mistral |
0.69
|
| 19 |
GPT-5.6 Terra (no reasoning)OpenAI |
0.72
|
| 20 |
Gemma 4 31B (off)Google |
0.80
|
| 21 |
GPT-6 Luna (no reasoning)OpenAI |
0.82
|
| 22 |
Nova 2 Lite (no reasoning)Amazon |
0.83
|
| 23 |
DeepSeek V4.1 Flash (off)DeepSeek |
0.84
|
| 24 |
DeepSeek V4 Pro 0423 (no reasoning)DeepSeek |
0.91
|
| 25 |
Qwen3 Coder NextAlibaba |
0.91
|
| 26 |
Qwen3 Omni 30B A3B InstructAlibaba |
0.93
|
| 27 |
Qwen3.5 Omni FlashAlibaba |
1.01
|
| 28 |
Phi 4Microsoft |
1.03
|
| 29 |
Qwen3.6 35B A3B (no reasoning)Alibaba |
1.04
|
| 30 |
Qwen3 Next 80B A3B InstructAlibaba |
1.05
|
| 31 |
Qwen3.5-122B-A10B (no reasoning)Alibaba |
1.05
|
| 32 |
GPT-5.6 Terra (low reasoning)OpenAI |
1.13
|
| 33 |
Qwen3.8 27B (off)Alibaba |
1.22
|
| 34 |
Qwen3.5 Omni PlusAlibaba |
1.24
|
| 35 |
GPT-5.6 Terra (medium reasoning)OpenAI |
1.31
|
| 36 |
GPT-6 Astra (low reasoning)OpenAI |
1.41
|
| 37 |
gemma-4-12b-it (no reasoning)Google |
1.42
|
| 38 |
GPT-6.1 Sol (low)OpenAI |
1.51
|
| 39 |
Qwen3.5 397B A17B (no reasoning)Alibaba |
1.57
|
| 40 |
Claude Opus 5.5 (low reasoning)Anthropic |
1.61
|
| 41 |
GPT-5.6 Terra (high reasoning)OpenAI |
1.86
|
| 42 |
GPT-6 Luna (low)OpenAI |
1.91
|
| 43 |
Claude Fable 5.1 (low reasoning)Anthropic |
2.11
|
| 44 |
Claude Fable 5.1 (medium reasoning)Anthropic |
2.84
|
| 45 |
GPT-6 Astra (medium reasoning)OpenAI |
3.57
|
| 46 |
GPT-6.1 Sol (medium)OpenAI |
3.94
|
| 47 |
o3OpenAI |
3.96
|
| 48 |
Grok 4.7 (low)xAI |
4.25
|
| 49 |
GPT-5.6 Terra (xhigh)OpenAI |
4.36
|
| 50 |
Claude Fable 5.1 (high reasoning)Anthropic |
4.88
|
| 51 |
Grok 4.7 (high)xAI |
6.16
|
| 52 |
Claude Sonnet 5.5 (high)Anthropic |
6.20
|
| 53 |
Trinity Large ThinkingArcee Ai |
6.88
|
| 54 |
Gemini 3.5 Flash LiteGoogle |
7.04
|
| 55 |
Claude Haiku 4.5 (thinking reasoning)Anthropic |
7.09
|
| 56 |
Grok 4.7 (xhigh)xAI |
7.72
|
| 57 |
gpt-oss-20b (low reasoning)OpenAI |
8.49
|
| 58 |
GPT-6 Luna (high)OpenAI |
9.64
|
| 59 |
DeepSeek V4 Flash Vision (experimental, max)DeepSeek |
10.0
|
| 60 |
DeepSeek V4.1 Flash (max)DeepSeek |
10.1
|
| 61 |
gpt-oss-120b (low)OpenAI |
10.5
|
| 62 |
gpt-oss-20b (high reasoning)OpenAI |
10.8
|
| 63 |
Claude Sonnet 5.5 (xhigh)Anthropic |
11.0
|
| 64 |
Claude Opus 5.5 (medium reasoning)Anthropic |
11.1
|
| 65 |
Gemini 3.8 Flash (high)Google |
11.2
|
| 66 |
gpt-oss-120b (high)OpenAI |
11.3
|
| 67 |
Nemotron 3 Nano 30B A3B (thinking reasoning)NVIDIA |
11.3
|
| 68 |
Qwen3 Next 80B A3B ThinkingAlibaba |
12.1
|
| 69 |
Mistral Small 4Mistral |
12.7
|
| 70 |
Mistral Medium 3.5Mistral |
12.9
|
| 71 |
Nova 2 Lite (low reasoning)Amazon |
13.4
|
| 72 |
Inkling SmallThinkingmachines |
13.6
|
| 73 |
Nemotron 3 SuperNVIDIA |
14.1
|
| 74 |
Inkling (xhigh)Thinkingmachines |
14.3
|
| 75 |
Claude Opus 5.5 (high)Anthropic |
15.0
|
| 76 |
Gemini 3.1 Pro PreviewGoogle |
15.2
|
| 77 |
gemma-4-12b-itGoogle |
15.3
|
| 78 |
Qwen3.5-122B-A10BAlibaba |
16.0
|
| 79 |
GPT-6 Luna (xhigh)OpenAI |
16.5
|
| 80 |
GPT-5.5 Instant (26 June)OpenAI |
17.9
|
| 81 |
Nova 2 Lite (medium reasoning)Amazon |
18.4
|
| 82 |
qwen3-omni-30b-a3b-thinkingAlibaba |
18.8
|
| 83 |
Nova 2 Lite (high reasoning)Amazon |
19.5
|
| 84 |
muse-glimmer (high)Meta |
19.7
|
| 85 |
Mistral Large 4Mistral |
19.8
|
| 86 |
GPT-6.1 Sol (high)OpenAI |
20.1
|
| 87 |
Solar Pro 4Upstage |
21.1
|
| 88 |
Reka Flash 3Rekaai |
21.9
|
| 89 |
DeepSeek V4 Pro 0423 (max)DeepSeek |
22.3
|
| 90 |
Hy3Tencent |
24.4
|
| 91 |
MiniMax M3MiniMax |
25.9
|
| 92 |
GPT-6 Astra (high reasoning)OpenAI |
25.9
|
| 93 |
Kimi K2.7 CodeMoonshot AI |
28.0
|
| 94 |
GLM 5.3 (max)Z.ai |
29.3
|
| 95 |
GLM 5.3 (low reasoning)Z.ai |
30.0
|
| 96 |
Qwen3.5-9BAlibaba |
30.3
|
| 97 |
GPT-5.3-Codex (xhigh)OpenAI |
34.5
|
| 98 |
Claude Fable 5.1 (xhigh)Anthropic |
34.6
|
| 99 |
Qwen3.5 397B A17BAlibaba |
37.1
|
| 100 |
Qwen3.7 PlusAlibaba |
37.7
|
| 101 |
Qwen3.8 Flash NextAlibaba |
37.7
|
| 102 |
Granite 4.2 8BIBM |
39.7
|
| 103 |
Qwen3.8 27B (low)Alibaba |
40.1
|
| 104 |
Muse Spark 1.3 (xhigh)Meta |
41.3
|
| 105 |
Qwen3.6 35B A3BAlibaba |
41.7
|
| 106 |
MiMo-V2.6-FlashXiaomi |
41.8
|
| 107 |
Qwen3.8 27B (medium)Alibaba |
42.8
|
| 108 |
GLM 5.3 FlashZ.ai |
43.3
|
| 109 |
Qwen3.8 27B (xhigh)Alibaba |
43.8
|
| 110 |
Muse Spark 1.3 (max)Meta |
44.7
|
| 111 |
Kimi K3 (max)Moonshot AI |
47.0
|
| 112 |
Gemma 4 31BGoogle |
50.1
|
| 113 |
Kimi K3 (low reasoning)Moonshot AI |
50.6
|
| 114 |
Qwen3.8 Max (0902)Alibaba |
53.3
|
| 115 |
Qwen3.8 2.4T A95BAlibaba |
54.3
|
| 116 |
MiMo-V2.6-ProXiaomi |
54.8
|
| 117 |
GPT-6 Luna (max)OpenAI |
56.8
|
| 118 |
GPT-6.1 Sol (xhigh)OpenAI |
58.6
|
| 119 |
Claude Opus 5.5 (xhigh reasoning)Anthropic |
63.2
|
| 120 |
GPT-6 Astra (xhigh reasoning)OpenAI |
68.0
|
| 121 |
gemma-4-e4b-itGoogle |
73.0
|
| 122 |
GPT-5.6 Terra (max)OpenAI |
76.6
|
| 123 |
qwen3.5-4b (thinking reasoning)Alibaba |
106
|
| 124 |
Claude Fable 5.1 (max)Anthropic |
132
|
| 125 |
GPT-6.1 Sol (max)OpenAI |
177
|
| 126 |
GPT-6 Astra (max)OpenAI |
202
|
| 127 |
Claude Sonnet 5.5 (max)Anthropic |
269
|
| 128 |
Claude Opus 5.5 (max reasoning)Anthropic |
489
|