HyperAI
Command Palette
Search for a command to run...
𝜏²-Bench Telecom
一项双控对话式 AI 基准，模拟技术支持场景，智能体与用户必须协同操作以解决电信服务问题。
- 模型
- 147 当前，440 总共
- 中位数
- 36.3%
- 范围
- 0.0%–99.1%
| 排名 | 比较 | 模型 | 评分 |
|---|---|---|---|
| 101 | 21.3% | ||
| 102 |  Gemma 4 E2B (Reasoning) Google | 20.8% | |
| 103 | Gemma 4 E4B (Reasoning) Google | 20.8% | |
| 104 |  K2-V2 (Low) Institute of Foundation Models | 20.8% | |
| 105 |  Exaone 4.0 1.2B (Non-reasoning) LG AI Research | 20.5% | |
| 106 | 19.6% | ||
| 107 |  LFM2.5-1.2B-Thinking Liquid AI | 19.6% | |
| 108 | 19.3% | ||
| 109 | 19.0% | ||
| 110 | 18.4% | ||
| 111 |  Llama 4 Maverick Meta | 17.8% | |
| 112 | EXAONE 4.0 32B (Reasoning) LG AI Research | 17.3% | |
| 113 | 17.3% | ||
| 114 | Exaone 4.0 1.2B (Reasoning) LG AI Research | 16.4% | |
| 115 |  Qwen3 Omni 30B A3B Instruct Alibaba | 16.4% | |
| 116 | LFM2.5-8B-A1B Liquid AI | 16.1% | |
| 117 |  Step3 VL 10B StepFun | 16.1% | |
| 118 |  Jamba Reasoning 3B AI21 Labs | 15.8% | |
| 119 | Llama 4 Scout Meta | 15.5% | |
| 120 |  Command A Cohere | 15.2% | |
| 121 | 14.6% | ||
| 122 | 14.6% | ||
| 123 |  Nova Micro Amazon | 14.0% | |
| 124 | Jamba 1.7 Large AI21 Labs | 13.5% | |
| 125 | 13.2% | ||
| 126 |  Apertus 70B Instruct Swiss AI Initiative | 12.9% | |
| 127 | 12.6% | ||
| 128 | Jamba 1.7 Mini AI21 Labs | 12.6% | |
| 129 |  Olmo 3 7B Instruct Allen Institute for AI | 12.6% | |
| 130 | Apertus 8B Instruct Swiss AI Initiative | 11.4% | |
| 131 | 11.4% | ||
| 132 | LFM2 24B A2B Liquid AI | 11.1% | |
| 133 | LFM2.5-1.2B-Instruct Liquid AI | 10.8% | |
| 134 | Gemma 3 270M Google | 9.1% | |
| 135 | LFM2.5-VL-1.6B Liquid AI | 8.5% | |
| 136 |  Phi-4 Mini Instruct Microsoft | 8.2% | |
| 137 | EXAONE 4.0 32B (Non-reasoning) LG AI Research | 4.1% | |
| 138 |  ERNIE 4.5 300B A47B Baidu | 0.0% | |
| 139 | 0.0% | ||
| 140 | Molmo 7B-D Allen Institute for AI | 0.0% | |
| 141 | Molmo2-8B Allen Institute for AI | 0.0% | |
| 142 | Olmo 3.1 32B Think Allen Institute for AI | 0.0% | |
| 143 | Olmo 3 7B Think Allen Institute for AI | 0.0% | |
| 144 | Phi-4 Microsoft | 0.0% | |
| 145 |  Reka Flash 3 Reka AI | 0.0% | |
| 146 |  Ring-flash-2.0 InclusionAI | 0.0% | |
| 147 | Tiny Aya Global Cohere | 0.0% |
