Agentic Task on tau2 Telecom
92.3AccuracyHyperCLOVA X 32B Think
Evaluation Results
| Method | Links | |
|---|---|---|
| HyperCLOVA X 32B ThinkUser-simulator=GPT-5.12026.01 | 92.3 | |
| GPT-5.1 (Medium)User-simulator=GPT-5.12026.01 | 81.2 | |
| GPT-5.1 (Medium)User-simulator=GPT-4.12026.01 | 80.9 | |
| GLM-4.6Thinking Mode=true, Parameters=357B-A32B, Lang=English, Sub-Dataset=tau2 Telecom2026.01 | 70.5 | |
| HyperCLOVA X 32B ThinkUser-simulator=GPT-4.12026.01 | 65.1 | |
| A.X K1Thinking Mode=true, Parameters=519B-A33B, Lang=English, Sub-Dataset=tau2 Telecom2026.01 | 58.1 | |
| gpt-oss-120bNumber of Parameters=117B, Optimization Level=high2026.01 | 57.7 | |
| Solar OpenNumber of Parameters=102B2026.01 | 55.6 | |
| gpt-oss-120bNumber of Parameters=117B, Optimization Level=medium2026.01 | 47.4 | |
| DeepSeek-V3.1Thinking Mode=true, Parameters=685B-A37B, Lang=English, Sub-Dataset=tau2 Telecom2026.01 | 37.4 | |
| GLM-4.5-AirNumber of Parameters=110B2026.01 | 28.1 | |
| Qwen3 235B-A22BUser-simulator=GPT-5.12026.01 | 24.2 | |
| Qwen3 235B-A22BUser-simulator=GPT-4.12026.01 | 20.5 |