Instruction-following clustering on SP500 (C2)
69.34V-measureGemini 2.5 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 2.5 ProModel Type=Reasoning Model2026.03 | 69.34 | |
| o3Model Type=Reasoning Model2026.03 | 66.51 | |
| QwQ-32BModel Type=Reasoning Model2026.03 | 63.23 | |
| C1-Qwen-7BModel Type=Our Model2026.03 | 60.28 | |
| C1-Qwen-14BModel Type=Our Model2026.03 | 59.44 | |
| GPT-oss-120BModel Type=Reasoning Model2026.03 | 51.84 | |
| Distill-Llama-70BModel Type=Reasoning Model2026.03 | 42.13 | |
| GPT-4.1Model Type=General Model2026.03 | 41.27 | |
| DeepSeek-R1Model Type=Reasoning Model2026.03 | 39.69 | |
| Distill-Qwen-32BModel Type=Reasoning Model2026.03 | 38.74 | |
| Qwen2.5-72B-InstructModel Type=General Model2026.03 | 33.28 | |
| Llama-3.1-70B-InstructModel Type=General Model2026.03 | 32.28 | |
| GPT-4oModel Type=General Model2026.03 | 31.15 |