Time series reasoning on TSandLang Two TS
57.3AccuracyTS-Agent
Evaluation Results
| Method | Links | |
|---|---|---|
| TS-Agent2025.10 | 57.3 | |
| GPT-4Modality=Vision2025.10 | 53.6 | |
| GPT-4Modality=Text2025.10 | 52.7 | |
| ChatTSModality=TS2025.10 | 49.3 | |
| Gemini-2.5Modality=Vision2025.10 | 48.3 | |
| GPT-4o-miniModality=Text2025.10 | 44.1 | |
| Gemini-2.5Modality=Text2025.10 | 44 | |
| GPT-4oModality=Text2025.10 | 43.1 | |
| Mistral-7BModality=Text2025.10 | 41 | |
| GPT-3.5Modality=Text2025.10 | 27.4 | |
| LLAMA-13BModality=Text2025.10 | 25.8 | |
| LLAMA-7BModality=Text2025.10 | 25.2 |