Time series reasoning on TSandLang One TS
92.3AccuracyGPT-4
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4Modality=Text2025.10 | 92.3 | |
| GPT-4Modality=Vision2025.10 | 91.8 | |
| GPT-3.5Modality=Text2025.10 | 88.2 | |
| LLAMA-13BModality=Text2025.10 | 82.5 | |
| TS-Agent2025.10 | 80.5 | |
| GPT-4oModality=Text2025.10 | 79.8 | |
| LLAMA-7BModality=Text2025.10 | 78.8 | |
| ChatTSModality=TS2025.10 | 71.3 | |
| Mistral-7BModality=Text2025.10 | 66.3 | |
| GPT-4o-miniModality=Text2025.10 | 61.6 | |
| Gemini-2.5Modality=Text2025.10 | 33.5 | |
| Gemini-2.5Modality=Vision2025.10 | 27.1 |