General Reasoning on BBH (TPS, Speedup, Performance Score)
66.1Performance Score (BBH General Reasoning)TinyR1-32B-Preview
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TinyR1-32B-Preview2025.03 | 66.1 | — | — | |
| Code Expert2025.03 | 63.5 | — | — | |
| Science Expert2025.03 | 61.1 | — | — | |
| DreamBase Model=Dream-7B-Base, Few-shot=32026.03 | 49.65 | 25.94 | 1 | |
| Math Expert2025.03 | 48.2 | — | — | |
| Base Model2025.03 | 47 | — | — | |
| DualCacheBase Model=Dream-7B-Base, Few-shot=32026.03 | 45.34 | 228.09 | 8.8 | |
| ES-dLLMBase Model=Dream-7B-Base, Few-shot=32026.03 | 44.28 | 281.65 | 10.9 |