GSM8K gsm8k Leaderboard
Auto-discovered from papers reporting GSM8K (Accuracy). · Metric: Accuracy (higher is better) · 🟢 Updated 2h ago
| # | Model | Accuracy | Paper |
|---|---|---|---|
| 1 | Qwen3-Next-80B-A3B-Instruct (Full) | 93.70 | link |
| 2 | ENP | 51.30 | link |
| 3 | Qwen1.5MoE-A2.7B (Retain=0%) | 51.30 | link |
| 4 | TENP(Rectange) | 51.10 | link |
| 5 | TENP+Rank(ShortGPT) | 49.70 | link |
| 6 | TENP+BI(MoDeGPT) | 49.10 | link |
| 7 | DeepSeek-V2-Lite (Retain=0%) | 22.10 | link |
| 8 | A JoLT for the KV Cache: Near-Lossless KV Cache Compression via Joint Tucker and JL-Residual Allocation for LLMs | 0.01 | link |