Reasoning on ZebraLogic (Score, TPF)
90.97ScoreQwen3-30B-A3B-Inst-2507
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-30B-A3B-Inst-25072026.02 | 90.97 | 100 | |
| LLaDA2.1-flashInference Mode=Q Mode2026.02 | 88.9 | 326 | |
| Ling-flash-2.02026.02 | 87.6 | 100 | |
| LLaDA2.1-flashInference Mode=S Mode2026.02 | 84.2 | 580 | |
| LLaDA2.0-flash2026.02 | 82.3 | 274 | |
| Ling-mini-2.02026.02 | 79.85 | — | |
| LLaDA2.1-minimode=Q Mode2026.02 | 77.1 | 2.93 | |
| LLaDA2.1-minimode=S Mode2026.02 | 68.5 | 5.38 | |
| LLaDA2.0-mini2026.02 | 64.2 | 2.3 | |
| Qwen3-8Bno think=true2026.02 | 37.48 | — | |
| Pair+Backbone=olmo-3-7b-it-sft2026.04 | 30.9 | — | |
| BaseBackbone=olmo-3.1-32b-it-sft2026.04 | 30 | — | |
| MPO+Backbone=olmo-3-7b-it-sft2026.04 | 28.8 | — | |
| RFTBackbone=olmo-3.1-32b-it-sft2026.04 | 28.3 | — | |
| Mag+Backbone=olmo-3-7b-it-sft2026.04 | 26.4 | — | |
| Soft+Backbone=olmo-3-7b-it-sft2026.04 | 25.5 | — | |
| DPOBackbone=olmo-3-7b-it-sft2026.04 | 25.4 | — | |
| Pair+Backbone=olmo-3.1-32b-it-sft2026.04 | 21.5 | — | |
| DPOBackbone=olmo-3.1-32b-it-sft2026.04 | 19.1 | — | |
| DPOBackbone=gemma-3-4b-sft2026.04 | 18.9 | — | |
| Soft+Backbone=olmo-3.1-32b-it-sft2026.04 | 18.6 | — | |
| Soft+Backbone=gemma-3-4b-sft2026.04 | 18.3 | — | |
| MPO+Backbone=olmo-3.1-32b-it-sft2026.04 | 18.2 | — | |
| BaseBackbone=olmo-3-7b-it-sft2026.04 | 17.5 | — | |
| Mag+Backbone=gemma-3-4b-sft2026.04 | 16.9 | — | |
| Pair+Backbone=gemma-3-4b-sft2026.04 | 16.9 | — | |
| MPO+Backbone=gemma-3-4b-sft2026.04 | 16.8 | — | |
| RFTBackbone=olmo-3-7b-it-sft2026.04 | 15.8 | — | |
| BaseBackbone=gemma-3-4b-sft2026.04 | 15.6 | — | |
| Mag+Backbone=olmo-3.1-32b-it-sft2026.04 | 12.1 | — | |
| RFTBackbone=gemma-3-4b-sft2026.04 | 11.7 | — |