| ERNIE 5.0 Technical Report | | 96.5 | ERNIE 5.0 | 2026-02-04 |
| VERGE: Formal Refinement and Guidance Engine for Verifiable LLM Reasoning | | 91.0 | VERGE (GPT-OSS-120B) | 2026-01-27 |
| LongCat-Flash Technical Report | ✓ Link | 89.30 | LongCat-Flash | 2025-09-01 |
| Kimi K2: Open Agentic Intelligence | ✓ Link | 89.0 | Kimi K2 | 2025-07-28 |
| LLaDA2.1: Speeding Up Text Diffusion via Token Editing | ✓ Link | 88.9 | LLaDA2.1-flash (100B) | 2026-02-09 |
| JudgeRLVR: Judge First, Generate Second for Efficient Reasoning | | 87.4 | Mixed Strategy (Qwen3-30B-A3B) | 2026-01-13 |
| Hunyuan Dense V1 | ✓ Link | 85.1 | Hunyuan-7B-Instruct | 2025-07-30 |
| LLaDA2.0: Scaling Up Diffusion Language Models to 100B | ✓ Link | 82.3 | LLaDA2.0-flash | 2025-12-10 |
| Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning | ✓ Link | 81.7 | NPR (Qwen3-4B-Instruct-2507) | 2025-12-08 |
| Qwen3 Technical Report | ✓ Link | 80.3 | Qwen3-235B-A22B (Thinking) | 2025-05-14 |
| Olmo 3 | ✓ Link | 80.1 | Olmo 3.1 Think 32B | 2025-12-15 |
| Qwen3-Omni Technical Report | ✓ Link | 76.1 | Qwen3-Omni-Flash-Instruct | 2025-09-22 |
| ERNIE 4.5 Technical Report | ✓ Link | 58.1 | ERNIE-4.5-300B-A47B | 2025-06-30 |
| Qwen3 Technical Report | ✓ Link | 37.7 | Qwen3-235B-A22B (Non-thinking) | 2025-05-14 |
| TEMPO: Scaling Test-time Training for Large Reasoning Models | ✓ Link | 35.1 | OLMO3-7B-Base (TEMPO) | 2026-04-21 |
| Generation as Auxiliary Supervision: Enhancing Visual Understanding at Zero Inference Overhead via Decoupled Embedding Prediction | | 20.8 | GAS (2B) | 2026-08-12 |