| Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models | ✓ Link | 90.4 | 86.7 | 85.6 | 80.2 | AutoIF (Llama3 70B) | 2024-06-19 |
| Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models | ✓ Link | 88 | 86.1 | 82.3 | 80.2 | AutoIF (Qwen2 72B) | 2024-06-19 |
| Instruction-Following Evaluation for Large Language Models | ✓ Link | 85.37 | 83.57 | 79.3 | 76.89 | GPT-4 | 2023-11-14 |
| Instruction-Following Evaluation for Large Language Models | ✓ Link | 59.11 | 55.76 | 46.95 | 43.07 | PaLM 2 S | 2023-11-14 |
| DeepSeek-V3 Technical Report | ✓ Link | | | | 86.1 | DeepSeek-V3 | 2024-12-27 |
| Nemotron-4 340B Technical Report | ✓ Link | | | | 79.9 | Nemotron-4-340B-Instruct | 2024-06-17 |
| DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning | ✓ Link | | | | 83.3 | DeepSeek-R1 | 2025-01-22 |
| Qwen2 Technical Report | ✓ Link | | | | 77.6 | Qwen2-72B-Instruct | 2024-07-15 |
| Qwen2 Technical Report | ✓ Link | | | | 59.9 | Qwen2-57B-A14B-Instruct | 2024-07-15 |
| Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models | ✓ Link | | | | 81.2 | Youtu-LLM 2B | 2025-12-31 |
| Kimi K2: Open Agentic Intelligence | ✓ Link | | | | 89.8 | Kimi K2 | 2025-07-28 |
| Announcing Bonsai 27B: The First 27B-Class Model to Run on a Phone | ✓ Link | | | | 85.03 | Ternary Bonsai 27B | 2026-07-14 |
| NGM: A Plug-and-Play Training-Free Memory Module for LLMs | ✓ Link | | | | 83.92 | Qwen3-14B + NGM | 2026-05-16 |
| SmolLM - blazingly fast and remarkably powerful | ✓ Link | | | | 18.48 | SmolLM 1.7B Instruct | 2024-07-16 |
| J-Zero: Unified Challenger--Solver--Judge Co-Evolution from Zero Data | ✓ Link | | | | 50.65 | J-Zero-4B | 2026-08-27 |
| Knowing When Not to Reuse: Conditional Experience Transfer in Autonomous LLM Post-Training | | | | | 76.7 | Qwen3-4B + BCIT (prompt-level IFEval) | 2026-08-27 |
| OmniEdu: Open Foundation Models for Learning and Teaching | ✓ Link | | | | 82.07 | OmniEdu-27B | 2026-09-19 |
| Beyond Teacher Assignment: Domain-Normalized Multi-Teacher On-Policy Distillation | ✓ Link | | | | 84.5 | DN-MOPD-9B | 2026-09-28 |
| Rufus-Air: An Open LLM Post-Training Recipe | | | | | 95.4 | Rufus-Air | 2026-09-24 |
| DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model | ✓ Link | | | 63.8 | | DeepSeek-V2 Chat (RL) (0-shot) | 2024-05-07 |
| DeepSeek LLM: Scaling Open-Source Language Models with Longtermism | ✓ Link | | | 59.1 | | DeepSeek LLM 67B Chat | 2024-01-05 |
| 2 OLMo 2 Furious | ✓ Link | | | 72.3 | | OLMo 2 7B Instruct | 2024-12-31 |
| 2 OLMo 2 Furious | ✓ Link | | | 82.6 | | OLMo 2 13B Instruct | 2024-12-31 |
| 2 OLMo 2 Furious | ✓ Link | | | 85.6 | | OLMo 2 32B Instruct | 2024-12-31 |
| OLMoE: Open Mixture-of-Experts Language Models | ✓ Link | | | 48.1 | | OLMoE-1B-7B-Instruct | 2024-09-03 |
| Jais 2: A Family of Arabic-Centric Open Large Language Models | | | 78.93 | | | Jais-2-70B | 2025-12-09 |
| CoRT: Counterfactual Replay for Token-Level Rubric-Guided Policy Optimization | | | 93.48 | | | Qwen3-14B + CoRT (AON) | 2026-07-28 |
| Carbon-X: When Natural Language Meets DNA | ✓ Link | | | | 79.3 | Carbon-X | 2026-10-22 |