| Introducing Beam: Reflection's 501B open-weight model | | 65.5 | Beam | 2026-10-05 |
| Hy3 preview | ✓ Link | 65.4 | Hy3 preview | 2026-04-23 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 63.2 | Qwen3.5-397B-A17B | 2026-02-16 |
| Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning | ✓ Link | 61.9 | Nemotron 3 Ultra | 2026-06-03 |
| Kimi K2.5: Visual Agentic Intelligence | ✓ Link | 61.0 | Kimi K2.5 | 2026-02-02 |
| MAI-Thinking-1: Building a Hill-Climbing Machine | | 61 | MAI-Thinking-1 | 2026-06-02 |
| MiMo-V2-Flash Technical Report | ✓ Link | 60.6 | MiMo-V2-Flash (Thinking) | 2026-01-06 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 60.6 | Qwen3.5-27B | 2026-02-16 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 60.2 | Qwen3.5-122B-A10B | 2026-02-16 |
| Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent | ✓ Link | 60.2 | Agents-A1 | 2026-06-29 |
| Qwen3.5-Omni Technical Report | | 59.6 | Qwen3.5-Omni-Plus | 2026-04-21 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 59.0 | Qwen3.5-35B-A3B | 2026-02-16 |
| Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters | ✓ Link | 57.5 | Step-3.5-Flash | 2026-02-11 |
| MiniMax-01: Scaling Foundation Models with Lightning Attention | ✓ Link | 56.5 | MiniMax-Text-01 | 2025-01-14 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 55.2 | Qwen3.5-9B | 2026-02-16 |
| Kimi K2: Open Agentic Intelligence | ✓ Link | 49.1 | Kimi K2 | 2025-07-28 |
| DeepSeek-V3 Technical Report | ✓ Link | 48.7 | DeepSeek-V3 | 2024-12-27 |
| Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale | ✓ Link | 48.31 | Ling-2.6-1T | 2026-06-13 |
| DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression | ✓ Link | 45.2 | DeepSeek-V4.1-Flash Base | 2026-09-17 |
| MiniCPM5-2B | ✓ Link | 43.7 | MiniCPM5-2B | 2026-09-07 |
| Hunyuan Dense V1 | ✓ Link | 43 | Hunyuan-7B-Instruct | 2025-07-30 |
| Introducing LFM-7B: Setting New Standards for Efficient Language Models | | 36.1 | LFM-7B | 2025-01-20 |
| Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models | ✓ Link | 27.2 | Youtu-LLM (2B, base) | 2025-12-31 |