| Paper | Code | Accuracy | ModelName | ReleaseDate |
|---|---|---|---|---|
| Introducing efficiency focused "Thinking CAP" model series | ✓ Link | 84.4 | ThinkingCap-Qwen3.6-27B | 2026-07-06 |
| Occamy-1.0: Open Pareto-frontier 35B Intelligence for Co-work | ✓ Link | 82.2 | Occamy-1.0 | 2026-09-04 |
| Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding | ✓ Link | 77.1 | Ornith-1.0-397B | 2026-06-21 |
| Qwen3.6 | ✓ Link | 72.4 | Qwen3.6-27B | 2026-04-21 |
| Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding | ✓ Link | 69.8 | Ornith-1.0-35B | 2026-06-21 |
| Mach-Mind-4-Flash Technical Report | 69.49 | Mach-Mind-4-Flash | 2026-07-10 | |
| Qwen3.6 | ✓ Link | 68.7 | Qwen3.6-35B-A3B | 2026-04-21 |
| WEFT: Scaling Tool-Use Post-Training for General-Purpose Agents | 64.0 | WEFT-35B-A3B | 2026-09-29 | |
| Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding | ✓ Link | 63.1 | Ornith-1.0-9B | 2026-06-21 |
| SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture | ✓ Link | 58.50 | SenseNova-U1-A3B-MoT | 2026-05-12 |