| Paper | Code | Success Rate | ModelName | ReleaseDate |
|---|---|---|---|---|
| Introducing Claude Opus 5 | ✓ Link | 70.57 | Claude Opus 5 (Anthropic system card) | 2026-07-24 |
| GLM-5.3-Flash: Frontier Intelligence, Flash Cost | ✓ Link | 59.1 | GLM-5.3-Flash | 2026-08-26 |
| Kimi K3: Open Frontier Intelligence | ✓ Link | 58.3 | Kimi K3 Max | 2026-07-27 |
| Introducing Muse Spark 1.3 | 32.0 | Muse Spark 1.3 (max) | 2026-09-02 | |
| StateAct: Program State, before Pixels, for Long-Horizon Computer-Use Agents | 26.9 | StateAct (Claude Opus 4.8) | 2026-07-24 | |
| Qwen-CUA: Native Computer Use for (almost) Everything | ✓ Link | 21.2 | Qwen-CUA-Max | 2026-08-02 |
| Qwen3.8-Max: A New Bar for Coding and Cowork | ✓ Link | 19.4 | Qwen3.8-Max | 2026-08-03 |
| Introducing Muse Spark 1.1 | ✓ Link | 14.2 | Muse Spark 1.1 | 2026-07-09 |
| LongHorizon-Harness: Advancing Long-Horizon Agents for Real-World Tasks | ✓ Link | 8.3 | Qwen 3.7-Plus + LongHorizon-Harness | 2026-08-03 |