| Paper | Code | Pass Rate | ModelName | ReleaseDate |
|---|---|---|---|---|
| DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression | ✓ Link | 35.4 | DeepSeek-V4.1-Flash-1017 | 2026-09-17 |
| DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression | ✓ Link | 31.8 | DeepSeek-V4.1-Flash Max | 2026-09-17 |
| Step 5 Preview | 29.5 | Step 5 Preview (High) | 2026-09-20 | |
| GLM-5.3: Frontier Coding with Emergent Cyber Capabilities | ✓ Link | 28.5 | GLM-5.3 (Claude Code) | 2026-08-14 |
| Qwen3.8-Max: A New Bar for Coding and Cowork | ✓ Link | 27.0 | Qwen3.8-Max | 2026-08-03 |
| GLM-5.3-Flash: Frontier Intelligence, Flash Cost | ✓ Link | 26.3 | GLM-5.3-Flash (Claude Code) | 2026-08-26 |
| DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence | ✓ Link | 25.7 | DeepSeek-V4-Pro-0813 (Max) | 2026-04-26 |
| DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence | ✓ Link | 25.2 | DeepSeek-V4-Flash-0731 (Max) | 2026-04-26 |
| Hy4 preview | ✓ Link | 22.8 | Hy4 preview (Claude Code) | 2026-08-28 |