| computer-use-agents-on-osworld-verified | Qwen3.8-27B | #3 | Success Rate: 84.3 |
| visual-question-answering-vqa-on-realworldqa | Qwen3.8-27B | #3 | Accuracy: 85.9 |
| code-generation-on-livecodebench-v6 | Qwen3.8-27B | #4 | Pass@1: 90.3 |
| computer-use-agents-on-androidworld | Qwen3.8-27B | #5 | Success Rate: 81.9 |
| instruction-following-on-ifbench | Qwen3.8-27B | #6 | Accuracy: 79.5 |
| code-generation-on-swe-bench-pro | Qwen3.8-27B | #10 | Resolved: 61.7 |
| question-answering-on-humanity-s-last-exam-hle | Qwen3.8-27B | #10 | Accuracy: 30.8 |
| optical-character-recognition-ocr-on-omnidocbench-v1-5 | Qwen3.8-27B | #11 | Accuracy: 91.1 |
| code-generation-on-nl2repo | Qwen3.8-27B | #15 | Accuracy: 42.3 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | Qwen3.8-27B | #15 | Accuracy: 89.2 |
| code-generation-on-terminal-bench-2-1 | Qwen3.8-27B (Terminus) | #23 | Accuracy: 73.0 |
| code-generation-on-deepswe | Qwen3.8-27B (Claude Code) | #24 | Pass@1: 0.422 |