| instruction-following-on-ifbench | Inkling-Small | #1 | Accuracy: 82.2 |
| mathematical-reasoning-on-hmmt-feb-2026 | Inkling-Small | #4 | Accuracy: 90.2 |
| physics-question-answering-on-critpt | Inkling-Small | #4 | Accuracy: 8.3 |
| code-generation-on-scicode | Inkling-Small | #6 | Accuracy: 48.7 |
| mathematical-reasoning-on-aime-2026 | Inkling-Small | #6 | Accuracy: 95.5 |
| code-generation-on-swe-bench-verified | Inkling-Small (Bash-only) | #10 | Accuracy: 80.2 |
| agents-on-toolathlon-verified-108-tasks | Inkling-Small | #11 | Pass@1: 0.544 |
| abstract-reasoning-on-arc-agi | Inkling-Small | #13 | Accuracy: 84.0 |
| abstract-reasoning-on-arc-agi-2 | Inkling-Small | #13 | Accuracy: 40.1 |
| multimodal-reasoning-on-mmmu-pro | Inkling-Small | #13 | Accuracy: 74.0 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | Inkling-Small | #14 | Accuracy: 89.5 |
| deep-research-agents-on-browsecomp | Inkling-Small (with context management) | #17 | Accuracy: 77.4 |
| code-generation-on-swe-bench-pro | Inkling-Small | #24 | Resolved: 55.9 |
| code-generation-on-terminal-bench-2-1 | Inkling-Small (Internal coding harness) | #29 | Accuracy: 64.7 |