{"data":{"kind":"file","path":"README.md","version_id":"v9v1uacy2u6sm8e8zvr42t1r","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":524,"modified_at":"2026-09-13T04:06:06.933000","content_hash":"54c9e1386253e2b3618f267186adf2136f6f0bd2160316788934b8c7367cd41d"},"entries":[],"content":"# aime25\n\nAIME 2025 competition math tasks (parts I and II). The model reasons through each task and boxes an integer answer, scored by math equivalence (`math-verify`) of the boxed answer against the gold.\n\n## Taskset\n\n- **Source:** [`opencompass/AIME2025`](https://huggingface.co/datasets/opencompass/AIME2025)\n- **Size:** 30 tasks (AIME2025-I and AIME2025-II combined)\n\n## Changelog\n\n- 2026-08-31: Yield task records on demand so bounded evaluations construct only the requested prefix.\n- 2026-06-23: Initial v1 taskset.\n","encoding":"utf-8","truncated":false,"total_bytes":524},"status":null}