MiMo-V2-Flash model card (Hugging Face) 16 Dec 2025
- MMLU-Pro 84.9
- GPQA-Diamond 83.7
- HLE (no tools) 22.1
- AIME 2025 94.1
- HMMT Feb. 2025 84.4
- LiveCodeBench-v6 80.6
- Arena-Hard (Hard Prompt) 54.1, Creative Writing 86.2 54.1
- LongBench V2 60.6
- MRCR 45.7
- SWE-Bench Verified 73.4
- SWE-Bench Multilingual 71.7
- Terminal-Bench Hard 30.5
- Terminal-Bench 2.0 38.5
- Best of BrowseComp 45.4 / BrowseComp (w/ Context Manage) 58.3 58.3
- τ²-Bench 80.3