Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Top downloads
Select 2-5 models
| Attribute |
XiaomiMiMo
|
|---|---|
| Company | XiaomiMiMo |
| Release Date | 2025-06-04 |
| Parameters (B) | 7 |
| Architecture | Vision-Language Transformer |
| Context Window | 32768 |
| Input Modalities | text, image, video |
| Open Source | Yes |
| License | mit |
| Score | 0 |
| VRAM (GB) | 24 |
| Compute | Single NVIDIA GPU with 24GB+ VRAM (e.g., RTX 4090, A10) for BF16 inference; compatible with Qwen2_5_VLForConditionalGeneration |
| Benchmark: AIME-2024 | 67.5 |
| Benchmark: AIME-2025 | 52.5 |
| Benchmark: Arena-Elo | 1093.9 |
| Benchmark: DROP | 85.1 |
| Benchmark: GPQA-Diamond | 58.3 |
| Benchmark: IFEval | 75.9 |
| Benchmark: MATH-500 | 95.4 |
| Benchmark: MMLU-Pro | 64.8 |
| Benchmark: MMMU | 66.7 |
| Benchmark: MMMU-Pro | 46.2 |
| Pricing (Input) | — |
| Pricing (Output) | — |