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 |
LiquidAI
|
|---|---|
| Company | LiquidAI |
| Release Date | 2025-10-22 |
| Parameters (B) | 3 |
| Architecture | Hybrid Conv-Attention (LFM2-VL) |
| Context Window | 32768 |
| Input Modalities | text, image |
| Open Source | Yes |
| License | other |
| Score | 0 |
| VRAM (GB) | 6.8 |
| Compute | Single GPU with ~7 GB VRAM at bfloat16; GGUF Q4/Q8 quantizations run on edge GPUs and mobile devices with under 3 GB |
| Benchmark: GPQA | 30 |
| Benchmark: IFEval | 51.8 |
| Benchmark: MMLU | 63 |
| Benchmark: MMMU | 45.3 |
| Benchmark: MMMU-Pro | 28.7 |
| Benchmark: SimpleQA | 33 |
| Pricing (Input) | — |
| Pricing (Output) | — |