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 |
OpenGVLab
|
|---|---|
| Company | OpenGVLab |
| Release Date | 2024-12-06 |
| Parameters (B) | 2.2 |
| Architecture | ViT-MLP-LLM |
| Context Window | 16384 |
| Input Modalities | text, image |
| Open Source | Yes |
| License | mit |
| Score | 0 |
| VRAM (GB) | 8 |
| Compute | Single NVIDIA GPU with 8GB+ VRAM for bf16 inference (~4.2GB weights); 8-bit quantization supported for lower-memory deployment |
| Benchmark: BBH | 40.9 |
| Benchmark: C-Eval | 56.2 |
| Benchmark: GSM8K | 55.1 |
| Benchmark: HellaSwag | 68.2 |
| Benchmark: HumanEval | 52.4 |
| Benchmark: MATH | 33.5 |
| Benchmark: MBPP | 50.6 |
| Benchmark: MMLU | 52.6 |
| Benchmark: MMMU | 43.6 |
| Benchmark: MMMU-Pro | 23.7 |
| Pricing (Input) | — |
| Pricing (Output) | — |